- Email: [email protected]
Patched represses the Hedgehog signalling pathway by promoting modification of the Smoothened protein
Brief Communication
1315
Patched represses the Hedgehog signalling pathway by promoting modification of the Smoothened protein P.W. Ingham, S. Nystedt*, Y. Nakano, W. Brown, D. Stark, M. van den Heuvel† and A.M. Taylor Hedgehog (Hh) signalling plays a central role in many developmental processes in both vertebrates and invertebrates [1]. The multipass membrane-spanning proteins Patched (Ptc) [2–4] and Smoothened (Smo) [5–7] have been proposed to act as subunits of a putative Hh receptor complex. According to this view, Smo functions as the transducing subunit, the activity of which is blocked by a direct interaction with the ligand-binding subunit, Ptc [8]. Activation of the intracellular signalling pathway occurs when Hh binds to Ptc [8–11], an event assumed to release Smo from Ptc-mediated inhibition. Evidence for a physical interaction between Smo and Ptc is so far limited to studies of the vertebrate versions of these proteins when overexpressed in tissue culture systems [8,12]. To test this model, we have overexpressed the Drosophila Smo protein in vivo and found that increasing the levels of Smo protein per se was not sufficient for activation of the pathway. Immunohistochemical staining of wildtype and transgenic embryos revealed distinct patterns of Smo distribution, depending on which region of the protein was detected by the antibody. Our findings suggest that Smo is modified to yield a non-functional form and this modification is promoted by Ptc in a nonstoichiometric manner. Address: MRC Intercellular Signalling Group, Centre for Developmental Genetics, University of Sheffield, UK. Present addresses: *Department of Cell and Molecular Biology, Section for Developmental Biology, Lund University, PO Box 94, S-22100 Lund, Sweden. †MRC Functional Genetics Unit, Department of Human Anatomy and Genetics, University of Oxford, South Parks Road, Oxford OX1 3QU, UK. Correspondence: P.W. Ingham E-mail: [email protected] Received: 28 July 2000 Revised: 23 August 2000 Accepted: 23 August 2000 Published: 6 October 2000 Current Biology 2000, 10:1315–1318 0960-9822/00/$ – see front matter © 2000 Elsevier Science Ltd. All rights reserved.
Results and discussion Various lines of evidence indicate that Hh activates its intracellular pathway by binding to Ptc [7–12], but the precise manner in which Ptc regulates the pathway remains unclear. Genetic studies have shown that Smo is
Figure 1
(a)
(c)
(b)
(d)
Current Biology
(a) Stage 10 h–Gal4;UAS–lacZ embryo stained with monoclonal anti-β-galactosidase antibody to reveal the expression of GAL4 driven by the hairy gene control elements. Note the seven (1–7) principal expression domains; expression in domains 2–4 was consistently weaker than in the other domains. (b) Stage 11 h–Gal4;UAS–smo embryo stained with monoclonal antibody 4D4 to reveal the pattern of Wg protein expression. This was identical to that of wild-type embryos, showing no evidence of the expansion of domains caused by ectopic activation of the Hh pathway. (c) Cuticular preparation of a smo3 homozygote, showing the typical zygotic null phenotype. Note the continuous ‘lawn’ of denticles covering most of the ventral surface. (d) Cuticular preparation of a smo3 homozygote expressing the Flag–smo transgene under h–Gal4 control. Note the restoration of naked cuticle in alternate segments (indicated in the abdominal region by dots), corresponding to the regions in which the transgene was expressed.
essential for Hh signalling and that, in the absence of Ptc, Smo is constitutively active [5,6]. Murone et al. [8] have shown that overexpression of vertebrate Smo in Hhresponsive tissue culture cells is sufficient to activate transcription of a Gli-dependent reporter gene, an effect that can be abrogated by co-transfecting the cells with a Ptc cDNA. Moreover, it was reported that Ptc and Smo could be co-immunoprecipitated from such cells [8], suggesting that Ptc suppresses Smo activity by a direct physical interaction. We have addressed this ‘sequestration’ model in vivo by overexpressing smo in alternate segments of the developing Drosophila embryo. Sequence encoding the FLAG epitope tag was added near the 5′ end of the fulllength Drosophila smo cDNA, which was then cloned into the UAS transformation vector pUAST [13]. Transgenic flies carrying this construct were crossed to flies homozygous for h–Gal4, which in the embryo drives expression of the GAL4 transactivator in alternating segment-wide
1316
Current Biology Vol 10 No 20
(a)
(e)
(b)
(f)
(c)
(g)
(d)
(h)
Figure 2
(i) W & P
E
E
Hh
P
W & P
E
E
P
Hh Current Biology
domains under the control of regulatory elements of the pair-rule gene hairy (h) [13] (see Figures 1a and 2i). Transcription of the wingless (wg) gene, a principal target of Hh signalling in the embryo, is restricted to cells immediately adjacent to those expressing Hh (see Figure 2i). Elimination of Ptc activity or ectopic expression of Hh results in the expansion of each wg domain, such that they come to occupy approximately half of every segment [14,15]. We therefore analysed wg expression in the h–Gal4;UAS–smo embryos to determine whether wg expression is ectopically activated in alternate segments in response to the overexpression of Smo. Contrary to expectation, we found no evidence of ectopic activation of wg, the embryos displaying an essentially wild-type pattern of expression (Figure 1b). Moreover, such h–Gal4;UAS–smo animals completed embryogenesis and developed into viable adults (data not shown). A trivial explanation of the failure of the overexpressed cDNA to activate the Hh pathway could be that the activity of the encoded protein has been compromised in some way, for instance, by addition of the epitope tag. To exclude this possibility, we tested its ability to rescue smo loss-of-function mutations. Expression of the tagged cDNA under the control of the ubiquitous arm–Gal4 driver [16] rescued fully the cuticle phenotype of homozygous null smo embryos (data not
