wingless

Directional transport of extracellular Wingless

Wg and Wnt molecules tightly associate with membrane and extracellular matrix and appear not to be readily soluble. Thus, it is unlikely that these proteins freely diffuse through extracellular spaces. Rather, Wg appears to be transported via active cellular processes. This phenomenon was first demonstrated using the shibire^ts (shi^ts) mutation to block endocytosis. shi encodes the fly dynamin homolog, a GTPase required for clathrin-coated vesicle formation. Rather than the broad, punctate Wg protein distribution normally found over several cell diameters on either side of the wg-expressing cells, shi mutant embryos show high level accumulation of Wg around the wg-expressing cells. Structure/function analysis of the Wg molecule further supports the idea that active transport of the ligand is essential. Four mutations within wg have been isolated that specifically disrupt Wg transport without abolishing signaling activity. These mutant molecules generate a restricted response within the segment, as assayed by both cuticular pattern elements and molecular events. Homozygous mutant embryos produce naked cuticle but little denticle diversity, and show narrowed domains of Wg protein distribution and Arm stabilization. Three of these four mutations are single amino acid substitutions; each affects a residue that is highly conserved throughout the Wnt family, suggesting that ligand transport may be an important general aspect of Wnt function (Moline, 1999 and references).

To assess the functional consequences of this broad Wg distribution, a means has been devised of perturbing endocytosis in spatially restricted domains within the embryo. A transgene expressing a dominant negative form of shibire (shi), the fly dynamin homolog, was constructed. When this transgene is expressed using the GAL4-UAS system, Wg protein distribution within the domain of transgene expression is limited and Wg-dependent epidermal patterning events surrounding the domain of expression are disrupted in a directional fashion. These results indicate that Wg transport in an anterior direction generates the normal expanse of naked cuticle within the segment and that movement of Wg in a posterior direction specifies diverse denticle cell fates in the anterior portion of the adjacent segment (Moline, 1999).

Interfering with posterior movement of Wg rescues the excessive naked cuticle specification observed in naked (nkd) mutant embryos. It is proposed that the nkd segment polarity phenotype results from unregulated posterior transport of Wg protein and therefore that wild-type Nkd function may contribute to the control of Wg movement within the epidermal cells of the segment (Moline, 1999).

Using en-Gal4-driven shi^D expression to reduce posterior movement of Wg suppresses the phenotype of the segment polarity mutation, naked. nkd mutant embryos secrete denticle belts that have essentially normal denticle type diversity but that are replaced to varying degrees by naked cuticle. This excess naked cuticle depends upon Wg activity levels. The wg;nkd double mutant shows no naked cuticle across the ventral region and reducing the dosage of wg in a nkd mutant restores denticle belts. Thus wild-type nkd gene function appears to be involved in limiting Wg signaling activity within the segment. Consistent with this idea, Wg target genes become ectopically expressed in nkd mutant embryos. The en expression domain expands 2-3 cell diameters during stage 9, and an ectopic stripe of wg expression arises at stage 10, in the row of cells posterior to this expanded en domain. The posterior expansion of en expression suggests that nkd might play a role in restricting the movement of Wg protein in a posterior direction. Indeed, when nkd mutant embryos were generated that express shi^D at moderate levels in the en domain, there was a dramatic reduction in the amount of naked cuticle specified. These embryos are very similar in appearance to wild-type embryos in which en-Gal4 drives shi^D expression except that the nkd mutant head defect is not fully rescued. en-Gal4-driven shi^D expression also prevents the ectopic activation of en expression in nkd mutants. The stripes of en expression in the thoracic and abdominal segments are restored to the normal width, although some expansion is still observed in the head segments (Moline, 1999).

Since wild-type Wg signaling activity is required for stabilization of en expression, En stripes of normal width indicate that sufficient functional Wg contacts both rows of en-expressing cells to produce normal target gene regulation. This result demonstrates that expression of shi^D does not interfere with Wg signal transduction and supports the idea that moderate level shi^D expression reduces, but does not eliminate, transport of Wg across the affected domain. In contrast, embryos expressing high level shi^D in the en domain show a narrowed stripe of En antibody staining, suggesting that Wg can no longer traverse the first row of en-expressing cells to stabilize en in the second row. However, because of the severe effects of a more complete endocytotic block, these embryos do not secrete cuticle properly and so the effects on cuticle pattern are not interpretable (Moline, 1999).

During early stages of wild-type embryogenesis, Wg protein can be detected at high levels in cells both anterior and posterior to the wg-expressing row of cells. Diversity of denticle types, as well as stabilization of en expression in the adjacent cells, are specified by Wg activity during these early stages of embryonic development. By mid-stage 10, when Wg is no longer required for denticle specification or en stabilization, the Wg protein distribution shifts and Wg appears to be excluded from the en-expressing cells. This exclusion is not observed in nkd mutants at the same stage. Rather, Wg protein continues to be detected in cells on either side of the wg-expressing row of cells and the levels become substantially higher due to the ectopic stripe of wg expression. These results suggest that nkd gene function may play a role in the posterior restriction of Wg protein that occurs during stage 10. Hence the mutant phenotype is rescued dramatically when this restriction is produced artificially, by expressing shi^D in the en-expressing row of cells. All stage 11 and 12 embryos derived from this cross show posterior restriction of Wg protein, indicating that the nkd homozygotes do not exhibit excess posterior movement of Wg under these conditions. It is suspected that, in wild-type embryos, this restrictive function is not limited to the en-expressing cells. If this were the case, then one would expect to observe excess naked cuticle replacing denticle belts when wg+ is expressed in the en domain. Instead, en-Gal4-driven wg+, either alone or when co-expressed with shi^D, does not produce substantial amounts of ectopic naked cuticle. Thus, it seems likely that some ability to restrict posterior Wg movement during later stages is shared by the rows of cells at the anterior of each segment (Moline, 1999).

It is believed that this analysis of Wg transport by perturbing endocytosis is physiologically relevant because a similar inhibition of transport can be produced by overexpressing Dfz2, the cognate receptor for Wg. It is presumed that these effects result from sequestering ligand, because pattern defects are observed only when Wg levels are limiting. No change from the wild-type cuticle pattern is detected when Dfz2 is driven at ubiquitous high levels of expression with E22C-Gal4. However, in embryos heterozygous for a null mutation of wg, significant pattern defects are observed at a frequency of 60%. Ectopic denticles appear in the domain of cells that normally secrete naked cuticle, similar to what is observed in segments where anterior Wg transport is perturbed by shi^D. These pattern defects caused by Dfz2 overexpression are accompanied by a restricted Wg protein distribution and by a narrowed domain of Arm stabilization. However, it is not possible to directly compare Dfz2 with shi^D in this experiment. E22C-Gal4-driven expression of shi^D , even with UAS lines that express at low levels, results in cell death and failure to secrete cuticle as was the case with the original shi^ts mutation at restrictive temperature (Moline, 1999).

Patterning and growth control by membrane-tethered Wingless

Wnts are evolutionarily conserved secreted signalling proteins that, in various developmental contexts, spread from their site of synthesis to form a gradient and activate target-gene expression at a distance. However, the requirement for Wnts to spread has never been directly tested. This study used genome engineering to replace the endogenous wingless gene, which encodes the main Drosophila Wnt, with one that expresses a membrane-tethered form of the protein. Surprisingly, the resulting flies were viable and produced normally patterned appendages of nearly the right size, albeit with a delay. In the prospective wing, prolonged wingless transcription followed by memory of earlier signalling allows persistent expression of relevant target genes. It is suggested therefore that the spread of Wingless is dispensable for patterning and growth even though it probably contributes to increasing cell proliferation (Alexandre, 2004).

Engrailed and Hedgehog make the range of Wingless asymmetric in Drosophila embryos

The spread of Wingless within the embryonic epidermis of Drosophila was examined in order to understand the role of Wingless in the segmentation. Using two assays for Wingless activity (specification of naked cuticle and repression of rhomboid transcription), it was found that Wingless acts at a different range in the anterior and posterior directions. This asymmetry follows in part from differential distribution of the Wingless protein. Transport or stability is reduced within engrailed-expressing cells, and farther posteriorward Wingless movement is blocked at the presumptive segment boundary and perhaps beyond. The role of hedgehog in the formation of this barrier is demonstrated (Sanson, 1999).

It is proposed that asymmetric Wingless distribution ensures the establishment of well-differentiated cell fates on either side of the engrailed domain. Anteriorly, at the wingless source, rhomboid expression is repressed. In contrast, reduced Wingless movement and/or stability within the engrailed domain allows nascent posterior rhomboid expression. Around this time (stage 11), a barrier to Wingless that requires hedgehog signaling forms at the posterior of the engrailed domain and ensures that Wingless does not foray across and repress rhomboid. rhomboid then activates the Egfr pathway within its expression domain and in adjacent cells. It may be that rhomboid itself contributes to barrier formation and thus builds a line of defense against invasion by its repressor. In addition, activation of the EGF pathway by rhomboid would antagonize any Wingless leaking through. Denticle formation requires transcription of shavenbaby, which is under positive regulation by the Egfr pathway and negative regulation by the wingless pathway. Activated EGFR and the absence of Wingless posterior to the engrailed domain allow shavenbaby expression and hence denticle formation. At the anterior side, converse conditions exist, since Wingless is present at high levels and the Egfr pathway is inactive. Therefore, it is proposed that polarization of Wingless transport by engrailed and hedgehog guarantees the naked fate anterior to the engrailed domain and the denticle fate posteriorly, and thus establishes the anteroposterior polarity of each segment (Sanson, 1999 and references).

In the ventral abdominal region of the Drosophila embryo, wingless is expressed in single cell-wide stripes. To assess the range of Wingless in specifying the naked fate, these stripes were mapped onto the final cuticle pattern. Unexpectedly, the wingless stripes were found to be are eccentric within each expanse of naked cuticle. Naked cuticle is made over a distance of approximately 3-4 cell diameters anterior to the wingless source. In contrast, posterior to it, only the adjoining cells make naked cuticle; these cells are the most anterior of each engrailed stripe. The denticle fate of more posterior engrailed-expressing cells could be explained if they were unable to respond to wingless. Therefore the responsiveness to Wingless of all epidermal cells was assessed. To ask whether Wingless is required beyond one cell diameter posterior to its source, in individual cells the ability to respond to Wingless was removed and the phenotypic consequences was analyzed at single-cell resolution. Armadillo is a downstream effector of Wingless signal transduction and is also a component of Cadherin-based adhesion complexes. Armadillo is titrated out of its signaling pool when Shotgun (DE-Cadherin) is overexpressed, and in the embryonic epidermis, this leads to a phenocopy of a wingless mutation. Overexpressed Shotgun, therefore, blocks the response to Wingless in a cell-autonomous manner. Shotgun was overexpressed in single cells with the "Flp-on Gal4" system. Clones of cells expressing Gal4, which in turn activates expression of Shotgun and GFP, were induced during early stage 11 and scored at the end of embryogenesis. GFP-positive cells within naked regions make ectopic denticles, confirming that wingless signaling is required everywhere naked cuticle is made. All Shotgun-overexpressing cells located within belts make denticle of the type and size expected for their position. These results show that wingless signaling (via Armadillo) is not required in the presumptive denticle belts. Wingless signaling is only required within the naked domain, and this requirement is asymmetric relative to the wingless source (Sanson, 1999).

Two types of mechanisms could account for the asymmetry of wingless action: (1) the wingless protein itself could be unequally distributed in the anterior and posterior directions, or (2) Wingless could be distributed symmetrically, but downstream signaling would be repressed posterior to the source. Immunocytochemistry has revealed an asymmetric distribution of the wingless-containing vesicles within each segment at stage 11. By colabeling with anti-Engrailed, it has been shown that the posterior transition in Wingless distribution occurs at the interface with engrailed-expressing cells. Although Wingless protein can be detected in the wingless domain and anterior to it, only a small number of Wingless-containing vesicles can be seen in the most anterior row of engrailed cells, and none can be seen more posteriorly. Thus, Wingless distribution is asymmetric, and this could explain the asymmetry of wingless action (Sanson, 1999).

It is conceivable that undetectable yet active Wingless is present in cells posterior to engrailed stripes. To demonstrate the absence of active Wingless there, a functional assay was used, based on the finding that wingless signaling represses rhomboid expression. From stage 11 onward, rhomboid is expressed in stripes just posterior to each engrailed domain, and this expression is abolished by continuous and uniform expression of Wingless. Later ectopic expression, induced at late stage 11, inhibits rhomboid transcription only in the midventral region, and if sibling embryos are left to develop, they make ectopic naked cuticle in the same region. Therefore, Wingless can repress rhomboid transcription in the same time window as it specifies naked cuticle. Wingless is not only sufficient for rhomboid repression, it is also necessary since wingless null mutants have an additional rhomboid stripe in each abdominal segment. The position of these extra stripes relative to landmarks in the CNS suggests that they form at the anterior of the domain of extinct engrailed expression, where wingless would normally be expressed. Thus, in the wild type, the presence of Wingless at the anterior of each engrailed stripe maintains the silence of rhomboid expression there. Significantly, rhomboid is expressed posterior to the engrailed domain of wild-type embryos. Therefore, active Wingless is not present in these cells, at least at late stage 11; if it were, rhomboid would not be expressed. These cells are located only two cell diameters posterior to the Wingless source (Sanson, 1999).

The asymmetric distribution of Wingless could be explained by decreased transport/stability either within the engrailed domain or at its posterior edge, where the segment boundary forms. To explore this, wingless was misexpressed directly in the engrailed domain (posterior to endogenous wingless) and it was determined whether the range of Wingless was shifted posteriorly. Wingless was expressed with the engrailed-Gal4 driver in otherwise wild-type embryos. The only effect on the cuticle pattern is the loss of row 1 denticles. Remarkably, no other denticles are lost. In particular, row 2 denticles are present even though they are adjacent to the Wingless-misexpressing cells. Thus wingless expressed at the anterior side of the presumptive segment boundary does not affect the fate of cells on the posterior side. To confirm this finding, rhomboid expression was used as an early molecular marker for the absence of Wingless. In the wild-type larva, rhomboid is expressed in the cells secreting rows 2-4. In en-Gal4/UAS-wg larvae, this expression is unchanged, indicating that the wingless pathway is not operative in the cells immediately posterior to the wingless-misexpressing cells. Thus, it appears that Wingless cannot cross the posterior edge of the engrailed domain. This was verified by looking directly at the distribution of Wingless protein in en-Gal4/UAS-wg embryos. In these embryos, Wingless is present within the domain of wingless misexpression, as expected. However, it is not detectable posterior to the engrailed-expressing cells. It is concluded that a barrier to Wingless protein movement exists at the presumptive segment boundary (Sanson, 1999).

Among various candidate genes, hedgehog was found to be required for the posterior barrier to Wingless. Assaying the range of Wingless in a hedgehog mutant is not straightforward, since wingless expression requires hedgehog signaling. Therefore, wingless expression was maintained artificially in hedgehog null mutants using en-Gal4. Normally the engrailed promoter also turns off in a hedgehog mutant for lack of wingless, but in en-Gal4/UAS-wg; hh^- embryos, this is remedied by exogenous Wingless. Thus, a hedgehog-independent positive feedback loop is established between engrailed and wingless, and stripes coexpressing Wingless and Engrailed are obtained. The distribution of the Wingless protein in en-Gal4/UAS-wg; hh^- embryos is different from that seen in en-Gal4/UAS-wg control embryos. Wingless spreads posterior to the engrailed domain as if a barrier had been lifted or Wingless movement enhanced. The resulting protein distribution is symmetrical, and this is reflected in the cuticle pattern: in contrast to en-Gal4/UAS-wg embryos, en-Gal4/UAS-wg; hh^- embryos lack rows 2-4 and, instead, have an extra expanse of naked cuticle. At the positions where rows 5 and 6 normally form, lies a thin stripe of small denticles. Naked cuticle is specified equally in the anterior and posterior directions, as shown by marking the wingless-expressing cells with GFP. Thus, in the absence of hedgehog, wingless action is symmetric (Sanson, 1999).

