Gene name - sightless
Synonyms - central missing (cmn), skinny hedgehog, rasp Cytological map position - 63B5 Function - enzyme Keywords - Hedgehog pathway, segment polarity |
Symbol - sit
FlyBase ID: FBgn0035406 Genetic map position - 3-4.3 Classification - acyltransferase Cellular location - cytoplasmic |
Hedgehog (Hh) proteins are synthesized as full-length precursors that are autocatalytically cleaved by their C-terminal domains to release the signaling N-terminal domains. The addition of a cholesterol molecule to the C terminus of the signaling domain is concomitant with cleavage. Vertebrate Sonic hedgehog (Shh) proteins have also been shown to acquire a fatty acid chain on the N-terminal cysteine of this domain, which is required for a subset of their in vivo functions. A mutation of the corresponding cysteine in Drosophila Hh transforms it into a dominant-negative protein. In a mosaic screen for novel genes required for Drosophila photoreceptor differentiation, two alleles were identified of a gene that has been named sightless (sit) for its effect on photoreceptor development. sit is required for the activity of Drosophila Hh in the eye and wing imaginal discs and in embryonic segmentation. sit acts in the cells that produce Hh, but does not affect hh transcription, Hh cleavage, or the accumulation of Hh protein. sit encodes a conserved transmembrane protein with homology to a family of membrane-bound acyltransferases. The Sit protein could act by acylating Hh or by promoting other modifications or trafficking events necessary for its function (Lee, 2001b).
Clones of sit mutant cells in the eye disc show a reduction in the number of Elav-expressing photoreceptors that is most pronounced near the center of large clones, suggesting that sit might act nonautonomously. Because both sit alleles cause pupal lethality, the eye discs of third instar larvae transheterozygous for two sit alleles could be examined. In these discs, only a few cells are able to differentiate as photoreceptors. Rescue by adjacent wild-type tissue may thus contribute to the differentiation observed in sit mutant clones (Lee, 2001b).
One of the critical signals triggering photoreceptor development is Hedgehog (Hh), which is expressed at the posterior margin of the disc prior to differentiation and subsequently in the differentiating photoreceptors. Hh activates the expression of decapentaplegic (dpp) in a stripe at the front of differentiation, or morphogenetic furrow; Dpp signaling also promotes photoreceptor formation. dpp expression is lost from the morphogenetic furrow in sit mutant eye discs. Another target of Hh signaling, the proneural gene atonal, also requires sit for its expression. Despite this lack of Hh target gene expression, a hh-lacZ enhancer trap is expressed at the posterior margin of sit mutant eye discs, indicating that hh expression is established normally. This suggests that the sit phenotype could be due to a defect in Hh signaling (Lee, 2001b).
Hh signaling has been extensively studied in the wing disc, where hh is expressed in the posterior compartment and signals to cells just anterior to the compartment boundary to upregulate the expression of dpp and patched (ptc). The Hh signal is mediated by the stabilization and activation of the full-length form of the transcription factor Cubitus interruptus (Ci). This stabilization can be detected with an antibody directed against the C-terminal region of Ci, which fails to recognize the cleaved form of Ci produced in the absence of Hh signaling. sit mutant wing discs show defects consistent with a lack of Hh pathway function; ptc expression is not upregulated at the compartment boundary, and dpp expression is almost completely absent. In addition, no stabilization of full-length Ci could be detected at the compartment boundary. However, hh-lacZ is expressed at wild-type levels in sit mutant discs, indicating that hh transcription is unaffected. This implicates Sit in the Hh pathway downstream of hh transcription and upstream of Ci stabilization (Lee, 2001b).
The defects in sit mutant discs appear to be specific to anterior-posterior patterning; wingless (wg), which marks the dorsal-ventral compartment boundary, and its target gene Distal-less are still expressed in sit mutant wing discs. In addition, the phenotype is not as severe as the complete loss of hh from early stages of larval development: although sit discs are smaller than wild-type discs, they are larger and more normally shaped than hh mutant discs. The sit alleles are therefore likely to cause an incomplete or late loss of Hh signaling: since they appear to be nulls at the molecular level, this may be due to the activity of maternally contributed sit (Lee, 2001b).
Hh signaling is also required for normal embryonic segmentation -- hh mutant embryos show a loss of naked cuticle and of wg expression. When the maternal contribution of sit is removed by making germline clones, a loss of naked cuticle strongly resembling the hh phenotype is observed, however, a wild-type copy of sit provided on the paternal chromosome is able to fully rescue the phenotype. In embryos lacking both maternal and paternal sit, stripes of Wg expression are lost from the ectoderm by stage 11. Thus, sit is required for the expression of Hh target genes in the embryo as well as in the eye and wing discs (Lee, 2001b).
sit is required in the Hh-producing cells but does not affect the level of Hh protein. sit might affect Hh signaling by promoting the production of functional Hh or by allowing cells to respond to the Hh signal. To distinguish between these possibilities, mosaic analysis was used to determine in which cells sit function is required: in the wing disc, Hh-producing cells are restricted to the posterior compartment, and Hh-responding cells are restricted to the anterior compartment. Small clones of cells homozygous for sit have no effect on ptc or dpp expression in the wing disc, consistent with the nonautonomy of sit function in the eye disc. When the Minute technique was used to generate larger clones lacking sit, it was found that sit function is not required in the ptc-expressing cells or anywhere in the anterior compartment for ptc upregulation, provided that sit is present in the posterior compartment. The loss of sit from the posterior compartment prevents ptc upregulation in adjacent anterior cells even if they themselves are wild-type for sit. Thus, sit function in cells of the posterior compartment is both necessary and sufficient to upregulate ptc in anterior compartment cells. This suggests that sit may be required for the production, activity, or release of Hh protein (Lee, 2001b).