(a) Wild-type stage 11 embryo hybridised with a full-length smo antisense probe, revealing the uniform distribution of smo mRNA. (b) Stage 10 wild-type embryo probed with the rabbit anti-SmoC antibody. Note the segmentally reiterated pattern of staining, the antibody-positive cells spanning each parasegmental groove. (c) Stage 10 h–Gal4;UAS–Flag–smo embryo probed with the rabbit anti-SmoC antibody. Note the elevated levels of staining in six stripes in alternating segments in the thoracic and abdominal region (dots) corresponding to parts of hairy stripes 2–7 (see Figure 1a), and the two additional domains of elevated expression in the cephalic region (asterisks), which fall within hairy domain 1 (see Figure 1a). (d) h–Gal4;UAS–Flag–smo embryo at a slightly later stage to that shown in (c), stained with rabbit anti-SmoC antibody (blue) and anti-En monoclonal antibody 4D9 (brown). Each stripe of Smo staining was centred over an En stripe. In the cephalic region, the two strong stripes of Smo expression within hairy domain 1 (asterisks) both coincided with an En stripe. (e) Stage 10 embryo homozygous for the ptc null allele ptcG12, stained with rabbit anti-SmoC antibody. Note the uniform staining. (f) Stage 10 Kr–Gal4;UAS–hh embryo stained with rabbit anti-SmoC antibody. Note the uniform staining between parasegments 5–9 (indicated by dots). (g) Stage 10 h–Gal4;UAS–Flag–smo embryo stained with rabbit anti-SmoC antibody (blue) and anti-Wg monoclonal antibody 4D4 (brown). Each stripe of Smo staining overlapped a Wg stripe at its anterior margin. (h) Detail of embryo shown in (d). (i) Schematic representation of gene expression domains in the ectoderm of a stage 10 embryo. Two parasegmental boundaries are shown, defined by the interface between cells expressing Wg and Ptc (green) and those expressing En and Hh (orange). Note that Hh secreted by the En-expressing cells signals in both directions, stimulating transcription of wg and ptc in more anterior cells and ptc (deep yellow) in more posterior cells; ptc is also expressed at lower levels in all other non-En-expressing cells (pale yellow). As Ptc is inactivated by Hh, these cells represent the domain of Ptc activity (pale yellow bar). Blue bars, domains in which Smo is potentially activated in response to Hh signalling and in which Smo protein was detected using the rabbit anti-SmoC antibody; magenta bar, a domain of hairy expression.
shown). Expression of the same transgene under the control of h–Gal4 in a smo null background rescued the cuticular phenotype in alternate segments (Figure 1c,d). To analyse the expression of the endogenous Smo protein, we raised an antibody against the membrane-proximal portion of the putative intracellular carboxy-terminal tail of Smo (anti-SmoC antibody; see Figure 3d) and used this to stain wild-type and transgenic embryos. In contrast to the ubiquitous distribution of the smo mRNA (Figure 2a), immunohistochemical staining of wild-type embryos with the anti-SmoC antibody revealed a modulated distribution of the protein (Figure 2b). Smo accumulates in a series of sequentially repeating stripes, each of which is about onehalf a segment in width and spans the parasegmental boundary, the site of Hh activity (see Figure 2i). Staining of h–Gal4;UAS–smo embryos with the anti-SmoC antibody revealed a significant increase in the levels of Smo protein in the GAL4-expressing segments (Figure 2c); strikingly, however, this staining was also restricted to cells flanking the parasegmental boundary. To determine the precise location of these Smo-positive cells, we probed embryos
Brief Communication
1317
Figure 3 (a) Late stage 10 h–Gal4;UAS–Flag–smo embryo stained with anti-FLAG M2 antibody. The transgenic protein was detected in alternate segments throughout each hairy expression domain. (b) This contrasted with the distribution detected in genotypically identical embryos using the rabbit anti-SmoC antibody. Note, in particular, the single broad stripe of staining (asterisk) in the region corresponding to hairy domain 1 in (a) compared with the two narrow stripes (asterisks) in the same region in (b). (c) Stage 10 h–Gal4;UAS–smo–3HA embryo stained with anti-HA antibody. The stripes of HA-positive cells were broader than those detected with the anti-SmoC antibody, though those corresponding to hairy domains 2–4 were frequently weaker and narrower than the rest, mirroring the lower levels of GAL4 expression in these domains. Note, however, the continuous broad patch of staining in hairy domain 1 (asterisk), indicative of the Hh-independent nature of the expression pattern. (d) Schematic representation of the Smo protein showing the location of the FLAG tag (green), the HA tag (blue) and the region of the carboxy-terminal tail (red) against which the
(a)
(d) Anti-FLAG antibody
(b) Anti-HA antibody
Anti-HA antibody
Anti-FLAG antibody
Anti-SMO-C antibody
(c)
Current Biology
rabbit anti-SmoC antibody was raised. The protein is shown in two hypothetical conformations, inactive (left) and active (right). Both of the epitope tags appear equally