Thus, loss of hedgehog signaling increases the range of Wingless. Now it was asked whether the converse is true. To assay the range of Wingless in the presence of excess hedgehog signaling, endogenous wingless must be removed because hedgehog signaling activates wingless expression and this would confuse the assay. Therefore, the en-Gal4/UAS-wg combination was used again, but this time in a wingless mutant background. The sole source of Wingless in these embryos is in the engrailed domain. The wingless mutant phenotype is significantly rescued: the normal alternation of denticle belts and naked cuticle is restored, and many belts are nearly wild type, except for the loss of row 1. In these embryos, the width of the band of naked cuticle is 4-5 cells, and this provides an assay for the anterior range of Wingless. This assay was validated with a version of Wingless expected to act only at short range. If a membrane-tethered form of Wingless is expressed instead of the wild-type protein, an expanse of naked cuticle only 1-2 cells wide is found. This demonstrates that Wingless has to be physically transported from cell to cell to specify a band of naked cuticle of the normal size, and that there is no relay mechanism. Next, the assay was used to find out the effect of increasing hedgehog signaling on the range of Wingless. Increased hedgehog signaling can be achieved either by overexpressing Hedgehog or by removing patched activity. wg^- en-Gal4/UAS-wg embryos carrying in addition UAS-hedgehog have significantly narrower naked domains. Likewise, wg^- ptc^- en-Gal4/UAS-wg embryos have narrow naked bands as well. This suggests that excess hedgehog signaling reduces the range of Wingless, although excess Hedgehog signaling could also induce ectopic rhomboid, which would in turn antagonize Wingless signaling and bring about the loss of naked cuticle (Sanson, 1999).

It is known that Wingless sustains engrailed expression only in adjoining cells, suggesting that Wingless is not readily transported across the engrailed domain. This is supported by the asymmetric distribution of the protein. Immunostaining reveals the presence of Wingless anterior to its source, whereas very little is detected posteriorly; posterior to wingless-expressing cells, in the engrailed domain, some Wingless staining is found but only in the most anterior cells (nearest the Wingless source). Thus, engrailed-expressing cells appear to restrict Wingless movement. Restricted Wingless transport through the engrailed domain could be explained by the downregulation of a specialized transport receptor in the engrailed cells; the existence of such a receptor has been hypothesized. Alternatively, inefficient transport could follow from selective instability of Wingless or its sequestration within the engrailed domain. In wing imaginal discs, the stability and range of Wingless increase in response to overexpression of its receptor Frizzled2. By analogy, and conversely, Wingless might be particularly unstable within engrailed stripes for lack of a receptor there. Alternatively, the surface or extracellular matrix surrounding engrailed cells might trap Wingless and impede its movement. A receptor of the proteoglycan type could possibly mediate this activity. Indeed, in mutants for the gene encoding UDP-glucose dehydrogenase, that lack HSPGs, embryonic engrailed stripes are temporarily widened, implying an increased range of Wingless. Identification of the relevant receptors and their pattern of expression will be required to discriminate between the above alternatives (Sanson, 1999 and references).

Not only is Wingless movement restricted within the engrailed domain, but a barrier seems to exist at its posterior edge. This is especially evident in embryos that ectopically express Wingless in the engrailed domain. In these embryos, Wingless does not specify naked cuticle nor repress rhomboid posteriorly, even in adjacent cells. The lack of response is unlikely to be due to insufficient expression, since en-Gal4 is a robust driver. Also, uniform wingless expression (even at low levels and up to late stage 11) induces uniform naked cuticle (D) and represses rhomboid transcription. This suggests that all cells, including those posterior to each engrailed stripe, are responsive to Wingless (although it is formally possible that the latter cells are only responsive to autocrine signaling). Thus, the lack of posterior response in en-Gal4/UAS-wg embryos is probably because, in this experimental situation, Wingless does not reach posteriorly. Indeed, in the same embryos, immunostaining fails to detect Wingless protein posterior to the engrailed domain. The best interpretation of the results is that a barrier to Wingless movement exists at the segment boundary, although the possibility that movement is impeded throughout the rhomboid expression domain or that these cells are unable to respond to paracrine Wingless cannot completely excluded (Sanson, 1999).

The notion that Wingless movement is blocked at the forming segment boundary contrasts with an earlier proposal that Wingless spreads symmetrically. According to this view, posterior to its source, Wingless signaling is antagonized by active EGFR. The Egfr pathway is activated within and near the rhomboid stripe, which lies just posterior to the segment boundary. However, it is proposed that this segmental activation, which requires rhomboid, occurs after formation of the restrictions to Wingless movement. If wingless protein were present in the rhomboid cells at late stage 11, rhomboid expression would not be allowed there since wingless has been shown to repress rhomboid transcription. Subsequent establishment of rhomboid expression would further counteract activation of the Wingless pathway in prospective denticle belts (Sanson, 1999).

It is suggested that two mechanisms restrict posterior Wingless movement. The first restriction occurs within the engrailed domain and is unlikely to be under hedgehog control, since engrailed cells are not thought to respond to Hedgehog. Rather, engrailed could implement this restriction by controlling a gene involved in Wingless transport, sequestration, or stability. By contrast, the barrier at the posterior of the engrailed domain requires hedgehog signaling. Wingless produced ectopically in the engrailed domain of hedgehog mutants is allowed to invade posteriorly located cells and induce naked cuticle there. The finding that the same effects are seen in cubitus interruptus mutants indicates that the hedgehog signaling pathway is involved. The role of the hedgehog pathway is confirmed by "gain-of-function" experiments. Loss of patched results in overactivation of the hedgehog pathway and so does excessive hedgehog expression. Both situations reduce the range of Wingless in the anterior direction as if the spread of the protein were reduced. It is presumed that, in the wild type, a downstream Hedgehog target is upregulated at the posterior of each engrailed/hedgehog stripe and this would lead to Wingless destabilization or a block to transport there (Sanson, 1999).

The extracellular matrix, glycosaminoglycan and Wingless

When expressed in S2 cells, the majority (approximately 83%) of secreted Wingless protein (WG) is bound to the cell surface and extracellular matrix through specific, noncovalent interactions. The tethered WG can be released by addition of exogenous heparin sulfate and chondroitin sulfate glycosaminoglycans. WG also binds directly to heparin agarose beads with high affinity. These data suggest that WG can bind to the cell surface via naturally occurring sulfated proteoglycans. Two lines of evidence indicate that extracellular glycosaminoglycans on the receiving cells also play a functional role in WG signaling: (1) treatment of WG-responsive cells with glycosaminoglycan lyases reduces WG activity by 50%; (2) when WG-responsive cells are preincubated with 1 mM chlorate, which blocks sulfation, WG activity is inhibited to near-basal levels. Addition of exogenous heparin to the chlorate-treated cells restores WG activity. Based on these results, it is proposed that WG belongs to the group of growth factor ligands whose actions are mediated by extracellular proteoglycan molecules (Reichsman, 1996).

The Drosophila UDP-glucose dehydrogenase gene is involved in wingless signaling. UDP-glucose dehydrogenase is an enzyme responsible for the production of UDP-glucuronate. UDP-glucuronate, a uronic acid, is a precursor in glycosaminoglycan (GAG) biosynthesis. GAG is a complex carbohydrate which is linked to a distinct class of glycoproteins called proteoglycans. A mutation in UDP-glucose dehydrogenase, called kiwi, generates a phenotype identical to that of wingless. By rescuing the kiwi phenotype with both UDP-glucuronate and the glycosaminoglycan heparan sulfate, it has been shown that kiwi function in the embryo is crucial for the production of heparan sulfate in the extracellular matrix. Injection of heparin degrading enzyme, heparinase (and not chondroitin, dermatan or hyaluronic acid degrading enzyme) into wild-type embryos leads to the degradation of heparin-like glycosaminoglycans and a 'wingless-like' cuticular phenotype. Heparin-like GAGs are complex acidic polysaccharides that are important soluble components of the extracellular matrix. They provide the necessary hydration to the ECM scaffold and hence solubilize and modulate the function of such proteins as fibroblast growth factor. This study provides the first genetic evidence for the involvement of heparin-like glycosaminoglycans in signal transduction (Binari, 1997).

In vitro experiments suggest that glycosaminoglycans (GAGs) and the proteins to which they are attached (proteoglycans) are important for modulating growth factor signaling. However, in vivo evidence to support this view has been lacking, in part because mutations that disrupt the production of GAG polymers and the core proteins have not been available. The identification and characterization of Drosophila mutants in the suppenkasper (ska) gene is described. The ska gene encodes UDP-glucose dehydrogenase, which produces glucuronic acid, an essential component for the synthesis of heparan and chondroitin sulfate. ska mutants fail to put heparan side chains on proteoglycans such as Syndecan. Surprisingly, mutant embryos produced by germ-line clones of this general metabolic gene exhibit embryonic cuticle phenotypes strikingly similar to those that result from loss-of-function mutations in genes of the Wingless (Wg) signaling pathway. Zygotic loss of ska leads to reduced growth of imaginal discs and pattern defects similar to wg mutants. In addition, genetic interactions of ska with wg and dishevelled mutants are observed. These data demonstrate the importance of proteoglycans and GAGs in Wg signaling in vivo and suggest that Wnt-like growth factors may be particularly sensitive to perturbations of GAG biosynthesis (Haerry, 1997).

The Drosophila gene sugarless encodes a homolog of vertebrate UDP-glucose dehydrogenase. This enzyme is essential for the biosynthesis of various proteoglycans. In the absence of both maternal and zygotic activities of this gene, mutant embryos develop with segment polarity phenotypes reminiscent of the loss of either Wingless or Hedgehog signaling. To analyze the function of Sugarless in cell-cell interaction processes, attention has been focused on the requirement of Sugarless for Wingless signaling in different tissues. sugarless mutations impair signaling by Wingless, suggesting that proteoglycans contribute to the reception of Wingless. Overexpression of Wingless can bypass the requirement for sugarless, suggesting that proteoglycans modulate signaling by Wingless, possibly by limiting its diffusion and thereby facilitating the binding of Wingless to its receptor. Tissue-specific regulation of proteoglycans may be involved in regulating both Wingless short- or long-range effects. For example, Dally (a Glypican-related heparin sulfate proteoglycan) mutations show wing notching with loss of wing margin structures, an effect seen in wingless and dishevelled mutants, suggesting the potential involvement of Dally in Wg signaling. A Drosophila homolog of vertebrate Syndecans has also been characterized. Syndecan is a transmembrane HSPG and represents the major source of HSPGs in epithelial cells. Syndecan has been demonstrated to function as a coreceptor for FGF signaling. Further genetic and biochemical studies should reveal whether Dally and/or Syndecan play a direct role in Wg signaling (Hacker, 1997).

Wingless is a member of the Wnt family of growth factors, secreted proteins that control proliferation and differentiation during development. Studies in Drosophila have shown that responses to Wg require cell-surface heparan sulphate, a glycosaminoglycan component of proteoglycans. These findings suggest that a cell-surface proteoglycan is a component of a Wg/Wnt receptor complex. The protein encoded by the division abnormally delayed (dally) gene is a cell-surface, heparan-sulphate-modified proteoglycan. dally partial loss-of-function mutations compromise Wg-directed events, and disruption of dally function with RNA interference produces phenotypes comparable to those found with RNA interference of wg or frizzled/Dfz2. Ectopic expression of Dally potentiates Wg signalling without altering levels of Wg and can rescue a wg partial loss-of-function mutant. Dally, a regulator of Decapentaplegic (Dpp) signalling during post-embryonic development, has tissue-specific effects on Wg and Dpp signalling. Dally can therefore differentially influence signalling mediated by two growth factors, and may form a regulatory component of both Wg and Dpp receptor complexes (Tsuda, 1999).

Heparan sulfate proteoglycans are critical for the organization of the extracellular distribution of Wingless

Recent studies in Drosophila have shown that heparan sulfate proteoglycans (HSPGs) are required for Wingless (Wg/Wnt) signaling. In addition, genetic and phenotypic analyses have implicated the glypican gene dally in this process. Another Drosophila glypican gene, dally-like (dly) has been identified and it is also involved in Wg signaling. Inhibition of dly gene activity implicates a function for DLY in Wg reception -- overexpression of DLY leads to an accumulation of extracellular Wg. It is proposed that DLY plays a role in the extracellular distribution of Wg. Consistent with this model, a dramatic decrease of extracellular Wg was detected in clones of cells that are deficient in proper glycosaminoglycan biosynthesis. It is concluded that HSPGs play an important role in organizing the extracellular distribution of Wg (Baeg, 2001).

One glypican molecule, Dally, has been implicated in Wg signaling. However, because numerous glypican genes are present in other animals, a search was carried out of the Drosophila database for additional glypican family members. One EST clone showed some sequence similarity to dally and a full length cDNA was cloned. The sequence of the cDNA revealed a potential open reading frame of 765 amino acid residues, with 22% and 35% identity to Dally and mouse K-glypican, respectively. The predicted primary structure of the molecule exhibits the hallmarks of a glypican protein. The hydrophilicity plot of the new molecule is similar to those of the other members of the glypican family, which is characterized by the presence of NH2- and COOH-terminal hydrophobic signal sequences. In addition, this molecule possesses four consensus serine/glycine dipeptide sequences for glycosaminoglycan (GAG) attachment sites, and a signal sequence for a GPI-moiety attachment site at the COOH-terminal region. Moreover, the number and position of cysteine residues, which are a unique feature of glypican family members, are almost completely conserved in the predicted protein. These results led to the identification of Dally-like. Hybridization using a dly-specific probe to polytene chromosomes from salivary glands localized the dly gene to the cytological division 70F on the third chromosome. Finally, Northern blot analysis revealed that dly encodes a single major 3.8 kb transcript (Baeg, 2001).

To discover the function of Dly, its expression in embryos was determined by in situ hybridization. dly transcripts are uniformly expressed at early embryonic stages, but by stage 8 they are enriched in stripes. Double staining for dly mRNA and Wg protein shows that dly transcripts are preferentially expressed in three to four cells anterior to the wg-expressing cells. Interestingly, this expression pattern is similar to that of both dally and frizzled 2, the Wingless receptor (Baeg, 2001).

In an attempt to assess the function of Dly during embryogenesis, the RNA interference (RNAi) method was used to perturb Dly protein synthesis. Embryos were injected with a dly double-stranded RNA (dsRNA). These embryos, referred to as dly dsRNA embryos, show the absence of naked cuticle. This phenotype is reminiscent of loss of either wg or hh gene activities. The segment polarity phenotype is also found when the activity of dally, which is required for Wg signaling in the embryo, is disrupted by RNAi, though the effect is less severe in dally dsRNA embryos than in dly dsRNA embryos. However, when compared with embryos injected with either dly dsRNA or dally dsRNA alone, embryos injected with an equimolar mixture of dly and dally dsRNAs show more severe segment polarity phenotypes. These embryos are smaller, particularly in the tail region. They also show an entire transformation of naked cuticle into cuticle with denticles, which is observed in wg or hh null mutations. Because dally does not appear to play a role in Hh signaling, the interaction between dally and dly observed in the RNAi interference experiment suggests that Dly and Dally function synergistically in Wg signaling. Altogether, these results suggest that dly is a novel segment polarity gene that potentiates Wg signaling (Baeg, 2001).