To determine whether Hh protein can be produced in the absence of sit function, sit mutant clones in the wing disc were stained with an antibody to the N-terminal domain of Hh. No change in the intensity of staining is apparent in sit mutant clones compared to adjacent wild-type tissue. Thus, sit is not required for Hh translation or stability. In clones that are mutant for dispatched (disp), which encodes a protein required for Hh release from the cell, Hh protein accumulates to high levels. No such accumulation of Hh is observed in sit mutant clones, suggesting that unlike Disp, Sit does not act at the level of Hh release (Lee, 2001b).
Hh is synthesized as a full-length precursor that is then cleaved by the autocatalytic activity of its C-terminal domain to release the N-terminal signaling domain. sit does not appear to be required for this cleavage, because similar proportions of full-length Hh and its cleaved N-terminal domain are detected on Western blots of extracts from sit mutant and wild-type third instar larvae. Whether the expression of an N-terminally truncated form of Hh (Hh-N) could rescue sit mutants was also tested; this form of the protein is not cholesterol-modified or restricted in its diffusion and does not require disp for its release from the cell. The expression of UAS-Hh-N with eyeless-GAL4 can induce premature photoreceptor differentiation in wild-type eye discs but does not alter the phenotype of sit mutant eye discs. These results suggest that sit is required for Hh activity, but not for its cleavage, cholesterol modification, or secretion (Lee, 2001b).
sit transcript is expressed uniformly at low levels in the imaginal discs and early embryo. The encoded protein has ten predicted transmembrane domains and shows homology to human, mouse, and C. elegans proteins present in the database. Its closest human homolog is BAA91772, to which it shows 28% identity and 45% similarity. In addition, the Sit protein shows more distant homology to a family of proteins that have been shown to transfer acyl chains onto hydroxyl groups of membrane-bound lipid or protein targets. An invariant histidine that has been suggested as a possible active-site residue is conserved in the Sit sequence, and both sit mutations truncate the protein prior to the region of acyltransferase homology (Lee, 2001b).
Since sit does not alter the level of Hh protein present in the cell, it is unlikely to affect Hh translation or release. The cleavage of Hh to release the N-terminal signaling domain also does not require sit, and an exogenously provided Hh-N domain is inactive in the absence of sit. sit is unlikely to be required for cholesterol addition to the C terminus of the signaling domain; bacterially produced Hh protein becomes cholesterol-modified in vitro, and this modification restricts Hh localization, but does not increase its activity in vivo. Human and rodent Sonic hedgehog (Shh) proteins have been shown to acquire a palmitoyl modification on the N-terminal cysteine of the signaling domain in cell culture. Mutation of this cysteine to serine in human Shh prevents its palmitoylation and greatly reduces its ability to ventralize the mouse forebrain. The corresponding cysteine to serine mutation in Drosophila Hh (Hh-C84S) completely abolishes its activity; however, the mutant protein appears to be secreted and appears to block the effects of wild-type Hh in the extracellular space. Together with the homology of Sit to acyltransferases, this raises the intriguing possibility that Sit might be the enzyme responsible for the palmitoylation of Hh. However, this would represent a difference in specificity between Sit and the other acyltransferases of this family, which acylate hydroxyl groups; the inactivity of Hh-C84S indicates that a hydroxyl group cannot act as a substrate in this case. Further biochemical analysis will be needed to determine whether Drosophila Hh is in fact palmitoylated and whether this palmitoylation requires sit function. Alternatively, Sit could be required for another modification or trafficking event required for Hh activity. Sit is distantly related to the Porcupine (Porc) protein, which is required in Wg-producing cells for Wg activity. Porc is localized to the endoplasmic reticulum and alters the glycosylation state of Wg. Interestingly, Porc also has homology to the membrane-bound O-acyltransferase family. The requirement of Porc for Wg function and Sit for Hh function suggests that modifications that are essential for the activity of signaling proteins may be more widespread than previously believed (Lee, 2001b).
CG11495, corresponding to sightless, is predicted to encode a protein that is 500 aa long and contains 10 transmembrane domains. Blast search identified homologs in mouse and human that share over 25% sequence identity to CG11495; the conservation spans the whole proteins. Sequence analysis of the CG11495-encoded protein suggests that it belongs to a family of membrane bound acyltransferases (Hofmann, 2000). Interestingly, this family also includes Porcupine, which is required for Wg secretion (Amanai, 2001).
date revised: 26 February 2002
Home page: The Interactive Fly © 1995, 1996 Thomas B. Brody, Ph.D.
The Interactive Fly resides on the
Society for Developmental Biology's Web server.