simultaneously with the anti-SmoC antibody and monoclonal antibodies for Engrailed (En) [17], a marker of Hhexpressing cells [18], and Wg [19], a marker of Hh-responding cells [14]. Double staining with the antiEn antibody revealed that Smo accumulates in and around cells expressing En (Figure 2d,h); double staining with the anti-Wg antibody showed that Smo also accumulates in cells expressing Wg (Figure 2g). The accumulation of Smo in and around cells secreting Hh protein, strongly suggests that the translation and/or stability of Smo is promoted by Hh activity. To test this possibility, we analysed Smo distribution in embryos in which Hh is ectopically expressed under the control of the Kr promoter [20]. Such embryos displayed a ubiquitous expression of Smo between parasegments 5 and 9, precisely the region in which ectopic Hh expression is driven by the Kr–Gal4 driver (Figure 2f). As Hh acts by inhibiting Ptc activity, the effects of Hh on Smo would be expected to be mediated by Ptc. In embryos homozygous for a ptc loss-of-function mutation, the modulated pattern of staining typical of the wild type was lost, indicating that Smo protein accumulates uniformly in the absence of Ptc activity (Figure 2e). The simplest interpretation of these data is that Ptc functions to block the translation or promote the degradation of Smo. And, as the spatial distribution of Smo was unaltered in h–Gal4;UAS–smo transgenic embryos, it would follow that Ptc activity can suppress accumulation of Smo
accessible in both conformations, but the epitope(s) recognised by the anti-SmoC antibody are accessible only in the active form, which is promoted by Hh signalling.
protein independently of the levels at which the gene is transcribed. This effect of Hh/Ptc-mediated signalling on Smo accumulation provides a simple explanation for the lack of an effect of ectopic smo expression, namely that the exogenous protein never accumulates outside the normal domain of Smo activity. Surprisingly, however, when h–Gal4;UAS–smo embryos were probed with the anti-FLAG antibody, a strikingly different pattern of exogenous protein accumulation was observed (Figure 3a,b). In contrast to the narrow stripes detected by the anti-SmoC antibody, staining was seen throughout each h–Gal4 expression domain. This indicates that the smo mRNA is translated in all cells in which it is transcribed. It follows that the Ptc-dependent staining pattern revealed by the anti-SmoC antibody reflects a posttranslational modification of the Smo protein. One possibility is that Ptc could promote the cleavage of Smo, yielding a relatively stable but functionally inert truncated form of the protein in cells not exposed to Hh. In this connection, it is interesting to note that the SREPB cleavage activating protein (SCAP), with which Ptc shares some homology, acts to promote the cleavage of SREBP [21] by chaperoning the latter from the endoplasmic reticulum to the Golgi [22,23]. Alternatively, however, it could be that Ptc induces a modification of the Smo protein that masks the epitope recognised by the anti-SmoC antibody; binding of Hh to Ptc would suppress this modification, activating the protein and making it accessible to the antibody. To discriminate between these two possibilities, we generated a
1318
Current Biology Vol 10 No 20
second tagged form of Smo, in this case inserting a haemagglutinin (HA) tag at the end of the carboxy-terminal tail (see Figure 3d). When embryos expressing this construct under h–Gal4 control were stained with an anti-HA antibody, a similar broad distribution of the tagged protein to that revealed by the anti-FLAG antibody was seen (Figure 3c). This argues against the cleavage model, but instead suggests that the carboxy-terminal tail undergoes a Ptc-dependent modification. As our anti-SmoC antibody was raised against an unmodified bacterially expressed protein, it seems most likely that this modification results in a conformational change that exposes epitopes recognised by the antibody. In this regard, it is notable that a putative dominant gain-of-function mutation in the human Smo protein is predicted to change the conformation of the equivalent region of the carboxy-terminal tail against which the anti-SmoC antibody is directed [24].
Materials and methods Generation of epitope tagged Smo transgenes A full-length smo cDNA was assembled from genomic clones [6]. A fragment encoding the prolactin signal peptide and FLAG tag was amplified from thrombin receptor cDNA (a generous gift from Shaun R. Coughlin, University of California at San Francisco) and used to replace the endogenous sequence encoding Smo signal peptide. The smo–HA transgene was generated by cloning the triple HA tag present in the shuttle plasmid pRD67 [11] into the XhoI site of the smo cDNA.
Generation and purification of anti-Smo antibody A fragment that encodes a region of the carboxy-terminal tail of Smo from residues 646 to 881 was amplified from the smo cDNA and ligated in-frame with NdeI/BamHI-digested pET14b (Novagen). ITPG induction of bacteria transformed with this plasmid yielded a His-tagged fusion protein of ~30 kD. Inclusion bodies were purified from induced cultures following standard protocols [25] and used to immunise rabbits. Antibodies specific for Smo were purified from the serum by immunoaffinity purification [25] using the bacterially expressed fragment fusion protein coupled to an NHS-activated Sepharose column (Hi Trap NHS-activated, Amersham-Pharmacia Biotech). Acidic and basic elutions were pooled and concentrated using an Amicon concentrator.