To further examine the role of Dly in Wg signaling, the function of Dly during wing imaginal disc development was examined. dly transcripts are uniformly expressed in wing discs. In the third instar imaginal disc, wg is expressed at the DV compartment border and acts over short and long ranges to pattern the wing disc. Short range Wg signaling induces the expression of the proneural gene achaete (ac) in a stripe on each side of the DV boundary, while long range Wg signaling controls the expression of Distal-less (Dll) within the wing blade. It was reasoned that overexpression of Dly might activate Wg signaling because dly dsRNA-injected embryos resemble those that have lost Wg activity. Interestingly, overexpression of dly (using the C96-Gal4 driver), which is highly expressed at the DV boundary of the wing disc, results in severe wing margin defects and loss of sensory bristles. These phenotypes are reminiscent of the phenotypes seen when Wg activity is reduced in the wing. Consistent with the adult wing phenotype, Ac expression is dramatically decreased in wing discs overexpressing Dly. Furthermore, when dly is overexpressed using the engrailed-Gal4 (en-Gal4) driver, the expression of Dll is reduced in the posterior compartment (Baeg, 2001).

A test was performed to see whether overexpression of transducers of the Wg signal can rescue the loss-of-function wg-like phenotypes associated with dly overexpression. Ectopic expression of either Wg or a gain-of-function Armadillo (Arm) can rescue the wing margin defects, and induced ectopic bristles that are characteristic of ectopic expression of the Wg pathway. Altogether, these results suggest that overexpression of Dly blocks Wg signaling in the wing disc, and that of Dly acts upstream of Arm (Baeg, 2001).

Since Dly is an extracellular GPI-linked molecule, it was reasoned that patterning defects associated with Dly overexpression might reflect the ability of Dly to sequester Wg, and thus prevent it from accessing and activating Frizzled 2. To visualize the effect of overexpressed Dly on Wg distribution, Dly was overexpressed using various Gal4 lines and the wing discs were stained with anti-Wg monoclonal antibodies. Two different staining protocols were used to detect Wg distribution. The first one, involved fixing the tissue before staining, and thus it detects mostly cytoplasmic Wg present in either secretory or internalized vesicles. In the second protocol, the tissue is incubated with the antibody prior to fixation and mostly detects extracellular Wg (Baeg, 2001).

Using the first protocol, in wild-type discs, Wg protein is found at high levels in a narrow stripe of three to five cells straddling the DV boundary. Following overexpression of Dly using either en-Gal4 or ptc-Gal4 a striking increased accumulation of Wg protein is observed. Using the second protocol, extracellular Wg is organized in a gradient at the basolateral surface of wg-expressing and nearby cells. Following overexpression of Dly, an increased accumulation of extracellular Wg protein is detected. This result indicates that Dly can affect extracellular Wg distribution (Baeg, 2001).

HSPGs are required to increase the local concentration of Wg ligand for its receptors. Considering the roles of Dally and Dly, there are at least two HSPGs at the cell surface of wing disc cells involved in Wg signaling. To generate mutant cells that lack all GAGs and determine the role of the HSPGs in Wg signaling, mutant clones of cells were generated that do not properly synthesize GAGs. sugarless (sgl) mutations cannot be used for this analysis because sgl acts in a cell non-autonomous manner, presumably because the enzyme synthesizes glucuronic acid that diffuses between cells. However, sulfateless (sfl) is involved in GAG modification and in its absence, the GAG chains are not synthesized properly. To determine whether the GAG chains of Dly are modified by sfl, Dly proteins expressed in wild-type or sfl mutant pupae were examined by Western blot analysis using anti-Dly antibodies. The predicted size for Dly is 80 kDa, and in wild-type pupae, Dly protein appears as a broad band that migrates to around 80-110 kDa, which presumably results from addition of GAG chains onto the core protein. In sfl mutant pupae, the modified form of Dly protein is significantly reduced and the sharp band of the core protein is increased. This results indicate that sfl plays a role in Dly modification. Using the staining protocol that detects cytoplasmic Wg, an alteration in the expression of intracellular Wg could not be detected in sfl mutant clones, indicating that sfl mutant cells normally transcribe wg and do not accumulate Wg (Baeg, 2001).

However, using the extracellular staining method, a dramatic decrease of extracellular Wg was detected in sfl mutant cells. Extracellular Wg has been shown to be mainly associated with the basolateral surface of cells, and GPI-anchored protein is thought to be primarily attached to the basal part of the cells. Together, these results suggest that the HSPGs are involved in extracellular Wg accumulation (Baeg, 2001).

Importantly, high accumulation of extracellular Wg can be detected on sfl mutant cells located adjacent to wild-type cells, suggesting that HSPGs act locally in a cell non-autonomous manner. Consistent with this observation, adult wing patterning in sfl mutant clones show local cell non-autonomy as well. Clones of sfl mutant cells are associated with wing margin defects, suggesting that sfl is required for Wg signaling, however some of the sfl mutant cells located near wild-type margin cells have a wild-type morphology (Baeg, 2001).

Taken together, these results suggest that HSPGs are involved in restricting Wg diffusion. Further, the local cell non-autonomy observed in sfl mutant clones may indicate that HSPGs may not be absolutely required for the binding of Wg to its transducing receptor(s) (Baeg, 2001).

HSPG modification by the secreted enzyme Notum shapes the Wingless morphogen gradient

The secreted signaling protein Wingless acts as a morphogen to pattern the imaginal discs of Drosophila. A secreted repressor of Wingless activity, called Notum, is described. Loss of Notum function leads to increased Wingless activity by altering the shape of the Wingless protein gradient. When overexpressed, Notum blocks Wingless activity. Notum encodes a member of the alpha/ß-hydrolase superfamily, with similarity to pectin acetylesterases. Evidence is presented that Notum influences Wingless protein distribution by modifying the heparan sulfate proteoglycans Dally-like and Dally. High levels of Wingless signaling induce Notum expression. Thus, Wingless contributes to shaping its own gradient by regulating expression of a protein that modifies its interaction with cell surface proteoglycans (Giráldez, 2002).

The Notum gene was identified in a gain-of-function genetic screen that caused loss of the wing and duplication of the dorsal thorax when expressed under sd^Gal4. Replacement of the wing by a duplicated dorsal thorax resembles the defect caused by the wg¹ mutant, and can also be produced by Gal4-driven overexpression of the GSK3 homolog, shaggy/ZW3, an intracellular repressor of Wg signaling. Expression of Notum throughout the embryo under tubulin^Gal4 control causes expansion of the denticle-producing zone at the expense of naked cuticle. Similar phenotypes can also be obtained by overactivation of the Egf receptor. The balance between Wg and Egfr activity determines the domain in which naked cuticle forms by regulating expression of shavenbaby in the embryonic ectoderm. Likewise, subdivision of the wing disc into wing and thorax territories occurs during the second larval instar and depends on the balance between Wg and Egfr signaling. Ventral expression of either Notum or an activated Egfr under wg^Gal4 control produces similar leg axis duplications. Thus, Notum might act as a repressor of the Wg pathway or as an activator of Egfr signaling when overexpressed. To distinguish between these possibilities, the effects of Notum expression were examined on wing vein formation, where the effects of Wg and Egfr signals differ. During third instar, ectopic Egfr signaling leads to formation of extra vein tissue in the wing blade. In contrast, Notum expression produces scalloping of the wing but has no effect on vein formation. This phenotype resembles late loss of Wg activity. Thus, it is concluded that overexpression of Notum does not work by activation of the Egfr pathway, but rather interferes with Wg activity (Giráldez, 2002).

A number of genes that modulate Wg activity are spatially regulated by Wg signaling in the embryo, including Dfz2, Dally, and naked. It was therefore asked whether Wg regulates Notum expression. In situ hybridization showed that Notum is expressed in a segmentally repeated pattern in two rows of cells anteriorly adjacent to the Engrailed-expressing cells. This corresponds to the Wg stripe in the embryo. Notum expression also mirrors Wg expression in the wing disc. Ectopic activation of the Wg pathway by expression of a constitutively active form of Armadillo induces ectopic Notum expression. Thus, high levels of Wg activity induce Notum expression, which in turn serves as a Wg antagonist (Giráldez, 2002).

To isolate loss-of-function mutations in the Notum gene, a chemical mutagenesis screen was performed for reversion of the thorax duplication caused by expression of EP-Notum. EP-Notum males were treated with EMS and crossed to sd^Gal4 females. Among 15,000 progeny, three flies were recovered that had normal wings despite overexpressing the endogenous Notum gene. These three alleles formed a single complementation group. l(3)72Da and l(3)72CDf were subsequently identified as Notum alleles and have been renamed Notum⁴ and Notum⁵. Sequence analysis revealed an alteration of the splice donor site of exon 1 in Notum¹ and alterations in the coding sequence for the other alleles analyzed (Giráldez, 2002).

Notum zygotic mutant embryos produced a variable naked cuticle phenotype, typical of excess Wg activity. The embryonic phenotype is weak, perhaps due to maternal contribution, wing discs from Notum mutant larvae were used to examine the effects of removing Notum activity. The primordia of the adult wing and thoracic body wall can be visualized in the wing disc by expression of Nubbin and Teashirt proteins . Nubbin is a POU homeodomain protein expressed in the presumptive wing blade and wing hinge. Tsh is a zinc finger protein expressed in the presumptive thorax. Ectopic Wg activity can lead to duplication of wing structures at the expense of thorax. Notum mutant discs showed this phenotype. The severity of these defects ranged from duplication of the wing pouch and hinge associated with a reduced thorax to almost complete loss of thorax associated with a severely abnormal wing duplication. The same range of phenotypes was obtained by activation of the Wg pathway in the early wing disc using Arm^S10 (Giráldez, 2002).

Wg forms a long-range protein gradient and regulates several target genes in different spatial domains in the wing disc. Achaete-Scute is a high-threshold Wg target, expressed in cells close to the DV boundary. Achaete-Scute expression specifies the proneural region in the anterior wing margin in which the single row of sense organ precursor (SOP) cells will form. In wild-type discs, a single row of SOPs forms on each side of the Wg stripe in the anterior compartment. Large clones of cells mutant for Notum³ form additional rows of SOPs and produce extra mechanosensory bristles. Thus, the region in which cells show a high-threshold response to Wg is broadened in the Notum mutant tissue. This defect occurs only when large clones affect both sides of the DV boundary. The defect is rescued when cells on one side of the boundary are able to produce Notum. This suggests that Notum can act nonautonomously. In addition, it was noted that some of the ectopic bristles derived from wild-type cells. This indicates that loss of Notum activity in a mutant clone can lead to increased Wg signaling in nearby wild-type cells. These observations suggest that Notum acts nonautonomously to affect the range of Wg action (Giráldez, 2002).

To examine the effect of large clones of Notum mutant cells on Wg, clones were produced using engrailed^Gal4 to drive Flp recombinase in posterior cells. The mutant posterior compartment was overgrown and the size of the Distal-less (Dll) expression domain increased, suggesting an increased range of Wg action. Antibody labeling showed that the level of Wg protein is elevated in the posterior compartment. The number and brightness of Wg protein vesicles was used to visualize the Wg gradient. Both were increased in the Notum mutant tissue. Thus, reducing Notum activity increases the level of Wg protein and broadens its distribution. The effects on Wg distribution do not make an abrupt transition at the AP compartment boundary. Instead, it increases with distance from the boundary, suggesting that the mutant phenotype may be partially rescued by Notum protein produced by the wild-type anterior cells (Giráldez, 2002).

The results of the clonal analyses suggest that Notum acts nonautonomously. The predicted Notum protein contains a hydrophobic sequence near its amino terminus that might function as a signal peptide. To determine whether Notum is secreted, S2 cells were transfected with expression constructs for wild-type Notum, the Notum² mutant, and a Golgi-tethered form of Notum, in which the signal peptide was replaced with the transmembrane domain from a Golgi-resident enzyme (Notum-GT). Secretion was assayed by immunoprecipitation from cell lysates and from the medium in which the cells were grown. Wild-type Notum protein was recovered from the conditioned medium. Notum-GT was not recovered from conditioned medium. The Notum² mutant protein was recovered at low levels from the medium, suggesting impaired secretion (Giráldez, 2002).

The activities of wild-type and Golgi-tethered Notum were compared in vivo to verify that Notum can act nonautonomously. In wild-type discs, Dll and Hnt are expressed symmetrically in D and V compartments in response to Wg. When overexpressed in the D compartment under apterous^GAL4 control, Notum protein is seen at elevated levels throughout the disc, indicating that it is secreted in vivo. Hnt expression is lost and Dll expression is reduced, symmetrically, in both D and V compartments. This leads to scalloping of the wing. In contrast, expression of Notum-GT affects only the D compartment. The reduced size of the Dll domain can be attributed to a reduced level and range of Wg in the D compartment. This leads to a small D compartment and loss of dorsal wing margin, without any effect on the ventral margin. Thus, overexpression of Notum-GT in the D compartment produces an asymmetric Wg gradient, whereas overexpression of the wild-type protein in the D compartment reduces the Wg gradient symmetrically in both compartments. These observations indicate that wild-type Notum can act as a secreted protein to reduce the effective range of the Wg gradient. However, Notum can also act when retained in the Golgi (Giráldez, 2002).

Pectin acetylesterases from plant pathogens act as secreted enzymes to deacetylate pectins in plant cell walls. Pectins are composed mainly of galacturonic acid residues, some of which are methylated or acetylated. Pectin acetylesterases hydrolyze the ester bond linking acetyl groups to galacturonic acid. Glycosaminoglycans consist of repeated glucuronic acid and GlcNAc disaccharide units. Although GAGs are different in structure from pectins, the similarity to pectin acetylesterases raised the possibility that Notum might act on the GAG side chains of HSPGs. As a first step toward addressing this possibility, it was asked whether Notum could modify Dally and Dlp when coexpressed in S2 cells. Coexpression of Dally-HA with Notum reduced the amount of Dally-HA recovered in S2 cell lysates. In sulfateless mutants, which lack N-deacetylase/N-sulfotransferase (NDST) activity, the level of Dally and Dlp proteins are also strongly reduced, perhaps indicating altered stability of the immature protein. Coexpression of HA-tagged Dlp with Notum altered the electrophoretic mobility of Dlp-HA, without causing substantial loss of the protein. Dlp-HA migrates as a broad band at ~115 kDa, with minor bands at 97 and 105 kDa. Coexpression with wild-type Notum increases the amount of the 97 kDa form (Dlp*), apparently at the expense of the 115 kDa form. The small amount of this band present in S2 cells expressing Dlp-HA may reflect activity of endogenous Notum protein, which was detected on longer exposures (Giráldez, 2002).