Antibody stainings and microscopy Embryos were fixed and stained using standard protocols. The affinitypurified anti-Smo antibody was used at 1:50–1:100. Monoclonal antiFLAG antibodies M1 and M2 (Sigma) were used at a dilution of 1:200. Monoclonal anti-En (4D9) and anti-Wg (4D4) were used at 1:100 dilution. Monoclonal anti-β-galactosidase was used at 1:1000. Specimens were examined and photographed using a Zeiss Axioplan 2 microscope. Figures were composed using Adobe Photoshop 4.0 software on Macintosh computers.
2
3. 4.
5.
6. 7. 8. 9. 10. 11. 12.
13. 14. 15. 16. 17. 18. 19. 20. 21. 22.
Supplementary material Supplementary material including full methodological details is available at http://current-biology.com/supmat/supmatin.htm.
23.
Acknowledgements
24.
This work was supported by a Wellcome Trust Programme grant to P.W.I. S.N. was a Research Fellow of the Swedish Foundation for International Cooperation in Research and Higher Education.
25.
References 1. Hammerschmidt M, Brook A, McMahon A: The world according to hedgehog. Trends Genet 1997, 13:14-21.
Nakano Y, Guerrero I, Hidalgo A, Taylor AM, Whittle JRS, Ingham PW: The Drosophila segment polarity gene patched encodes a protein with multiple potential membrane spanning domains. Nature 1989, 341:508-513. Hooper J, Scott MP: The Drosophila patched gene encodes a putative membrane protein required for segmental patterning. Cell 1989, 59:751-765. Goodrich LV, Johnson RL, Milenkovic L, McMahon JA, Scott MP: Conservation of the hedgehog/patched signaling pathway from flies to mice — induction of a mouse patched gene by hedgehog. Genes Dev 1996, 10:301-312. Alcedo J, Ayzenzon M, Vonohlen T, Noll M, Hooper JE: The Drosophila smoothened gene encodes a 7-pass membraneprotein, a putative receptor for the hedgehog signal. Cell 1996, 86:221-232. van den Heuvel M, Ingham PW: smoothened encodes a serpentine protein required for hedgehog signaling. Nature 1996, 382:547-551. Stone D, Hynes M, Armanini M, Swanson T, Gu Q, Johnson R, et al.: The tumor-suppressor gene patched encodes a candidate receptor for Sonic hedgehog. Nature 1996, 384:129-134. Murone M, Rosenthal A, de Sauvage F: Sonic hedgehog signaling by the patched–smoothened receptor complex. Curr Biol 1999, 9:76-84. Ingham PW, Taylor AM, Nakano Y: Role of the Drosophila patched gene in positional signalling. Nature 1991, 353:184-187. Chen Y, Struhl G: Dual roles for patched in sequestering and transducing hedgehog. Cell 1996, 87:553-563. Marigo V, Davey R, Zuo Y, Cunningham J, Tabin C: Biochemicalevidence that Patched is the Hedgehog receptor. Nature 1996, 384:176-179. Carpenter D, Stone DM, Brush J, Ryan A, Armanini M, Frantz G, et al.: Characterization of two patched receptors for the vertebrate hedgehog protein family. Proc Natl Acad Sci USA 1998, 95:13630-13634 Brand AH, Perrimon N: Targeted gene expression as a means of altering cell fates and generating dominant phenotypes. Development 1993, 118:401-415. Ingham PW: Localised hedgehog activity controls spatially restricted transcription of wingless in the Drosophila embryo. Nature 1993, 366:560-562. Martinez Arias A, Baker NE, Ingham PW: Role of segment polarity gene in the definition and maintenance of cell states in the Drosophila embryo. Development. 1988, 103:157-170. Vincent JP, Girdham C: Promoters to express cloned genes uniformly in Drosophila. Meth Mol Biol 1997, 62:385-392. Patel NH, Martin-Blanco E, Coleman KG, Poole SJ, Ellis MC, Kornberg TB, Goodman CS: Expression of engrailed proteins in arthropods, annelids, and chordates. Cell 1989, 58:955-968. Tabata T, Eaton S, Kornberg TB: The Drosophila hedgehog gene is expressed specifically in posterior compartment cells and is a target of engrailed regulation. Genes Dev 1992, 6:2635-2645. Brook WJ, Cohen SM: Antagonistic interaction between Wingless and Decapentaplegic responsible for dorsal-ventral pattern in the Drosophila leg. Science 1996, 273:1373-1377. Castelli-Gair J, Greig S, Micklem G, Akam M: Dissecting the temporal requirements for homeotic gene function. Development 1994, 120:1983-1995. Brown MS, Goldstein JL: A proteolytic pathway that controls the cholesterol content of membranes, cells, and blood. Proc Natl Acad Sci USA 1999, 96:11041-11048. DeBose-Boyd RA, Brown MS, Li WP, Nohturfft A, Goldstein JL, Espenshade PJ: Transport-dependent proteolysis of SREBP: relocation of site-1 protease from Golgi to ER obviates the need for SREBP transport to Golgi. Cell 1999, 23:703-712. Nohturfft A, DeBose-Boyd RA, Scheek S, Goldstein JL, Brown MS: Sterols regulate cycling of SREBP cleavage-activating protein (SCAP) between endoplasmic reticulum and Golgi. Proc Natl Acad Sci USA 1999, 96:11235-11240. Xie J, Murone M, Luoh SM, Ryan A, Gu Q, Zhang C, et al.: Activating Smoothened mutations in sporadic basal-cell carcinoma. Nature 1998, 391:90-92. Harlow E, Lane D: Antibodies: A Laboratory Manual. Cold Spring Harbor: Cold Spring Harbor Laboratory Press; 1988.
1315
Patched represses the Hedgehog signalling pathway by promoting modification of the Smoothened protein P.W. Ingham, S. Nystedt*, Y. Nakano, W. Brown, D. Stark, M. van den Heuvel† and A.M. Taylor Hedgehog (Hh) signalling plays a central role in many developmental processes in both vertebrates and invertebrates [1]. The multipass membrane-spanning proteins Patched (Ptc) [2–4] and Smoothened (Smo) [5–7] have been proposed to act as subunits of a putative Hh receptor complex. According to this view, Smo functions as the transducing subunit, the activity of which is blocked by a direct interaction with the ligand-binding subunit, Ptc [8]. Activation of the intracellular signalling pathway occurs when Hh binds to Ptc [8–11], an event assumed to release Smo from Ptc-mediated inhibition. Evidence for a physical interaction between Smo and Ptc is so far limited to studies of the vertebrate versions of these proteins when overexpressed in tissue culture systems [8,12]. To test this model, we have overexpressed the Drosophila Smo protein in vivo and found that increasing the levels of Smo protein per se was not sufficient for activation of the pathway. Immunohistochemical staining of wildtype and transgenic embryos revealed distinct patterns of Smo distribution, depending on which region of the protein was detected by the antibody. Our findings suggest that Smo is modified to yield a non-functional form and this modification is promoted by Ptc in a nonstoichiometric manner. Address: MRC Intercellular Signalling Group, Centre for Developmental Genetics, University of Sheffield, UK. Present addresses: *Department of Cell and Molecular Biology, Section for Developmental Biology, Lund University, PO Box 94, S-22100 Lund, Sweden. †MRC Functional Genetics Unit, Department of Human Anatomy and Genetics, University of Oxford, South Parks Road, Oxford OX1 3QU, UK. Correspondence: P.W. Ingham E-mail: [email protected] Received: 28 July 2000 Revised: 23 August 2000 Accepted: 23 August 2000 Published: 6 October 2000 Current Biology 2000, 10:1315–1318 0960-9822/00/$ – see front matter © 2000 Elsevier Science Ltd. All rights reserved.
Results and discussion Various lines of evidence indicate that Hh activates its intracellular pathway by binding to Ptc [7–12], but the precise manner in which Ptc regulates the pathway remains unclear. Genetic studies have shown that Smo is
Figure 1
(a)
(c)
(b)
(d)
Current Biology
(a) Stage 10 h–Gal4;UAS–lacZ embryo stained with monoclonal anti-β-galactosidase antibody to reveal the expression of GAL4 driven by the hairy gene control elements. Note the seven (1–7) principal expression domains; expression in domains 2–4 was consistently weaker than in the other domains. (b) Stage 11 h–Gal4;UAS–smo embryo stained with monoclonal antibody 4D4 to reveal the pattern of Wg protein expression. This was identical to that of wild-type embryos, showing no evidence of the expansion of domains caused by ectopic activation of the Hh pathway. (c) Cuticular preparation of a smo3 homozygote, showing the typical zygotic null phenotype. Note the continuous ‘lawn’ of denticles covering most of the ventral surface. (d) Cuticular preparation of a smo3 homozygote expressing the Flag–smo transgene under h–Gal4 control. Note the restoration of naked cuticle in alternate segments (indicated in the abdominal region by dots), corresponding to the regions in which the transgene was expressed.
essential for Hh signalling and that, in the absence of Ptc, Smo is constitutively active [5,6]. Murone et al. [8] have shown that overexpression of vertebrate Smo in Hhresponsive tissue culture cells is sufficient to activate transcription of a Gli-dependent reporter gene, an effect that can be abrogated by co-transfecting the cells with a Ptc cDNA. Moreover, it was reported that Ptc and Smo could be co-immunoprecipitated from such cells [8], suggesting that Ptc suppresses Smo activity by a direct physical interaction. We have addressed this ‘sequestration’ model in vivo by overexpressing smo in alternate segments of the developing Drosophila embryo. Sequence encoding the FLAG epitope tag was added near the 5′ end of the fulllength Drosophila smo cDNA, which was then cloned into the UAS transformation vector pUAST [13]. Transgenic flies carrying this construct were crossed to flies homozygous for h–Gal4, which in the embryo drives expression of the GAL4 transactivator in alternating segment-wide
1316
Current Biology Vol 10 No 20
(a)
(e)
(b)
(f)
(c)
(g)
(d)
(h)
Figure 2
(i) W & P
E
E
Hh
P
W & P
E
E
P
Hh Current Biology
domains under the control of regulatory elements of the pair-rule gene hairy (h) [13] (see Figures 1a and 2i). Transcription of the wingless (wg) gene, a principal target of Hh signalling in the embryo, is restricted to cells immediately adjacent to those expressing Hh (see Figure 2i). Elimination of Ptc activity or ectopic expression of Hh results in the expansion of each wg domain, such that they come to occupy approximately half of every segment [14,15]. We therefore analysed wg expression in the h–Gal4;UAS–smo embryos to determine whether wg expression is ectopically activated in alternate segments in response to the overexpression of Smo. Contrary to expectation, we found no evidence of ectopic activation of wg, the embryos displaying an essentially wild-type pattern of expression (Figure 1b). Moreover, such h–Gal4;UAS–smo animals completed embryogenesis and developed into viable adults (data not shown). A trivial explanation of the failure of the overexpressed cDNA to activate the Hh pathway could be that the activity of the encoded protein has been compromised in some way, for instance, by addition of the epitope tag. To exclude this possibility, we tested its ability to rescue smo loss-of-function mutations. Expression of the tagged cDNA under the control of the ubiquitous arm–Gal4 driver [16] rescued fully the cuticle phenotype of homozygous null smo embryos (data not