The GAG side chains of HSPGs consist mainly of repeated dimers of N-acetylglucosamine (GlcNAc) and glucuronic acid. The first step in modification of the side chains involves replacement of the acetyl moiety on GlcNAc with a sulfate moiety by N-deacetylase/N-sulfotransferase. NDST modifies ~50% of the GlcNAc residues, in blocks along the GAG side chain. The structural similarity between pectins and GAGs raised the possibility that Notum might act by removing acetyl groups from GlcNAc residues of GAGs. If this is the case, GlcNAc residues modified by NDST should not be a good substrate for Notum in cells. To address this possibility, S2 cells were transfected to express a constant amount of Notum and increasing amounts of NDST and the ability of Notum to modify Dlp was assayed. Increasing the ratio of NDST to Notum reduces the amount of Dlp*. These observations support the proposal that NDST and Notum could act on the same substrate. NDST is a Golgi-resident enzyme. In cells where the two proteins are coexpressed, NDST and Notum might compete for modification of GAG side chains, as illustrated by the activity of Notum-GT when expressed in the wing disc. Since there are few cells in the disc where Notum is expressed, it is suggested that secreted Notum may act on HSPGs at the cell surface to deacetylate the blocks of GlcNAc residues that were not modified by NDST during GAG biosynthesis (Giráldez, 2002).

The glypicans Dally and Dlp have been shown to bind and stabilize extracellular Wg, although Dlp is considerably more effective. To ask whether Notum modifies the ability of Dlp to bind Wg, Wg protein was examined in discs expressing Dlp-HA and Notum. Expression of Dlp-HA in a broad band of cells in the center of the wing disc under spalt^Gal4 control causes accumulation of Wg protein, mainly outlining the cell surface. Under these conditions, scalloping of the wing margin was observed, suggesting that Wg is partially sequestered by binding to Dlp and is less available for binding to its receptor. The distribution of Wg differs in discs expressing Dlp-HA and Notum. The total level of Wg accumulation is considerably lower, although Dlp is expressed at a comparable level. In addition, much of the Wg protein appears in intracellular vesicles, instead of outlining the cell surface. These findings support the proposal that Notum modifies Dlp to render it less able to bind and stabilize Wg. Thus, Notum contributes to shaping the Wg gradient by altering the ability of the cell surface glypican Dlp to stabilize extracellular Wg (Giráldez, 2002).

Wg, Dpp, and Hh each regulate the expression of their receptors in ways that can influence the shape of the gradient. These findings indicate that Wg can also influence formation of its own gradient by modulating the activity of cell surface HSPGs. Wg induces expression of Notum, which can act either during GAG biosynthesis or as a secreted protein to modify cell surface HSPGs. Reduced Notum activity allows excess accumulation of Wg protein, resulting in an increased range of Wg activity . Conversely, overexpression of Notum limits the ability of cells to bind and stabilize Wg, thereby limiting the ability of Wg to spread in the disc epithelium and form a long-range gradient. In the embryo, Notum overexpression produces segment polarity defects comparable to those caused by reduced Wg activity. In the wing disc, elevating Notum levels limit Wg movement and cause phenotypes ranging from scalloping of the wing to early failure of wing pouch specification. These changes in the shape of the Wg gradient can be attributed to the effect of Notum on Dally and Dlp. Coexpression of Notum with Dly limit the ability of overexpressed Dlp to accumulate Wg protein. It is suggested that the role of Notum is to limit the ability of Dlp and Dally to bind Wg (Giráldez, 2002).

Wingful/Notum, an extracellular feedback inhibitor of Wingless

A novel Drosophila Wingless (Wg) target gene, wingful (wf: termed Notum by FlyBase), encodes a potent extracellular feedback inhibitor of Wg. In contrast to the cytoplasmic protein Naked cuticle (Nkd), the only known Wg feedback antagonist, Wf functions during larval stages, when Nkd function is dispensable. It is proposed that Wf may provide feedback control for the long-range morphogen activities of Wg (Gerlitz, 2002).

A library of 2000 Gal4 enhancer trap P-element insertions was established, each of which reports a gene expression pattern in the wing imaginal disc. This collection was screened with a UAS-GFP reporter for lines that show a wg-like expression pattern. There were 11 insertions identified that reported wg-like gene expression in the embryonic epidermis and all imaginal discs. Four of these lines (S180, ND382, S476, S554) contained an insertion in the wg gene itself; the other seven lines (S141, S145, S163, S330, ND337, ND339, ND634) all carried a P-element insertion at cytological position 72D, only a few base pairs upstream of gene CG13076, referred to as wingful. These enhancer trap insertions indeed report the expression of wf, as revealed by RNA in situ hybridization. wf is ectopically expressed upon wg misexpression, indicating that wf is a Wg target throughout larval development (Gerlitz, 2002).

wf codes for a presumptive protein of 671 amino acids, with an N-terminally situated signal sequence. The wf product is readily secreted from transfected Drosophila cells and has a noticeable propensity to adhere to the surfaces of intact cells. The analysis of the Wf protein sequence reveals a significant structural homology to a subfamily of poorly characterized hydrolases related to plant pectin acetylesterase. Together, these results suggest that the product of the wf gene may catalyze the hydrolytic cleavage of an extracellular substrate (Gerlitz, 2002).

To test the hypothesis that the pan-Wg-target wf encodes an inhibitor of Wg activity, attempts were made to abolish wf function by genetic means. From a collection of six EMS-induced lethal mutations, located between the distal breakpoint of Df(3L)st-f13 and the proximal breakpoint of Df(3L)brm11, a putative null allele of wf was identified with a stop codon at amino acid position 141, encoding a severely truncated protein. Animals homozygous for wf¹⁴¹ or animals of the genotype wf¹⁴¹/Df3(3L)st-f13 die during pupal stages, and show various phenotypes. The most prominent of these are patterning defects in the wing imaginal disc. Wg signaling plays at least two distinct roles during wing development. Early reduction of wg activity results in a wing-to-notum transformation, indicating a requirement for Wg in defining the wing blade primordium, but later reductions cause the loss of wing margin and adjacent tissue, indicating its subsequent role in specifying the wing margin and organizing wing blade development. wf mutants show phenotypes opposite to both classes. Wing discs mutant for wf are enlarged with an extended wing blade region (hence the name wingful). Often these discs contain two wing pouches at the expense of notal structures. Although no apparent expansion of Distalless-lacZ (Dll-lacZ) expression was detected along the dorsoventral axis, there was a significant increase in the number of cells expressing neuralized, a high-threshold Wg target expressed in neural wing margin cells. Consistent with this observation, rare adult escapers mutant for wf show a dramatically increased number of mechanosensory bristles in the wing. Wg signaling also distinguishes between sternite and ventral pleura development in the adult abdomen. wf adult escapers show extra sternite bristles, an effect that was also observed with ectopic expression of wg. Finally, wf adults show extra dorsocentral bristles, sensory organs on the notum whose specification has been shown to depend on wg activity. Taken together, these results show that the absence of wf function causes a gain of Wg activity in developing adult tissues. Therefore, the function of the wild-type wf product is to limit Wg signaling activity (Gerlitz, 2002).

A further prediction of the assumption that Wf functions as a Wg feedback inhibitor is that wf overexpression should lead to wg loss-of-function phenotypes. Three lines of evidence are presented to show that this is, indeed, the case. (1) One of the wf enhancer trap P-element insertions was replaced with an EP element positioning 10 UAS sites upstream of the wf gene, rendering it transcriptionally responsive to Gal4 expression. Alterations of wf expression have unusually potent effects, since all commonly used Gal4 drivers caused lethality in combination with UAS-wf. The only exceptions were S168-Gal4 and scalloped-Gal4, which are expressed in the wing pouch and represent Wg targets, providing a self-regulating circuit in combination with UAS-wf. Adult animals carrying the S168-Gal4 and UAS-wf transgenes have severely reduced wings that lack all wing margin structures. (2) The expression of two target genes were analyzed in this context. S168-Gal4 expression was virtually abolished by UAS-wf expression, whereas the expression domain of wg itself is expanded. Wg is known to narrow its own domain of expression, because a reduction in Wg signal transduction causes ectopic wg transcription. (3) Finally, and perhaps most strikingly, driving expression of wf with scalloped-Gal4 results in a wing-to-notum transformation, the founding loss-of-function phenotype of the wg gene (Gerlitz, 2002).

Based on its structural features as a secreted protein with homologies to pectin acetylesterases, Wf could exert its function by modifying polysaccharide-based properties of cell surface proteins and thereby impeding the intercellular movement of the Wg protein. Alternatively, Wf could counteract Wg signaling by modifying the transducing properties of Wg or one of its receptors. To distinguish between these two possibilities, tests were performed to see whether Wf also antagonizes a derivative of Wg, Nrt-Wg, that is tethered to the cell surface and does not move through tissue. Expression of Wg or Nrt-Wg driven by dpp-Gal4 results in a robust activation of ectopic Dll-lacZ expression. Surprisingly, wf expression extinguishes Dll-lacZ expression induced by tethered Wg, as well as that induced by free Wg. From this experiment it can be ruled out that the primary function of Wf is to impede the extracellular transport of Wg. Therefore, Wf must interfere with the signaling activities of either Wg or its receptor components.Because no physical interaction between Wf and Wg, or between Wf and Frizzled-2 (Dfz2), or between Wf and Frizzled-2's LRP-like partner Arrow could be detected in tissue culture systems, the view is favored that Wf inhibits the activity of a coreceptor component, such as Dally or Dally-like (Dly), proteoglycans that appear to participate in Wg reception (Gerlitz, 2002). Wf may inhibit such receptor components via its presumptive esterase activity, for example, by modifying Dally or Dly glycosaminoglycan chains. A definitive proof for this mode of action could be achieved by the genetic demonstration that in larval dally;dly double-mutant situations, loss of wf function has no antagonistic effect (Gerlitz, 2002).

The discovery of Wingful as an essential Wg feedback antagonist may provide an explanation of why Nkd has no apparent role in imaginal tissues of Drosophila. The function of Nkd may be superseded by that of Wf, which functions in a powerful negative-feedback loop in adult development. Conversely, when both maternal and zygotic components of wf are removed, no obvious requirement was observed for Wf in embryonic development, possibly because the nkd system is operative at this stage of development. Both Naked and Wf can, however, inhibit Wg signaling throughout development if they are overexpressed, but each of them is operating more effectively at only one of the two stages. It may not be coincidence that Nkd, as the intracellular feedback antagonist, is used during embryonic patterning, (where Wg functions at short range), whereas Wf, as a secreted extracellular antagonist, primarily regulates patterning processes that depend on long-range Wg signaling. Wf functions nonautonomously and, like Argos (a secreted feedback antagonist of the Drosophila EGF system), may have a different range of action compared with the primary signal, providing an intricate means to shape the range and slope of the cellular responses to a morphogen gradient (Gerlitz, 2002).

slalom encodes an adenosine 3'-phosphate 5'-phosphosulfate transporter essential for development in Drosophila

Sulfation of all macromolecules entering the secretory pathway in higher organisms occurs in the Golgi and requires the high-energy sulfate donor adenosine 3'-phosphate 5'-phosphosulfate. A gene has been identified that encodes a transmembrane protein required to transport adenosine 3'-phosphate 5'-phosphosulfate from the cytosol into the Golgi lumen. Mutations in this gene, which has been called slalom, display defects in Wg and Hh signaling; these defects are likely due to the lack of sulfation of glycosaminoglycans by the sulfotransferase sulfateless. Analysis of mosaic mutant ovaries shows that sll function is also essential for dorsal-ventral axis determination, suggesting that sll transports the sulfate donor required for sulfotransferase activity of the dorsal-ventral determinant Pipe (Lüders, 2003).

Secreted signaling molecules of the FGF, Hh, TGF-ß and WNT families rely on proteoglycans (PGs) for efficient activation of their respective signaling pathways. PGs consist of secreted or transmembrane core proteins to which glycosaminoglycan (GAG) side chains are attached at specific consensus sites. In Drosophila, the secreted PG perlecan, the transmembrane PG syndecan and two members of the glypican family of glycosylphosphatidylinositol (GPI) anchored PGs have been identified. Phenotypes associated with loss of function mutations of the glypican-encoding genes dally and dally-like (dlp) have revealed the requirement of these PGs for efficient activation of several signal transduction pathways (Lüders, 2003).

The function of PGs is critically dependent on the integrity of the attached GAGs. GAGs are unbranched polysaccharide chains, which are synthesized on proteoglycan core proteins in the Golgi and undergo complex modification reactions before the PG to which they are attached is transported to the cell surface. In the case of glypican, heparan sulfate (HS) chains, consisting of a sugar backbone of alternating units of N-acetylglucosamine (GlcNAc) and glucuronic acid (GlcA), are synthesized on the core protein. The nucleotide sugar substrates for this reaction are synthesized in the cytoplasm and must be transported into the Golgi. In Drosophila, the activated precursor UDP-GlcA is synthesized from UDP-glucose by the enzymatic activity of the sugarless (sgl) gene product, a homolog of mammalian UDP-glucose dehydrogenases. The gene product of fringe connection (frc), a predicted ER/Golgi multipass transmembrane protein, has been shown to transport UDP-GlcA and UDP-GlcNAc from the cytosol into the Golgi. Mutations in either gene severely affect the Wg and FGF signaling pathways. Elongation of the HS chains requires the activity of HS polymerases. tout-velu (ttv) encodes a protein with homology to the mammalian HS co-polymerase EXT1 and has been demonstrated to be required specifically for Hh signaling. Subsequent to their synthesis, GAGs undergo multiple modifications such as epimerization and sulfation. Mutations in sulfateless (sfl), a homolog of vertebrate N-deacetylase/N-sulfotransferases, lead to a severe reduction in the activity of the Wg, Hh and FGF signaling pathways. A characteristic feature of all mutations in genes involved in GAG biosynthesis is that their segment polarity phenotypes can be rescued by ectopic expression of wg or hh, suggesting that GAGs are not essential components of the respective signaling cascades but accessory factors most likely required for the proper distribution of extracellular signaling molecules throughout morphogenetically active tissues in vivo (Lüders, 2003).

GAGs have also been proposed to play a role in the determination of the dorsal-ventral (D/V) axis of the Drosophila embryo. The D/V polarity of the embryo is established during oogenesis by asymmetric expression of the key D/V determinant pipe (pip) in the follicle cell epithelium. pip expression in the ventral follicle cell layer is necessary and sufficient to trigger a serine-protease cascade in the perivitelline space; this leads to the generation of an active ligand for the transmembrane receptor Toll (Tl). Activation of Tl on the ventral side of the embryo results in a gradient of nuclear localization of the transcription factor Dorsal, which patterns the D/V axis. Based on sequence similarity to a family of vertebrate enzymes and its localization in the Golgi apparatus, pip has been hypothesized to encode a heparan sulfate 2-O-sulfotransferase. However, neither the enzymatic activity nor the substrate specificity of pip have been demonstrated directly (Lüders, 2003).

Sulfation of secreted molecules occurs in the Golgi and requires the high-energy sulfate donor adenosine 3'-phosphate 5'-phosphosulfate (PAPS) to be present within that organelle. In Drosophila, PAPS is synthesized in the cytoplasm by PAPS-synthetase, which incorporates both ATP-sulfurylase and adenosine 5'-phosphosulfate-kinase (APS-kinase) activity. PAPS must be transported into the Golgi thereafter to serve as a substrate for sulfotransferases. This study reports the molecular identification and functional characterization of a PAPS transporter. Mutations in this gene, called slalom (sll), are associated with defects in multiple signaling pathways, including Wg and Hh signaling. A phenotypic analysis suggests that the effects of sll on signal transduction are caused by its requirement for GAG modification. Evidence is presented that sll is also required to supply PAPS to the machinery initiating the establishment of embryonic D/V polarity, supporting the view that Pipe protein is a sulfotransferase (Lüders, 2003).

A hydrophobicity analysis of the putative Sll protein predicts a hydrophobic polypeptide with at least 10 transmembrane regions, a structural characteristic of many nucleotide-sugar transporters. A BLAST search of non-redundant protein databases with the Sll sequence revealed that Sll is conserved throughout the animal kingdom, as well as in plants, and shares nearly 40% of amino acid sequence identity with predicted mouse and human proteins of unknown function. Sll is also similar to mammalian proteins that have been classified as nucleotide-sugar transporters on the basis of their homology to the human UDP-galactose transporter hUGT. However, Sll itself has no significant sequence similarity to hUGT. While these data suggest that Sll encodes a transmembrane transporter, they leave open what substrate Sll may be transporting (Lüders, 2003).

In order to demonstrate directly that Sll has transporter activity, and to determine its substrate specificity, an in vitro vesicle transport assay was used to examine the ability of Sll to facilitate the transport of defined nucleotide substrates into Saccharomyces cerevisae microsomes. Heterologous expression of sll in yeast was favored over other systems since procedures in this organism have been successfully used for transfection of nucleotide-sugar genes, high-level preparation of sealed microsomal vesicles and nucleotide-sugar assays. In vesicles derived from cells transfected with a sll-expression construct, a 5-fold increase in the uptake of PAPS was detected. Import of several other nucleotide-sugars did not increase after sll expression. Importantly, the sll-dependent PAPS transport activity could be efficiently competed by structurally similar substrate analogs but not by more distantly related nucleotide monophosphates, confirming that the transport was specific (Lüders, 2003).

PAPS is synthesized in the cytoplasm and serves as a substrate for sulfotransferases during the post-translational modification of macromolecules in the Golgi. Therefore, the result that sll encodes a PAPS transporter suggests that Sll should be localized to the Golgi membranes. To address this question, peptide antibodies were raised against Sll. Localization of Sll in salivary glands of wild-type embryos, using Sll antiserum, showed a punctate staining pattern in the perinuclear region of cells. Co-staining of salivary glands with antibodies against an endogenous Golgi protein revealed extensive co-localization of the two proteins, indicating that Sll is a resident Golgi protein. Together, these data suggest that Sll encodes a nucleotide transporter required to translocate the high-energy sulfate donor PAPS into the lumen of the Golgi (Lüders, 2003).

The phenotypical analysis of sll mutants shows that sll function is required for Wg and Hh signaling, and suggests that sll may also play a role in the activation of other signaling pathways, such as the TGF-ß pathway and possibly others. The broad requirement for sll raises the question of how sulfation influences the function of multiple secreted signaling factors. Proteins entering the secretory pathway are ubiquitously sulfated on specific tyrosine residues. However, according to sulfation consensus prediction algorithms, neither Wg nor Hh is predicted to be a substrate for tyrosine sulfation (see 'Sulfinator' at http://www.expasy.org/tools/sulfinator/). In addition, overexpression experiments show that wg and hh have the ability to at least partially activate their respective signaling cascades in sll mutants, which would be unlikely if the signaling factors themselves were non-functional due to lack of sulfate modification. Therefore, it is unlikely that lack of tyrosine sulfation of either Wg or Hh proteins is responsible for the defects in Wg or Hh signaling pathways in sll mutants (Lüders, 2003).

The Wg and Hh signaling pathways are activated through interaction of secreted ligands with their respective cell surface receptors. This activation relies on the action of GAGs attached to proteoglycan core proteins of the glypican family. In the case of Wg, the proteoglycans are thought to act as coreceptors, retaining the ligand at the cell surface in the vicinity of its receptor. Hh has been proposed to depend on GAGs for efficient transport to its target cells. In the absence of functional GAGs, the signaling molecules are not present in sufficiently high concentrations at the receptor to activate the signal transduction pathway. This defect can be compensated by overexpression of the signaling molecule. Therefore, a characteristic feature of mutations in genes involved in GAG metabolism is that their segment polarity defects can be rescued by overexpression of the ligand. These data show that Wg protein levels and Hh signaling activity are reduced in sll mutant clones. The segment polarity defects of sll mutants can be rescued by overexpression of wg or hh. In addition sll and dally mutants affect Wg signaling synergistically. The data also suggest that the ability of cells to sulfate GAGs is dramatically reduced in sll mutants because the sulfate donor necessary for sfl to modify the polysaccharide chains cannot be transported into the Golgi in sufficient amounts. Taken together, these observations strongly suggest that lack of functional GAGs is responsible for the segment polarity defects in sll mutants (Lüders, 2003).

Interestingly, the overall size of sugar chains attached to the glypican Dally does not appear to be affected in sll mutants. Consistent with this result, it has been observed that the overall level of HS, which is reduced to trace amounts in the sgl mutant that affects GAG biosynthesis, is not markedly changed in sfl mutants, which should affect sulfate modification but not synthesis of GAG chains. However, it cannot be excluded that residual sulfation of GAGs occurs in the cell culture assay owing to incomplete inhibition of the PAPS transport activity of sll by dsRNA interference. However, the phenotypic analysis of sll suggests that sll function is essential for at least two sulfotransferases, sfl and pip, and that the GAG chains present in sll mutants are not able to fulfill their normal function owing to altered sulfation patterns (Lüders, 2003).

The central position of sll in sulfate metabolism along the secretory pathway makes it an interesting tool for the identification of developmental pathways sensitive to sulfate modification. The results demonstrate that several cell-cell communication pathways are critically dependent on the sulfation of macromolecules, and highlight the importance of sulfation during pattern formation and development (Lüders, 2003).

Three Drosophila EXT genes shape morphogen gradients through synthesis of heparan sulfate proteoglycans

The signaling molecules Hedgehog (Hh), Decapentaplegic (Dpp) and Wingless (Wg) function as morphogens and organize wing patterning in Drosophila. In the screen for mutations that alter the morphogen activity, novel mutants of two Drosophila genes, sister of tout-velu (sotv) and brother of tout-velu (botv), and new alleles of toutvelu (ttv), were identified. The encoded proteins of these genes belong to an EXT family of proteins that have or are closely related to glycosyltransferase activities required for biosynthesis of heparan sulfate proteoglycans (HSPGs). Mutation in any of these genes impaired biosynthesis of HSPGs in vivo, indicating that, despite their structural similarity, they are not redundant in the HSPG biosynthesis. Protein levels and signaling activities of Hh, Dpp and Wg were reduced in the cells mutant for any of these EXT genes to a various degree, Wg signaling being the least sensitive. Moreover, all three morphogens were accumulated in the front of EXT mutant cells, suggesting that these morphogens require HSPGs to move efficiently. In contrast to previous reports that ttv is involved exclusively in Hh signaling, ttv mutations were also found to affect Dpp and Wg. These data lead to the conclusion that each of three EXT genes studied contribute to Hh, Dpp and Wg morphogen signaling. It is proposed that HSPGs facilitate the spreading of morphogens and therefore, function to generate morphogen concentration gradients (Takei, 2004).

In addition to monitoring signaling in EXT mutant cells, antibodies that recognize Hh, Dpp and Wg, and a GFP-tagged version of Dpp were used to analyze whether the levels or distribution of these morphogens had been affected. Levels of each of these proteins were significantly reduced in the mutant, both in the morphogen-expressing region and in the receiving region. For Hh, Dpp and Wg, similar results were observed in cells mutant singly for any of the EXT genes. Single mutation was not tested for the distribution of Dpp-GFP. In the morphogen-expressing region, hh expression was not downregulated, however levels of Hh protein were significantly decreased. This may indicate that Hh protein is destabilized and/or not retained efficiently on the cell surface in the absence of HSPGs. In contrast to hh, expression of the wg and dpp and levels of Wg and Dpp were decreased in the EXT clones. The decrease in dpp expression is easily accountable because Hh signaling is impaired in the absence of HSPGs. In contrast, the decrease in wg expression is not as readily explainable: cut and wg are both targets of Notch signaling, however the protein level of Cut was not altered in EXT clones. This suggests that wg is also regulated by an unknown mechanism dependent on HSPGs (Takei, 2004).

In the morphogen-receiving region, each of these proteins was significantly decreased in the clones of cells mutant for EXT genes, although a little leakage of morphogen molecules was seen even in the clones doubly mutant for ttv and botv. This suggests two possible mechanisms that do not exclude each other: in the absence of HSPGs these three morphogens are (1) destabilized and/or are not retained efficiently on the cell surface, like Hh in morphogen-expressing region, or (2) prevented from diffusing efficiently into the region consisting of EXT mutant cells. Intriguingly, close observation of the distribution of Hh strongly suggested a function for HSPGs in morphogen movement. In the wild-type discs, Hh protein synthesized in the posterior compartment appears to flow into the anterior compartment, with a moderate concentration gradient starting from the middle of the posterior compartment. However, Hh abnormally accumulates in the posterior compartment when the EXT mutant clone is in the anterior compartment along the A/P boundary. This effect is seen both in the ventral region and in the dorsal region. This suggests that Hh fails to move into the mutant cells and as a consequence accumulates in posterior cells instead. Dpp-GFP and Wg accumulation in front of the mutant clones was also apparent, however less pronounced compared with the case of Hh. Therefore it is concluded that the HSPG-dependent diffusion is the common mechanism for the movement of these three morphogens (Takei, 2004).

Abrogation of heparan sulfate synthesis in Drosophila disrupts the Wingless, Hedgehog and Decapentaplegic signaling pathways

Studies in Drosophila and vertebrate systems have demonstrated that heparan sulfate proteoglycans (HSPGs) play crucial roles in modulating growth factor signaling. Mutations have been isolated in sister of tout velu (sotv), a gene that encodes a co-polymerase that synthesizes HSPG glycosaminoglycan (GAG) chains. Phenotypic and biochemical analyses reveal that HS levels are dramatically reduced in the absence of Sotv or its partner co-polymerase Tout velu (Ttv), suggesting that both copolymerases are essential for GAG synthesis. Furthermore, mutations in sotv and ttv impair Hh, Wg and Decapentaplegic (Dpp) signaling. This contrasts with previous studies that suggested loss of ttv compromises only Hh signaling. These results may contribute to understanding the biological basis of hereditary multiple exostoses (HME), a disease associated with bone overgrowth that results from mutations in EXT1 and EXT2, the human orthologs of ttv and sotv (Bornemann, 2004).

A role for HSPGs in Wg signaling in Drosophila is well established. Addition of heparin sulfate (HS) or chondroitin sulfate (CS) GAGs to culture medium causes S2 cells to release secreted Wg protein, while heparinase treatment reduces their ability to respond to exogenous Wg. Furthermore, both Wg levels and signaling efficiency are sensitive to HSPG concentration; Wg activity is reduced when either HS synthesis or the core proteins for the glypicans Dally and Dlp are compromised. Severe disruption of extracellular Wg distribution and the altered expression of the downstream target Ac, in ttv and sotv mutant clones, demonstrates that HS chains are essential for establishing and/or maintaining the Wg gradient (Bornemann, 2004).

The effects of HSPGs on signaling are further modulated by secondary modifications to the GAG chains. For example in Drosophila, loss of either sfl or slalom (sll) compromises both Wg and Hh signaling. Sfl is an N-deacetyl/N-sulfotransferase, and Sll is required to transport a high-energy sulfate donor molecule to the golgi, where HSPGs such as dally and dlp are sulfate modified. Loss of GAG sulfation results in signaling defects, despite the fact that unsulfated HS-GAGs are still present in sfl mutants. Consequently, not only are HS-GAG chains required for signaling, but the extent of sulfation is also crucial. In vertebrates, Qsulf1 promotes Wg signaling by catalyzing removal of 6-O sulfate groups from HS chains of HSPGs, including the Dally homolog glypican 1. It has been proposed that in the absence of Qsulf1, Wg stays tightly bound to HSPGs, which prevents the ligand from interacting with receptors. However, in Qsulf1-expressing cells, selective 6-O desulfation reduces the binding affinity sufficiently to permit ligand-receptor interaction. In this context, the altered HS disaccharide distribution and abnormal sulfation patterns encountered in hypomorphic sotv alleles is intriguing. If partial loss of function EXT2 mutations in individuals with HME causes similar disruption of GAG modifications, they could also have complex and allele-specific effects on signaling (Bornemann, 2004).

Opposing activities of Dally-like Glypican at high and low levels of Wingless morphogen activity

The glypican family of heparan sulfate proteoglycans has been implicated in formation of morphogen gradients. The role of the glypican Dally-like protein (Dlp) in shaping the Wingless gradient was studied in the Drosophila wing disc. Surprisingly, Dlp has opposite effects at high and low levels of Wingless. Dlp promotes low-level Wingless activity but reduces high-level Wingless activity. Evidence is presented that the Wg antagonist Notum acts to induce cleavage of the Dlp glypican at the level of its GPI anchor, which leads to shedding of Dlp. Thus, spatially regulated modification of Dlp by Notum employs the ligand binding activity of Dlp to promote or inhibit signaling in a context-dependent manner. Notum-induced shedding of Dlp could convert Dlp from a membrane-tethered coreceptor to a secreted antagonist (Kreuger, 2004).

Notum encodes a secreted member of the alpha-hydrolase superfamily, which includes serine proteases, lipases, and other enzymes in which a Ser-Asp-His catalytic triad comprises the active site. Mutation of Ser237 to Ala was shown to remove Notum activity in vivo and in vitro (Giraldez, 2002). On the basis of its similarity to pectin acetylesterase enzymes from plants and plant pathogens and on the basis of its ability to compete with the N-deacetylase-N-sulfotransferase enzyme for use of Dlp as a substrate, it has been proposed that Notum might function as a GAG deacetylase. If this were the case, it would be expected that Notum would be unable to modify a form of Dlp lacking GAG side chains (Kreuger, 2004).

Dally and Dlp each contain five putative GAG addition sites and have been shown to carry predominantly heparan sulfate (HS, as shown for vertebrate glypicans). GAG biosynthesis is initiated by addition of a xylose residue to the hydroxyl group of a serine in a serine-glycine (SG) motif in the stem region of the glypican (Lindahl, 1998). Mutant forms of Dally and Dlp proteins were prepared in which all the putative GAG-addition sites were mutated. For Dally, the serine residues in the five SG motifs were mutated to alanine. For Dlp, four SG motifs were deleted by removing a block of 25 residues, and the fifth was mutated to AG (Kreuger, 2004).

The glycosylation states of epitope-tagged Dally and Dlp and the GAG addition site mutants were compared by anion exchange chromatography. Glypicans isolated from mammalian cells in culture are highly negatively charged and are typically retained on a strong anion exchange matrix up to 1 M NaCl. When expressed in Schneider S2 cells, a considerable fraction of Dally bound to Q-Sepharose in 0.3 M salt and some was able to bind in up to 1 M salt. Dally migrated as a broad band in SDS-PAGE. The more negatively charged forms that were bound to Q-Sepharose at higher salt exhibited a higher apparent molecular weight. This may reflect the presence of more and longer HS side chains as well as differences in charge due to sulfation. Dally behaves like a conventional glypican (Tsuda, 1999). In contrast, most of the Dlp protein bound to Q-Sepharose in 0.15 M salt, but none bound at higher salt concentrations. Dlp migrated as a more tightly resolved band in SDS-PAGE. These observations suggest that Dally and Dlp differ in the extent or quality of GAG modification. Removal of the GAG addition sites considerably reduced retention of both mutant proteins on Q-Sepharose. The proportion of the mutant form of Dally that bound was reduced at all salt levels and none was bound above 0.3 M salt. Very little of the mutant form of Dlp was able to bind to Q-Sepharose, even in physiological salt (Kreuger, 2004).