(a) Wild-type stage 11 embryo hybridised with a full-length smo antisense probe, revealing the uniform distribution of smo mRNA. (b) Stage 10 wild-type embryo probed with the rabbit anti-SmoC antibody. Note the segmentally reiterated pattern of staining, the antibody-positive cells spanning each parasegmental groove. (c) Stage 10 h–Gal4;UAS–Flag–smo embryo probed with the rabbit anti-SmoC antibody. Note the elevated levels of staining in six stripes in alternating segments in the thoracic and abdominal region (dots) corresponding to parts of hairy stripes 2–7 (see Figure 1a), and the two additional domains of elevated expression in the cephalic region (asterisks), which fall within hairy domain 1 (see Figure 1a). (d) h–Gal4;UAS–Flag–smo embryo at a slightly later stage to that shown in (c), stained with rabbit anti-SmoC antibody (blue) and anti-En monoclonal antibody 4D9 (brown). Each stripe of Smo staining was centred over an En stripe. In the cephalic region, the two strong stripes of Smo expression within hairy domain 1 (asterisks) both coincided with an En stripe. (e) Stage 10 embryo homozygous for the ptc null allele ptcG12, stained with rabbit anti-SmoC antibody. Note the uniform staining. (f) Stage 10 Kr–Gal4;UAS–hh embryo stained with rabbit anti-SmoC antibody. Note the uniform staining between parasegments 5–9 (indicated by dots). (g) Stage 10 h–Gal4;UAS–Flag–smo embryo stained with rabbit anti-SmoC antibody (blue) and anti-Wg monoclonal antibody 4D4 (brown). Each stripe of Smo staining overlapped a Wg stripe at its anterior margin. (h) Detail of embryo shown in (d). (i) Schematic representation of gene expression domains in the ectoderm of a stage 10 embryo. Two parasegmental boundaries are shown, defined by the interface between cells expressing Wg and Ptc (green) and those expressing En and Hh (orange). Note that Hh secreted by the En-expressing cells signals in both directions, stimulating transcription of wg and ptc in more anterior cells and ptc (deep yellow) in more posterior cells; ptc is also expressed at lower levels in all other non-En-expressing cells (pale yellow). As Ptc is inactivated by Hh, these cells represent the domain of Ptc activity (pale yellow bar). Blue bars, domains in which Smo is potentially activated in response to Hh signalling and in which Smo protein was detected using the rabbit anti-SmoC antibody; magenta bar, a domain of hairy expression.
shown). Expression of the same transgene under the control of h–Gal4 in a smo null background rescued the cuticular phenotype in alternate segments (Figure 1c,d). To analyse the expression of the endogenous Smo protein, we raised an antibody against the membrane-proximal portion of the putative intracellular carboxy-terminal tail of Smo (anti-SmoC antibody; see Figure 3d) and used this to stain wild-type and transgenic embryos. In contrast to the ubiquitous distribution of the smo mRNA (Figure 2a), immunohistochemical staining of wild-type embryos with the anti-SmoC antibody revealed a modulated distribution of the protein (Figure 2b). Smo accumulates in a series of sequentially repeating stripes, each of which is about onehalf a segment in width and spans the parasegmental boundary, the site of Hh activity (see Figure 2i). Staining of h–Gal4;UAS–smo embryos with the anti-SmoC antibody revealed a significant increase in the levels of Smo protein in the GAL4-expressing segments (Figure 2c); strikingly, however, this staining was also restricted to cells flanking the parasegmental boundary. To determine the precise location of these Smo-positive cells, we probed embryos
Brief Communication
1317
Figure 3 (a) Late stage 10 h–Gal4;UAS–Flag–smo embryo stained with anti-FLAG M2 antibody. The transgenic protein was detected in alternate segments throughout each hairy expression domain. (b) This contrasted with the distribution detected in genotypically identical embryos using the rabbit anti-SmoC antibody. Note, in particular, the single broad stripe of staining (asterisk) in the region corresponding to hairy domain 1 in (a) compared with the two narrow stripes (asterisks) in the same region in (b). (c) Stage 10 h–Gal4;UAS–smo–3HA embryo stained with anti-HA antibody. The stripes of HA-positive cells were broader than those detected with the anti-SmoC antibody, though those corresponding to hairy domains 2–4 were frequently weaker and narrower than the rest, mirroring the lower levels of GAL4 expression in these domains. Note, however, the continuous broad patch of staining in hairy domain 1 (asterisk), indicative of the Hh-independent nature of the expression pattern. (d) Schematic representation of the Smo protein showing the location of the FLAG tag (green), the HA tag (blue) and the region of the carboxy-terminal tail (red) against which the
(a)
(d) Anti-FLAG antibody
(b) Anti-HA antibody
Anti-HA antibody
Anti-FLAG antibody
Anti-SMO-C antibody
(c)
Current Biology
rabbit anti-SmoC antibody was raised. The protein is shown in two hypothetical conformations, inactive (left) and active (right). Both of the epitope tags appear equally
simultaneously with the anti-SmoC antibody and monoclonal antibodies for Engrailed (En) [17], a marker of Hhexpressing cells [18], and Wg [19], a marker of Hh-responding cells [14]. Double staining with the antiEn antibody revealed that Smo accumulates in and around cells expressing En (Figure 2d,h); double staining with the anti-Wg antibody showed that Smo also accumulates in cells expressing Wg (Figure 2g). The accumulation of Smo in and around cells secreting Hh protein, strongly suggests that the translation and/or stability of Smo is promoted by Hh activity. To test this possibility, we analysed Smo distribution in embryos in which Hh is ectopically expressed under the control of the Kr promoter [20]. Such embryos displayed a ubiquitous expression of Smo between parasegments 5 and 9, precisely the region in which ectopic Hh expression is driven by the Kr–Gal4 driver (Figure 2f). As Hh acts by inhibiting Ptc activity, the effects of Hh on Smo would be expected to be mediated by Ptc. In embryos homozygous for a ptc loss-of-function mutation, the modulated pattern of staining typical of the wild type was lost, indicating that Smo protein accumulates uniformly in the absence of Ptc activity (Figure 2e). The simplest interpretation of these data is that Ptc functions to block the translation or promote the degradation of Smo. And, as the spatial distribution of Smo was unaltered in h–Gal4;UAS–smo transgenic embryos, it would follow that Ptc activity can suppress accumulation of Smo
accessible in both conformations, but the epitope(s) recognised by the anti-SmoC antibody are accessible only in the active form, which is promoted by Hh signalling.
protein independently of the levels at which the gene is transcribed. This effect of Hh/Ptc-mediated signalling on Smo accumulation provides a simple explanation for the lack of an effect of ectopic smo expression, namely that the exogenous protein never accumulates outside the normal domain of Smo activity. Surprisingly, however, when h–Gal4;UAS–smo embryos were probed with the anti-FLAG antibody, a strikingly different pattern of exogenous protein accumulation was observed (Figure 3a,b). In contrast to the narrow stripes detected by the anti-SmoC antibody, staining was seen throughout each h–Gal4 expression domain. This indicates that the smo mRNA is translated in all cells in which it is transcribed. It follows that the Ptc-dependent staining pattern revealed by the anti-SmoC antibody reflects a posttranslational modification of the Smo protein. One possibility is that Ptc could promote the cleavage of Smo, yielding a relatively stable but functionally inert truncated form of the protein in cells not exposed to Hh. In this connection, it is interesting to note that the SREPB cleavage activating protein (SCAP), with which Ptc shares some homology, acts to promote the cleavage of SREBP [21] by chaperoning the latter from the endoplasmic reticulum to the Golgi [22,23]. Alternatively, however, it could be that Ptc induces a modification of the Smo protein that masks the epitope recognised by the anti-SmoC antibody; binding of Hh to Ptc would suppress this modification, activating the protein and making it accessible to the antibody. To discriminate between these two possibilities, we generated a
1318
Current Biology Vol 10 No 20
second tagged form of Smo, in this case inserting a haemagglutinin (HA) tag at the end of the carboxy-terminal tail (see Figure 3d). When embryos expressing this construct under h–Gal4 control were stained with an anti-HA antibody, a similar broad distribution of the tagged protein to that revealed by the anti-FLAG antibody was seen (Figure 3c). This argues against the cleavage model, but instead suggests that the carboxy-terminal tail undergoes a Ptc-dependent modification. As our anti-SmoC antibody was raised against an unmodified bacterially expressed protein, it seems most likely that this modification results in a conformational change that exposes epitopes recognised by the antibody. In this regard, it is notable that a putative dominant gain-of-function mutation in the human Smo protein is predicted to change the conformation of the equivalent region of the carboxy-terminal tail against which the anti-SmoC antibody is directed [24].
Materials and methods Generation of epitope tagged Smo transgenes A full-length smo cDNA was assembled from genomic clones [6]. A fragment encoding the prolactin signal peptide and FLAG tag was amplified from thrombin receptor cDNA (a generous gift from Shaun R. Coughlin, University of California at San Francisco) and used to replace the endogenous sequence encoding Smo signal peptide. The smo–HA transgene was generated by cloning the triple HA tag present in the shuttle plasmid pRD67 [11] into the XhoI site of the smo cDNA.
Generation and purification of anti-Smo antibody A fragment that encodes a region of the carboxy-terminal tail of Smo from residues 646 to 881 was amplified from the smo cDNA and ligated in-frame with NdeI/BamHI-digested pET14b (Novagen). ITPG induction of bacteria transformed with this plasmid yielded a His-tagged fusion protein of ~30 kD. Inclusion bodies were purified from induced cultures following standard protocols [25] and used to immunise rabbits. Antibodies specific for Smo were purified from the serum by immunoaffinity purification [25] using the bacterially expressed fragment fusion protein coupled to an NHS-activated Sepharose column (Hi Trap NHS-activated, Amersham-Pharmacia Biotech). Acidic and basic elutions were pooled and concentrated using an Amicon concentrator.
Antibody stainings and microscopy Embryos were fixed and stained using standard protocols. The affinitypurified anti-Smo antibody was used at 1:50–1:100. Monoclonal antiFLAG antibodies M1 and M2 (Sigma) were used at a dilution of 1:200. Monoclonal anti-En (4D9) and anti-Wg (4D4) were used at 1:100 dilution. Monoclonal anti-β-galactosidase was used at 1:1000. Specimens were examined and photographed using a Zeiss Axioplan 2 microscope. Figures were composed using Adobe Photoshop 4.0 software on Macintosh computers.