To further evaluate their GAG content, lysates of S2 cells transfected to express wild-type or mutant Dally or Dlp were digested with heparitinase, and the remaining glycan stub was detected with 3G10 antibody. Dally-expressing cells showed elevated GAG labeling in a broad band that comigrated with the major form of endogenous glypican in untransfected cells. Although the level of expression of the mutant form of Dally was comparable, GAG labeling was not detectable above background. The level of GAG modification on overexpressed Dlp was lower than that seen on the endogenous glypicans. This indicates that Dlp has a lower GAG content than Dally, consistent with its poor retention on Q sepharose. The mutant form of Dlp showed much reduced GAG labeling, barely distinguishable from background levels (Kreuger, 2004).

Dally resembles a conventional glypican, being extensively modified by GAG side chains. Dlp appears to have less heterogeneous and less extensive GAG modification than Dally. The mutant form of Dlp has low levels of residual GAG and would be expected to be a poor substrate for Notum, if Notum acts directly on the GAG side chains. Coexpression of Notum and HA-tagged Dlp in S2 cells caused a large shift in the electrophoretic mobility of Dlp in SDS-PAGE to a faster migrating species, and the Notum S237A mutant had no effect. S2 cells contain endogenous Notum, so there is always a background level of Dlp processing. Interestingly, Notum caused a shift in the mobility of the low-GAG form Dlp comparable in magnitude to the shift in the wild-type protein, though the processing efficiency was lower. This would be difficult to explain if the shift were due to alteration in the amount or in the negative charge of the remaining GAG due to deacetylation. Further, no increase in deacetylase activity was detected in lysates of S2 cells overexpressing Notum, nor was there any detectable decrease in the degree of HS acetylation in larvae overexpressing Notum. These observations suggest that Notum does not act as a GAG-modifying enzyme (Kreuger, 2004).

The mobility shift of Dlp caused by Notum is ~15 kDa. Proteolytic cleavage at the N or C termini of Dlp could cause the apparent size reduction, although other modifications such as removal or attachment of lipids, sugars, or phosphate groups also could cause anomalous migration in SDS-PAGE. To test the possibility of proteolytic processing, forms of Dlp were prepared with epitope tags close to the two ends. For Dlp-HA-C, an HA epitope tag was inserted at serine 732, just before the putative cleavage site for addition of the GPI anchor (S733). To produce GFP-Dlp-HA-C, the HA tag was placed at the corresponding position in GFP-Dlp, which contains a GFP moiety inserted in place of residue G68, close to the N terminus of the protein following removal of the signal peptide at residue 41. S2 cells were transfected with Dlp-HA, Dlp-HA-C, or GFP-Dlp-HA-C together with normal or mutant forms of Notum. Notum induced comparable mobility shifts in all three forms of Dlp. The C-terminally HA-tagged forms of Dlp were processed more efficiently than Dlp-HA, so that more of the faster migrating from was observed without addition of wild-type Notum (Notum is expressed in S2 cells). These observations are difficult to reconcile with Notum acting as a protease (Kreuger, 2004).

The possibility was considered that the shift in Dlp mobility could be due to cleavage of the GPI anchor. Lysates were prepared from S2 cells transfected with Dlp-HA with or without Notum, and half of each was treated with phospholipase-C (PI-PLC) to cleave the GPI anchor of Dlp. Interestingly, PI-PLC cleavage caused a mobility shift in Dlp comparable to that caused by Notum. Notum had no additional effect on PI-PLC-treated Dlp. Although the magnitude of this shift in apparent molecular weight is larger than would be expected from the mass of the GPI anchor alone, anomalous migration of proteins in SDS-PAGE has been observed following removal of GPI anchors (Kreuger, 2004).

To further evaluate the possibility that the mobility shift caused by Notum might be due to cleavage of the GPI anchor, a Triton X-114 phase separation analysis was performed. Following phase separation, integral membrane proteins and GPI-anchored proteins are recovered mainly in the detergent phase, while other proteins are mainly in the aqueous phase. In cells cotransfected to express Dlp and Notum, processing of Dlp was efficient and the faster migrating form of Dlp was mainly recovered in the aqueous phase. In cells transfected to express Dlp, the slower migrating form was mainly recovered in the detergent phase, consistent with it having a GPI anchor. The faster migrating form partitioned between aqueous and detergent phases. These observations are consistent with the suggestion that the faster migrating form of Dlp produced by Notum lacks a GPI anchor (Kreuger, 2004).

It was next asked if Notum could act on forms of Dlp that lack a GPI anchor. The predicted GPI anchor site of Dlp consists of the sequence SDA (at S733) followed by a characteristic hydrophobic motif. Mutation of any of the three residues of the GPI anchor site to proline has been shown to reduce GPI addition in other glypicans. The SDA motif was mutated to PDA in GFP-Dlp-S988P (residue numbers are for GFP-Dlp). A truncated version of GFP-Dlp was also prepared in which a stop codon was inserted after S987 to mimic the effects of cleavage of the protein chain at this position without addition of GPI. Coexpression with Notum had little or no effect on these forms of GFP-Dlp. GFP-Dlp-S988P migrated predominantly at an intermediate position, faster than the GPI-anchored form but slower than the Notum-processed form. This presumably reflects the full-length protein that has not been cleaved for GPI addition (this form appears as a minor band in most preparations). The major band of GFP-Dlp-S988P is unaffected by Notum, though a small amount of the protein does appear to have had GPI added (perhaps at S987) and was subject to Notum activity. The truncated GFP-Dlp-S987Z protein did not produce any of the slower migrating GPI anchored form and was not affected by Notum. GFP-Dlp-S987Z migrated slightly faster than the Notum-modified form of GFP-Dlp. In Triton X-114 phase separation analysis, GFP-Dlp-S987Z partitioned into the aqueous phase, also consistent with the lack of a GPI anchor. The form of Dlp cleaved by PI-PLC comigrated more closely with the Notum-processed form, consistent with the possibility that Notum cleavage was in the GPI moiety (Kreuger, 2004).

Next a version of Dlp was prepared that was not GPI anchored, in order to ask if it would be a substrate for Notum. GFP-Dlp was fused at residue S975 to residue R23 of the transmembrane protein CD2 and expressed in S2 cells. This protein contains the entire ectodomain (CRD and stem) of Dlp, but lacks the GPI addition signals (the fusion is at the equivalent position in Dlp to the S987Z construct; numbering differs because the HA tag was removed). GFP-Dlp-CD2 was not cleaved by PI-PLC, verifying that it does not contain a GPI anchor. GFP-Dlp-CD2 was also not a substrate for Notum. If Notum acted as a protease to cleave in the Dlp protein, it would be expected that GFP-Dlp-CD2 fusion would be a substrate for Notum. Likewise, if Notum acted on the GAG side chains or on another posttranslational modification, it would be expected that GFP-Dlp-CD2 would be affected. Taken together, these observations indicate that Notum can induce modification of Dlp in a manner that resembles cleavage of the GPI anchor. Given that the alpha-hydrolases includes lipases, it is possible that Notum might act directly to cleave within the GPI anchor. Alternatively, Notum could induce the activity of an endogenous phospholipase (Kreuger, 2004).

Can cleavage at the GPI anchor cause Dlp to be shed from the cell surface? The faster and slower migrating forms of Dlp-HA were efficiently recovered from the cell lysates. In addition, the faster migrating cleaved form of Dlp-HA was recovered from the culture medium of cells overexpressing Notum. The slower migrating GPI-anchored form of Dlp-HA was not recovered from the medium. Very little Dlp-HA was recovered from medium conditioned by cells transfected with Dlp alone or with the Notum mutant. This indicates that Notum can release Dlp from the cell surface, causing it to be shed into the medium. Notum does not induce detectable shedding of HA-tagged Dally (Kreuger, 2004).

Dlp-HA or Dally-HA were expressed under en-Gal4 control in the wing disc to compare the effects of the endogenous Notum protein on Dlp and Dally. Notum is expressed at the dorsoventral boundary under control of Wg. En-Gal4 produced a uniform level of Dally-HA across the wing pouch. In contrast, Dlp-HA protein levels were lower at the dorsoventral boundary. Notum caused loss of Dlp, but had little effect on Dally. This presumably reflects shedding of Dlp protein from cells in the disc in the region where Wg levels are highest (Kreuger, 2004).

The finding that Notum can cleave and release Dlp from cells raised the possibility that Dlp might be released together with bound Wg. This could explain how Dlp acts to reduce Wg activity where Wg and Notum levels are high. Therefore the possibility was tested of synergistic action of Notum and Dlp on Wg function. Overexpressed Dlp binds and sequesters Wg at the cell surface, leading to reduced Wg activity. Coexpression of Notum with Dlp reduced the ability of Dlp to bind and retain Wg (Giraldez, 2002). To examine the effects of Notum on Dlp activity, transgenes expressing Notum or Dlp were selected at levels that were on the threshold for altering wing morphology. When expressed under en-Gal4 control, each transgene caused loss of some of the posterior wing margin bristles, indicating mild reduction of Wg activity. Expression of the two transgenes together caused scalloping of the wing, a typical defect caused by more severe reduction of Wg activity. This observation suggests that expression of Notum enhanced the ability of Dlp to reduce Wg activity. In view of the finding that Notum can cause Dlp to be released from the cell surface, it is suggested that coexpression of Notum causes Dlp to be released together with bound Wg (Notum reduced the level of Wg bound in the disc; Giraldez, 2002). Release of Wg bound to Dlp could reduce the level of Wg available for signaling and cause the observed wing scalloping phenotype. Notum causes release of Dlp from cells at the DV boundary, where Wg levels are highest (Kreuger, 2004).

Therefore, Notum can induce cleavage of Dlp in a manner that resembles PI-PLC cleavage of the GPI anchor. Notum and PI-PLC produce comparable shifts in the electrophoretic mobility of Dlp. Although the magnitude of this shift is larger than would be expected from the mass of the GPI anchor, other studies have shown that the effects of removing GPI anchors are not predictable. Examples of increased or decreased mobility have been reported, and the magnitude of the shifts can be large. Notum-induced processing also renders Dlp soluble in the aqueous phase following detergent phase separation of soluble and membrane-associated proteins, consistent with removal of the GPI anchor. Also, a Dlp-CD2 fusion protein that is not GPI anchored, and is therefore not a substrate for PI-PLC, is insensitive to Notum. This would not be expected if Notum acted on the GAG side chains or if Notum was a protease cutting within the Dlp core protein. These observations are consistent with two modes of Notum action. Notum might act directly to cleave the GPI anchor or it might modify Dlp in some way that makes Dlp a substrate for an endogenous phospholipase. The exact nature of the Notum-induced cleavage remains to be determined; however, one intriguing possibility is that Notum might cleave within the glycan linker of the GPI anchor (Kreuger, 2004).

Can the effects of Notum on Dlp provide an explanation for the apparently opposing activities of Dlp at high and low levels of Wg? How could Notum-induced cleavage of Dlp lead to reduced Wg activity, whereas removal of Dlp by RNAi increases Wg activity? If Wg remains bound to Dlp when Dlp is cleaved and shed from the cell, bound Wg would also be shed and so become unavailable for signaling. Shedding of Dlp as a consequence of Notum-induced cleavage could reduce peak levels of available Wg. Notum is expressed at the source of Wg and so would be expected to shed Dlp and reduce Wg activity where Wg levels are highest. In this way, removal of Dlp protein as a consequence of Notum-induced cleavage could have a different effect than failure to express Dlp. In the absence of Dlp, Wg would not bind Dlp and be shed with it, so that more Wg might be available to interact with Dally and/or the Wg receptor complex. This could increase the effective concentration of Wg locally near the site of Wg production (Kreuger, 2004).

This model is consistent with the observed synergy between low-level expression of Dlp and Notum. Overexpression of Dlp is thought to increase the number of Wg binding sites and shift the equilibrium toward more Wg bound to Dlp. At high levels of Dlp, this can reduce the amount of Wg available for signaling and produce a Wg loss-of-function phenotype. A level of Dlp overexpression was chosen that produces a mild defect due to reduced availability of Wg. Coexpression of Notum would lead to shedding of the Dlp bound Wg and thus remove this fraction of Wg from the pool on the cell surface so that it would no longer be able to contribute to the pool of Wg in equilibrium with the receptor. This would be expected to further reduce Wg activity and increase the severity of the defect, as observed (Kreuger, 2004).

Localized cleavage of Dlp induced by regulated expression of Notum may provide a unifying explanation for the opposing effects of the Dlp glypican in different regions of the tissue. A mathematical model of Wg gradient formation has invoked a need for an elevated level of Wg turnover in cells close to the source of Wg. The mechanism described here would be sufficient to provide the reduction of Wg activity in cells closest to the source of the secreted ligand that is needed for formation of a robust morphogen gradient (Kreuger, 2004).

Spatial regulation of Wingless morphogen distribution and signaling by Dally-like protein

Wingless is a morphogen required for the patterning of many Drosophila tissues. Several lines of evidence implicate heparan sulfate-modified proteoglycans (HSPGs) such as Dally-like protein (Dlp) in the control of Wg distribution and signaling. dlp is required to limit Wg levels in the matrix, contrary to the expectation from overexpression studies. dlp mutants show ectopic activation of Wg signaling at the presumptive wing margin and a local increase in extracellular Wg levels. dlp somatic cell clones disrupt the gradient of extracellular Wg, producing ectopic activation of high threshold Wg targets but reducing the expression of lower threshold Wg targets where Wg is limiting. Notum encodes a secreted protein that also limits Wg distribution, and genetic interaction studies show that dlp and Notum cooperate to restrict Wg signaling. These findings suggest that modification of an HSPG by a secreted hydrolase can control morphogen levels in the matrix (Kirkpatrick, 2004).

By a number of cellular and molecular markers, compromising dlp function produced ectopic activation of high-threshold Wg target genes in the wing imaginal disc. dlp adult escapers have ectopic mechanosensory bristles and dlp third instar larvae show expanded expression of Ac, a marker for sensory organ precursor cells. Wg signaling was abnormally elevated along the DV boundary of dlp mutant discs by a number of other measures, including expansion of the zone of nonproliferating cells (ZNC) and ectopic expression of Cyclin A. These findings are consistent with the expansion of sensory organ precursor cells produced by RNAi inhibition of dlp function. These phenotypes show that Dlp serves to limit activation of high-threshold target genes near the DV boundary (Kirkpatrick, 2004).

The expansion of Wg signaling found in dlp mutants is accompanied by locally elevated levels of extracellular Wg. The increase in Wg in the matrix of dlp mutant wing discs is not associated with increases in Wg production or altered expression of D-fz2, a known regulator of Wg levels. These findings argue against the model that dlp controls Wg levels indirectly via control of Wg expression or other regulatory genes. It is well established that Wg binds heparan sulfate, and therefore it is likely that Dlp controls Wg levels through a direct interaction (Kirkpatrick, 2004).

The expansion of Wg-directed patterning in the wing disc of dlp mutants was a surprising finding on a number of counts: (1) mutations in genes encoding heparan sulfate biosynthetic enzymes compromise rather than increase the levels of Wg in the matrix; (2) ectopic expression of Dlp results in higher levels of Wg in the matrix. These findings suggest that loss of Dlp would destabilize Wg on cell surfaces and in the matrix, thereby reducing Wg signaling. However, exactly the opposite result was obtained in dlp mutants: loss of Dlp increases extracellular Wg and signaling. These findings emphasize that while mutations compromising synthesis of all HSPGs might destabilize Wg in the matrix, loss of any one proteoglycan can have very different effects (Kirkpatrick, 2004).

The phenotypes of dlp mutants suggest that Dlp serves to limit Wg distribution and/or alter Wg stability, thus affecting the Wg gradient in the wing disc. Genetic mosaic analysis of dlp clearly demonstrates that the ectopic activation of Wg signaling, as measured by Arm, Cyclin A, and Ac expression, occurs non-cell autonomously, providing support for this model. Recent mathematical modeling of morphogen gradients demonstrates that binding of morphogens to receptors or coreceptors on cell surfaces or in the matrix can have profound effects on diffusion and subsequently limit morphogen activity. Consistent with such modeling, it has been demonstrated experimentally that a nonheparin binding isoform of the secreted growth factor VEGF adopts a broader, shallower distribution than heparin binding forms and the differential localization of VEGF-A isoforms in the matrix controls the vascular branching pattern. It is also well established that heparan sulfate proteoglycans mediate endocytosis of extracellular ligands, and the findings suggest the possibility that Dlp might mediate endocytic control of Wg levels in the matrix. dlp mutants display locally elevated levels of extracellular Wg without an increase in production, and therefore Wg turnover must be affected in some way (Kirkpatrick, 2004).

In contrast to the higher levels of extracellular Wg found near the DV boundary in dlp mutant wing discs, there is no apparent change in the low level of Wg at the edges of the gradient. However, genetic mosaic analysis of dlp in wing development shows that loss of dlp can modestly reduce the activation of a low-threshold Wg target gene in those regions where Wg levels are low. Since this effect on Wg signaling is not a consequence of reduced levels of Wg in the matrix, it must result from an alteration of cell responsiveness or the bioactivity of the Wg that is present. The findings are consistent with studies showing that RNAi of dlp in the wing disc also inhibits transcriptional activation of Wg target genes in regions where Wg is low. It is interesting to note that dlp expression, like that of the Wg receptor D-fz2, is highest outside the presumptive wing margin. Perhaps the high level of dlp expression in this region serves, in part, to enhance cellular responses to Wg. The ability of Dlp to enhance Wg signaling as well as limit Wg distribution is not unprecedented: the other glypican in Drosophila, Dally, enhances Dpp signaling in regions of the wing disc where Dpp levels are modest in addition to regulating the distribution of Dpp protein (Kirkpatrick, 2004).

Mutations in Notum produce ectopic activation of Wg signaling and alter the Wg gradient, similar to what was observed in dlp mutants. Notum encodes a protein with homology to plant hydrolases and has been proposed to alter the structure of Dlp, thus affecting its affinity for Wg. Coexpression of Notum with Dlp decreases Dlp-mediated stabilization of Wg protein in the wing pouch. dlp and Notum mutants fail to complement: this is strong genetic evidence that Dlp and Notum cooperate to control Wg-mediated patterning. Biochemical studies show that Notum can cleave Dlp from the cell surface. The finding that heterozygosity for dlp ameliorates patterning abnormalities produced by ectopic Notum suggests that Dlp is a principal substrate for Notum activity. The results suggest that Notum-mediated release of Dlp from the cell surface is required for Dlp to limit Wg signaling (Kirkpatrick, 2004).

Notum and dlp are expressed in complementary patterns in the wing disc, with Notum levels highest near the DV boundary and dlp levels highest elsewhere. Thus, a gradient of Dlp cleavage may be established across the disc, with some Dlp remaining on cell surfaces in regions distant from the DV boundary, enhancing cellular responses to Wg. Cleavage of Dlp would remove this coreceptor activity, but shed Dlp-Wg complexes also could limit Wg signaling by promoting Wg clearance or could serve as dominant-negative inhibitors of signaling. Coexpression of Notum and Dlp in the wing disc produced more intracellular vesicles compared to expression of Dlp alone: this is consistent with a role for shed Dlp in promoting endocytosis and clearance of Wg from the matrix (Giraldez, 2002). Such a role is also consistent with the non-cell-autonomous effects of dlp somatic cell clones on extracellular Wg distribution. Loss of dlp in clones might locally reduce Wg turnover as well as reducing cell surface binding sites for Wg, permitting increased diffusion of Wg and accumulation on wild-type cells at clone boundaries. Spatial regulation of Dlp activity by Notum can explain how Dlp primarily limits Wg signaling near the DV boundary and enhances signaling away from the boundary, but other Wg-dependent factors may also influence the ability of Dlp and Notum to downregulate Wg levels. Notably, recent mathematical modeling suggests that elevated degradation of Wg close to its source is necessary to enhance the robustness of the morphogen gradient. Together, Dlp and Notum may provide this locally increased turnover and hence stabilize the Wg gradient. Enzymatic modification of a proteoglycan to influence its cell surface localization may enable it to play both positive and negative roles in signaling and provides another potential mechanism for regulating morphogen distribution in tissues (Kirkpatrick, 2004).

In humans, mutations in glypican-3 (GPC3) were identified as the genetic basis of a human overgrowth and tumor susceptibility syndrome, Simpson Golabi Behmel Dysmorphia (SGB). SGB patients display both prenatal and postnatal overgrowth, a number of morphological abnormalities including renal dysplasia and skeletal defects, as well as a high incidence of tumors. Subsequently, loss of GPC3 expression has been found associated with a number of cancers, including breast, mesothelioma, and ovarian neoplasias. Given the established role of proteoglycans as molecules facilitating growth factor signaling at the cell surface, the effects of loss of GPC3 on growth and tumor development are a bit puzzling. The current findings provide a molecular mechanism for understanding the capacity of glypicans to serve as tumor suppressors. Dlp restricts the cellular domain of Wg signaling during wing development, and loss of dlp results in ectopic growth factor signaling. Clearly, ectopic signaling produced by loss of GPC3 could readily contribute to tumor development and growth in humans. Recently, analysis of a gene-trap mutant in mouse Ext1 has provided further evidence for the capacity of heparan sulfate proteoglycans to limit the range of morphogen activity during chondrocyte differentiation. Mice partially defective for heparan sulfate biosynthesis as a consequence of hypomorphic mutations in Ext1 show ectopic Indian Hedgehog signaling and altered Hedgehog distribution at the growth plate (Kirkpatrick, 2004 and references therein).

The Wingless morphogen gradient is established by the cooperative action of Frizzled and Heparan Sulfate Proteoglycan receptors

The respective contribution of Heparan Sulfate Proteoglycans (HSPGs) and Frizzled (Fz) proteins in the establishment of the Wingless (Wg) morphogen gradient has been examined. From the analysis of mutant clones of sulfateless/N-deacetylase-sulphotransferase in the wing imaginal disc, it was found that lack of Heparan Sulfate (HS) causes a dramatic reduction of both extracellular and intracellular Wg in receiving cells. These studies reveal that the Glypican molecule Dally-like Protein (Dlp) is associated with both negative and positive roles in Wg short- and long-range signaling, respectively. In addition, analyses of the two Fz proteins indicate that the Fz and DFz2 receptors, in addition to transducing the signal, modulate the slope of the Wg gradient by regulating the amount of extracellular Wg. Taken together, this analysis illustrates how the coordinated activities of HSPGs and Fz/DFz2 shape the Wg morphogen gradient (Baeg, 2004).

Sfl encodes a homolog of the Golgi enzyme HS N-deacetylase/N-sulfotransferase that is required for the modification of HS. It was of interest to determine whether the retention of Wg at the cell surface involves HSPGs in receiving cells, since it has been proposed that HSPGs are unlikely to be required in Wg-receiving cells. In the wing blade, Wg, originating from the D/V border, is detected in an irregular pattern of puncta in receiving cells; these puncta correspond to the internalized Wg protein. The intensity and number of puncta decreases from the source of Wg. Furthermore, using an extracellular labeling method, a gradient of Wg protein that appears broader, shallower, and with less puncta is observed. Both in sfl mutant wing discs and in large sfl mutant clones, a striking decrease in the number of Wg puncta was observed. This decrease is not due to a change in wg transcription because it is not affected in sfl mutant cells. Further, lack of Sfl activity does not appear to disrupt the overall amount of Wg produced by wg-expressing cells but is associated with a dramatic decrease in extracellular Wg. These results suggest that HSPGs are required for sequestering extracellular Wg in receiving cells (Baeg, 2004).

To gain further insights into the role of HSPGs in receiving cells to shape the Wg gradient, an examination has been made of the distribution of Wg in patches of WT cells located within a large sfl mutant territory. In such cases, bright spots of Wg were be detected within the patch of WT cells, indicating that sfl-expressing cells are able to sequester extracellular Wg, unlike neighboring cells that lack sfl. This result is consistent through an analysis of more than 10 clones. Further, clones of cells that overexpress Notum-GT (Golgi-tethered), which acts cell autonomously in receiving cells, were generated. Notum, which encodes a member of the α/β-hydrolase superfamily, antagonizes Wg signaling and it has been proposed that Notum acts by altering the ability of the cell surface glypican molecules Dally and Dlp to stabilize extracellular Wg (Giraldez, 2002). Consistent with the conclusion that HSPGs are required in receiving cells to capture extracellular Wg, a decrease in the formation of Wg puncta was observed in cells overexpressing Notum-GT. These results are consistent with at least two nonexclusive models: (1) HSPGs could be required for Wg stability and/or trapping of Wg at the cell surface such that it does not diffuse away; (2) HSPGs could be involved in promoting Wg movement throughout tissues. The role of HSPGs in sequestering and/or stabilizing the ligand is supported by previous observations that overexpression of either Dlp or Dally results in the accumulation of extracellular Wg (Baeg, 2004).

Because HSPGs have been implicated in endocytosis of ligands such as FGF, possibly HSPGs also play a role in Wg internalization. A role for HSPG in Wg endocytosis would be consistent with the absence of puncta in sfl clones, and also the observed accumulation of extracellular Wg following overexpression of Dlp-HA (Giraldez, 2002). However, much of Wg proteins appeared in intracellular vesicles, instead of outlining the cell surface, in discs overexpressing both Dlp-HA and Notum. If HSPGs were directly involved in Wg internalization, fewer intracellular vesicles would be detected in discs overexpressing Dlp-HA and Notum since Notum acts to decease the affinity of Dlp for Wg. Furthermore, if the primary function of HSPGs were to internalize Wg, then extracellular Wg accumulation in cells lacking HSPGs activity would be expected, which is not the case. However, the data does not rule out the possibility that HSPGs play a direct role in Wg endocytosis, and, thus, further analysis will be required to clarify this issue. Taken together, these results suggest that the primary role of HSPGs is to trap and/or stabilize extracellular Wg in receiving cells where it is then able to interact with its signaling receptor as well as other factors that are responsible for its internalization, and thus contributes to shaping the Wg gradient (Baeg, 2004).

Previous ectopic expression studies have shown that Dlp can trap extracellular Wg and prevent activation of the Wg signaling pathway. Because Dlp appears to be a major HSPG required to regulate Wg signaling, its endogenous distribution was examined in the wing imaginal disc using a polyclonal antibody against Dlp and a staining method that primarily detects extracellular proteins. The specificity of the Dlp antibody was confirmed by misexpressing dlp using the ap-Gal4 driver. In the third instar wing imaginal disc, Dlp was detected throughout the disc; however, a significant decrease in the level of Dlp was detectable at the D/V border. This domain of low Dlp expression correlates with the region where high level of Wg signaling is required to induce the expression of short-range target genes. It is noted that since dlp mRNA expression is uniform throughout the disc, the down-regulation of Dlp at the D/V border must occur post-translationally. Interestingly, an optical cross section of the disc has revealed that endogenous Dlp localizes mostly on the basolateral surface of the cell where extracellular Wg is detected. The subcellular localization of Dlp protein was also examined using a GFP-dlp expressed under the control of a Gal4 driver. Consistent with the Dlp antibody result, it was found that GFP-Dlp localizes predominantly to the basolateral membrane. Altogether, these observations suggest that Dlp can bind to extracellular Wg and that Dlp levels need to be reduced for high-level Wg activity in cells near the D/V boundary (Baeg, 2004).

In a genome-wide RNAi screen in S2R+ cells to identify genes that either up- or down-regulate Wg signaling, Dlp was identified as both a negative and positive regulator of Wg signaling under stimulated and nonstimulated conditions, respectively. The cell-based assay devised consists of the activation of a Tcf/Arm-dependent Wg-reporter gene upon induction of S2R+ cells by expressing a wg cDNA by transient transfection. The activity of the Wg pathway and the effect of the addition of various dsRNAs on the pathway were assayed by monitoring Luciferase-reporter activity using a luminescence-based plate reader. Using this assay, the addition of dsRNAs of positive transducers of Wg signaling, such as arm, decrease the Top12X-HS-Luciferase reporter activity, while dsRNAs to negative Wg regulators, such as Daxin, increase its activity. Interestingly, under condition of Wg induction, dlp was found to act as a negative regulator of Wg signaling, since dlp dsRNA led to a twofold increase in luciferase activity. This increase is significant since it is comparable to that of daxin dsRNA. In the absence of Wg induction, dlp was found to positively regulate Wg signaling, since dlp dsRNA leads to a fivefold decrease in luciferase activity, a decrease that is similar to that observed by the addition of arm dsRNA. These results suggest Dlp acts as a positive regulator of the Wg pathway when Wg level is low and negatively influences signaling when Wg is abundant. These results are consistent with in vivo results that demonstrate that Dlp has both a positive and negative role in Wg signaling (Kirkpatrick, 2004). The observations in S2R+ are consistent with the hypothesis that low Dlp levels at the D/V boundary supports high-level Wg signaling, while further away from the D/V boundary where the Wg concentration is lower, Dlp positively influences Wg signaling. Overall, the result from the S2R+ RNAi experiments indicates that (1) Dlp is not an essential component of the Wg signal transduction pathway; and (2) Dlp can either have a negative or positive impact on Wg signaling depending on the level of Wg available. The negative effect of Dlp is consistent with previous in vivo studies that have shown ectopic Dlp expression can trap extracellular Wg and prevent activation of the Wg signaling pathway. Therefore, a reduction of Dlp levels at the D/V border would be expected to contribute to high-level Wg signaling. The positive effect of Dlp in Wg signaling needs to be understood in the context of the previous findings that loss of HSPG activity results in wg loss-of-function phenotypes, as shown by a decrease in dll expression in clones mutant for enzymes involved in GAG biosynthesis. One attractive model is that Dlp would act as a co-receptor that traps/stabilizes extracellular Wg and facilitates its association with the signal transducing Fz receptors in cells located at a distance from the D/V boundary where low level of Wg is available. Finally, these observations are consistent with a recent study (Kirkpatrick, 2004) that showed that (1) ectopic activation of Wg signaling at the wing margin occurs in dlp mutant tissues, and (2) a cell autonomous reduction in Wg signaling in dlp clones located distal to the Wg-producing cells (Baeg, 2004).