2
3. 4.
5.
6. 7. 8. 9. 10. 11. 12.
13. 14. 15. 16. 17. 18. 19. 20. 21. 22.
Supplementary material Supplementary material including full methodological details is available at http://current-biology.com/supmat/supmatin.htm.
23.
Acknowledgements
24.
This work was supported by a Wellcome Trust Programme grant to P.W.I. S.N. was a Research Fellow of the Swedish Foundation for International Cooperation in Research and Higher Education.
25.
References 1. Hammerschmidt M, Brook A, McMahon A: The world according to hedgehog. Trends Genet 1997, 13:14-21.
Nakano Y, Guerrero I, Hidalgo A, Taylor AM, Whittle JRS, Ingham PW: The Drosophila segment polarity gene patched encodes a protein with multiple potential membrane spanning domains. Nature 1989, 341:508-513. Hooper J, Scott MP: The Drosophila patched gene encodes a putative membrane protein required for segmental patterning. Cell 1989, 59:751-765. Goodrich LV, Johnson RL, Milenkovic L, McMahon JA, Scott MP: Conservation of the hedgehog/patched signaling pathway from flies to mice — induction of a mouse patched gene by hedgehog. Genes Dev 1996, 10:301-312. Alcedo J, Ayzenzon M, Vonohlen T, Noll M, Hooper JE: The Drosophila smoothened gene encodes a 7-pass membraneprotein, a putative receptor for the hedgehog signal. Cell 1996, 86:221-232. van den Heuvel M, Ingham PW: smoothened encodes a serpentine protein required for hedgehog signaling. Nature 1996, 382:547-551. Stone D, Hynes M, Armanini M, Swanson T, Gu Q, Johnson R, et al.: The tumor-suppressor gene patched encodes a candidate receptor for Sonic hedgehog. Nature 1996, 384:129-134. Murone M, Rosenthal A, de Sauvage F: Sonic hedgehog signaling by the patched–smoothened receptor complex. Curr Biol 1999, 9:76-84. Ingham PW, Taylor AM, Nakano Y: Role of the Drosophila patched gene in positional signalling. Nature 1991, 353:184-187. Chen Y, Struhl G: Dual roles for patched in sequestering and transducing hedgehog. Cell 1996, 87:553-563. Marigo V, Davey R, Zuo Y, Cunningham J, Tabin C: Biochemicalevidence that Patched is the Hedgehog receptor. Nature 1996, 384:176-179. Carpenter D, Stone DM, Brush J, Ryan A, Armanini M, Frantz G, et al.: Characterization of two patched receptors for the vertebrate hedgehog protein family. Proc Natl Acad Sci USA 1998, 95:13630-13634 Brand AH, Perrimon N: Targeted gene expression as a means of altering cell fates and generating dominant phenotypes. Development 1993, 118:401-415. Ingham PW: Localised hedgehog activity controls spatially restricted transcription of wingless in the Drosophila embryo. Nature 1993, 366:560-562. Martinez Arias A, Baker NE, Ingham PW: Role of segment polarity gene in the definition and maintenance of cell states in the Drosophila embryo. Development. 1988, 103:157-170. Vincent JP, Girdham C: Promoters to express cloned genes uniformly in Drosophila. Meth Mol Biol 1997, 62:385-392. Patel NH, Martin-Blanco E, Coleman KG, Poole SJ, Ellis MC, Kornberg TB, Goodman CS: Expression of engrailed proteins in arthropods, annelids, and chordates. Cell 1989, 58:955-968. Tabata T, Eaton S, Kornberg TB: The Drosophila hedgehog gene is expressed specifically in posterior compartment cells and is a target of engrailed regulation. Genes Dev 1992, 6:2635-2645. Brook WJ, Cohen SM: Antagonistic interaction between Wingless and Decapentaplegic responsible for dorsal-ventral pattern in the Drosophila leg. Science 1996, 273:1373-1377. Castelli-Gair J, Greig S, Micklem G, Akam M: Dissecting the temporal requirements for homeotic gene function. Development 1994, 120:1983-1995. Brown MS, Goldstein JL: A proteolytic pathway that controls the cholesterol content of membranes, cells, and blood. Proc Natl Acad Sci USA 1999, 96:11041-11048. DeBose-Boyd RA, Brown MS, Li WP, Nohturfft A, Goldstein JL, Espenshade PJ: Transport-dependent proteolysis of SREBP: relocation of site-1 protease from Golgi to ER obviates the need for SREBP transport to Golgi. Cell 1999, 23:703-712. Nohturfft A, DeBose-Boyd RA, Scheek S, Goldstein JL, Brown MS: Sterols regulate cycling of SREBP cleavage-activating protein (SCAP) between endoplasmic reticulum and Golgi. Proc Natl Acad Sci USA 1999, 96:11235-11240. Xie J, Murone M, Luoh SM, Ryan A, Gu Q, Zhang C, et al.: Activating Smoothened mutations in sporadic basal-cell carcinoma. Nature 1998, 391:90-92. Harlow E, Lane D: Antibodies: A Laboratory Manual. Cold Spring Harbor: Cold Spring Harbor Laboratory Press; 1988.