The distribution of Dlp protein is reminiscent of the down-regulation of Dfz2 transcription near the D/V border. Wg-mediated repression of DFz2 expression has been shown to affect the shape of the Wg gradient, resulting in a gradual decrease in Wg concentration. Because these results indicate that HSPGs affect Wg distribution, the functions of the two seven transmembrane Wg receptors, Fz and DFz2, were examined to evaluate how the signal transducing receptors cooperate with HSPGs in shaping the Wg gradient. To determine the role of Fz and DFz2 in Wg movement, the distribution of Wg in fz DFz2 double-mutant clones was examined. In these clones, an expansion of wg expression was observed, which is consistent with the previously described Wg “self-refinement” process, by which Wg signaling represses wg expression in cells adjacent to wg-expressing cells. Unexpectedly, within these clones, Wg puncta are still present, indicating that Fz/DFz2 receptor activities are not required for Wg spreading (Baeg, 2004).

To determine whether Wg is present in endosomes in the absence of Fz/DFz2 activities, wing discs were labeled with the endosomal marker Texas-red dextran. More than 50% of Wg puncta co-localize with red dextran, indicating that Wg is internalized in the absence of Fz/DFz2 activities. These observations are consistent with results in the embryo, and altogether suggest that internalization of Wg can be accomplished by proteins other than Fz/DFz2. Interestingly, this observation contrasts with the role of HSPGs in Wg distribution, since wing discs lacking GAGs show alteration in Wg puncta in receiving cells. The extracellular distribution of Wg was examined in fz DFz2 mutant clones. Interestingly, accumulation of extracellular Wg was detected throughout these clones, thus revealing that Wg can bind to the cell surface and that Fz/DFz2 receptors are required somehow for Wg degradation. To exclude the possibility that accumulation of extracellular Wg results from increased wg expression or secretion in fz DFz2 clones that cross D/V boundary, small clones that do not include the D/V boundary were generated. Accumulation of extracellular Wg was detected in these clones, which is reminiscent of the finding that overexpression of a dominant-negative form of DFz2 (ΔDFz2-GPI) driven by en-Gal4 in embryonic tissue prevents Wg decay within the en domain. It has been proposed that endocytosis of a Wg/receptor complex is responsible for down-regulating Wg levels. Further, because Wg is still organized in a graded manner in these clones, as shown by the distribution of the Wg puncta, it indicates that Wg movement can occur in the absence of Fz/DFz2 (Baeg, 2004).

There is a third member of the Frizzled family encoded by DFz3 that could influence the distribution of Wg in tissues. DFz3 expression is similar to that of wg, and a constitutively activated form of Arm up-regulates its expression in the wing disc, suggesting that DFz3 is transcriptionally regulated by Wg signaling. Based on these observations, little or no DFz3 protein would be expected to be present in cells that lack Fz/DFz2 activity, suggesting that internalization of Wg in Fz/DFz2 mutant cells is unlikely to be mediated by DFz3. Another candidate that could affect Wg distribution is Arrow, which is a Drosophila homolog of a low-density lipoprotein (LDL)-receptor-related protein (LRP) and has been shown to be essential in cells receiving the Wg signal. However, because a soluble form of the Arrow fails to bind Wg and Fz receptors in vitro, and because Arrow functions after DFz2 engages Wg, it is unlikely that Wg internalization in Fz/DFz2 mutant cells is mediated by Arrow. Finally, as is case for FGF endocytosis, HSPGs themselves possibly play a role in Wg internalization (Baeg, 2004).

In summary, there is a Fz/DFz2 receptor-independent mechanism that organizes Wg distribution, and Fz/DFz2 proteins play a role in Wg gradient formation by decreasing the level of extracellular Wg. Regulation of extracellular Wg levels by Fz/DFz2 may occur through receptor-mediated endocytosis, or by some other mechanisms. If Wg degradation occurs by receptor-mediated endocytosis, it indicates that there may exist more than one way to generate Wg puncta since these are still present in the absence of Fz/DFz2 receptor activity. These findings also emphasize that the amount of Fz/DFz2 receptors at the cell surface must be precisely regulated to achieve the proper spreading of Wg, an observation that is underscored by the transcriptional down-regulation of DFz2 expression near the source of Wg (Baeg, 2004).

To further examine the role of Fz/DFz2 receptors in Wg gradient formation, DFz2 was overexpressed at the D/V boundary using the C96-Gal4 driver, and the effect on Wg distribution and wing patterning was examined. Analysis was focused on DFz2 since DFz2 has been shown to bind Wg with high affinity and to stabilize it. Further, DFz2 expression is down-regulated by Wg signaling, and this regulation has been shown to play a critical role in the overall shape of the Wg gradient. Interestingly, ectopic expression of DFz2 results in wing notching and ectopic bristles at the wing margin of adult wing. Previous studies have shown that wing nick phenotypes result from an inhibition in Wg signaling activity while the presence of ectopic bristles on the wing blade corresponds to an increase in Wg signaling. Thus, based on the wing phenotypes, it appears that overexpression of DFz2 paradoxically both increases and decreases Wg signaling (Baeg, 2004).

Overexpression of the DFz2 could interfere with Wg signaling and its distribution in a number of ways. For example, an increase in DFz2 could increase the efficiency of Wg signaling, if the amount of receptor is limiting. Further, since wg expression in the wing disc is restricted to the D/V margin, and Wg diffuses from it, trapping of Wg near these cells most likely will have an effect on Wg short- and long-range activity since the shape of the Wg gradient will be disrupted. To distinguish between these possibilities, Wg distribution was examined in discs with clones of cells that overexpress DFz2 at the D/V boundary. These clones were associated with two effects on Wg distribution. (1) The level of Wg was increased in the clones of cells where DFz2 was overexpressed, indicating that an increase in the level of DFz2 in receiving cells leads to an increase in trapping extracellular Wg. This observation is consistent with the occurrence of extra bristles on the wing blade since they reflect high levels of Wg signaling activity. (2) A dramatic reduction in Wg puncta was detected in WT cells located adjacent to the cells overexpressing DFz2, suggesting that Wg movement from the D/V margin into the wing blade is impaired as a result of the excess trapping of Wg by cells that overexpress DFz2. To demonstrate that Wg accumulation correlates with an increase in Wg signaling and that the absence of Wg puncta correlate with an absence of Wg signaling, the effect of DFz2 overexpression on the expression of senseless was examined. sen expression is expanded in cells overexpressing DFz2, yet sen is not expressed in WT cells near a clone of cells overexpressing DFz2. This is consistent with the observation that more Wg can be detected in cells overexpressing DFz2 and that less Wg puncta are present in WT cells near a clone of cells overexpressing DFz2 (Baeg, 2004).

It has been proposed that Fz proteins contribute to Wg turnover. Thus, it is intriguing to note that overexpression of DFz2 leads to an accumulation of extracellular Wg. This may reflect saturation of the endocytotic pathway when DFz2 is overexpressed or an inability of the regulatory pathways that normally control Dfz2 endocytosis in the wing disc to appropriately respond under this overexpressed condition. Another possibility is HSPGs themselves might play an important role. Wg endocytosis and the stoichiometry of Fz to HSPGs is essential to promote proper Wg internalization. Detailed biochemical and cell biological studies are now required to clarify the role(s) of these receptors in Wg movement (Baeg, 2004).

Finally, whether overexpression of DFz2 at the D/V boundary could affect long-range activity of Wg was examined, using wing disc overexpressing DFz2 driven by C96-Gal4 driver. Interestingly, Dll expression is dramatically shortened in wing disc overexpressing DFz2 at the D/V boundary when compared to that of WT disc. It is concluded that DFz2 has multiple roles in Wg signaling: First, it transduces Wg signaling and its level is limiting in amount; and second, it affects Wg short- and long-range activity by modulating the availability of extracellular ligand (Baeg, 2004).

In this study, the respective roles of HSPGs and Fz/DFz2 receptors in Wg distribution and gradient formation were examined. Interestingly, it was found that loss of Dlp activity significantly increases the level of Wg activity in S2R+ cells upon Wg induction, indicating that Dlp acts as a negative regulator in Wg signaling and that it is not required for transducing the Wg signal. Interestingly, the in vivo results show that Dlp protein levels are low near the D/V boundary. Thus, low levels of Dlp near the source of Wg production may allow for activation of high threshold Wg target gene. It is of interest to note that Notum is highly expressed along the D/V boundary (Giralez, 2002), which would be predicted to further diminish HSPGs activity. In addition, it was found that Dlp positively influences Wg signaling in S2R+ cells when Wg is not induced, suggesting that it is required for Wg signaling in cells where Wg level is low. A possible explanation for this result is that Dlp may act as a co-receptor that traps/stabilizes extracellular Wg and facilitates its association with signal transducing Fz receptors. In the wing imaginal disc, given that Dlp is required for Wg signaling in cells where Wg levels are low, HSPG activity is possibly required for Wg signaling by somehow facilitating Wg movement. The binding of extracellular Wg to the low-affinity HSPG receptors in receiving cells may result in the association of Wg to cell membranes. Ligand movement could then occur by a mechanism that directly involves HPSGs where subsequent cycles of Wg dissociation/reassociation with HSPGs might promote the movement or require other yet to be identified extracellular molecules. To distinguish these possibilities, the role of HSPGs in Wg movement will require further detailed analysis. Regardless, these results clearly indicate that the primary role of HSPGs is to sequester and/or stabilize extracellular Wg in receiving cells. The imposition of the HSPG-mediated Wg accumulation and the Fz-dependent degradation mechanism would thus contribute to the Wg morphogen gradient. It is important to note that the expression levels of some of the critical components of each systems (i.e., dally, DFz2) are also regulated by the Wg pathway itself, indicating that the slope of the Wg gradient is established by the delicate balance between these two systems (Baeg, 2004).

Dally and Dally-like shape the extracellular Wingless morphogen gradient in the wing disc

During the wing development Wingless acts as a morphogen whose concentration gradient provides positional cues for wing patterning. The molecular mechanism(s) of Wg gradient formation is not fully understood. This study systematically analyzes the roles of glypicans Dally and Dally-like protein (Dlp), the Wg receptors Frizzled (Fz) and Fz2, and the Wg co-receptor Arrow (Arr) in Wg gradient formation in the wing disc. Both Dally and Dlp are essential and have different roles in Wg gradient formation. The specificities of Dally and Dlp in Wg gradient formation are at least partially achieved by their distinct expression patterns. Surprisingly, although Fz2 has been suggested to play an essential role in Wg gradient formation by ectopic expression studies, removal of Fz2 activity does not alter the extracellular Wg gradient. Interestingly, removal of both Fz and Fz2, or Arr causes enhanced extracellular Wg levels, which mainly results from upregulated Dlp levels. It is further shown that Notum, a negative regulator of Wg signaling, downregulates Wg signaling mainly by modifying Dally. Last, it is demonstrated that Wg movement is impeded by cells mutant for both dally and dlp. Together, these new findings suggest that the Wg morphogen gradient in the wing disc is mainly controlled by combined actions of Dally and Dlp. It is proposed that Wg establishes its concentration gradient by a restricted diffusion mechanism involving Dally and Dlp in the wing disc (Han, 2005).

One important finding in this work is that Dally and Dlp are required for Wg gradient formation. Several recent studies have shown that extracellular Wg distribution is compromised in clones mutant for HS biosynthesis enzymes, including sfl, slalom and members of the Drosophila EXT gene. However, it is unclear which HSPG cores are involved in this process. This study shows that Wg morphogen distribution is defective in either dally or dlp mutant clones. These new findings clearly establish the requirement of Dally and Dlp in Wg morphogen gradient formation. Thus, as in the case of Hh and Dpp, the glypican members Dally and Dlp, rather than Drosophila syndecan or perlecan, are the main HSPGs involved in Wg gradient formation (Han, 2005).

Interestingly, Dally and Dlp differentially regulate the Wg extracellular gradient in distinct regions of the wing disc. Both Dally and Dlp are glypican members of HSPG family. One would expect that differences in the structure of Dally and Dlp, and their attached HS GAG chains may determine their abilities to interact with Wg, thereby leading to their specificities. This is probably one of the factors, since overexpression of Dally and Dlp has very different effects on extracellular Wg gradient. Consistent with the data in this work, previous studies have shown that Dlp is much more potent in accumulating Wg protein than Dally when overexpressed. However, the regional effects of Dally and Dlp on extracellular Wg gradient correspond well to their expression patterns. The regions with higher expression levels of Dally or Dlp have stronger extracellular Wg defects when Dally or Dlp is removed, respectively. Based on these data, it is suggested that the differential roles of Dally and Dlp in extracellular Wg distribution are at least partially determined by their restricted expression (Han, 2005).

What exact roles do Dally and Dlp play in shaping the extracellular Wg gradient? Loss-of-function results suggest that removal of Dally or Dlp leads to reduced extracellular Wg levels on the cell membrane. Furthermore, extracellular Wg levels are reduced in wild-type cells behind sfl or dally-dlp clones. These data suggest that the primary function of Dally and Dlp in Wg gradient formation is to maintain extracellular Wg proteins so that locally concentrated Wg proteins can further move to more distal cells through diffusion (Han, 2005).

Despite a positive role of Dlp in extracellular Wg distribution, surprisingly, Dlp negatively regulates Wg signaling at the DV boundary. However, ectopic Wg signaling at the DV boundary of the dlp mutant is not as great as expected. This relatively weak effect is most probably due to the low level expression of Dlp, which is downregulated by Wg signaling. The results are consistent with a previous observation that overexpression of Dlp in the wing disc leads to a blockage of Wg signaling. Dlp may compete with Fz proteins for available Wg protein at the DV boundary, thereby inhibiting Wg signaling. However, the extract mechanism of Dlp-mediated Wg inhibition needs to be further determined (Han, 2005).

Previous studies have identified Notum as a secreted inhibitor for Wg signaling. Notum is expressed at the DV boundary and has been proposed to downregulate Wg signaling by modulating Dlp activity (Giraldez, 2002). Kreuger (2004) and Selleck (Kirkpatrick, 2004) proposed that Notum negatively regulates Wg signaling by shedding of Dlp, which converts Dlp from a membrane-tethered co-receptor to a secreted antagonist. Their conclusions are mainly based on two lines of experimental data: (1) biochemical experiments clearly demonstrated that Notum can modify Dlp in a manner that resembles cleavage of the GPI anchor (Kreuger, 2004); (2) Kirkpatrick (2004) showed that transheterozygous dlp/notum flies produced ectopic mechanosensory bristles which are not seen in dlp^+/- or notum^+/- alone, indicating that Dlp and Notum genetically collaborate in downregulating Wg signaling (Han, 2005).

However, on the basis of the current data, it is suggested that Notum inhibits Wg signaling mainly by modifying Dally in the wing disc. (1) Genetic interaction data shown by Kirkpatrick (2004) cannot distinguish whether Dlp and Notum work in the same pathway or in two independent pathways to downregulate Wg signaling at the DV boundary (Kirkpatrick, 2004). If Dlp is indeed the main substrate for Notum, it would be expected that ectopic Wg signaling activity in dlp-notum should be similar to that in dlp mutant. However, loss-of-function analysis demonstrates that ectopic Wg signaling in dlp-notum is similar to that in notum mutant, but much stronger than that in dlp mutant. However, dally-notum clones exhibit loss of Wg signaling activity that is similar to dally mutant. (2) Dlp expression is strikingly repressed by Wg signaling and this reduction is independent of Notum. Low/absent expression of Dlp is not consistent with the view that Dlp is the main substrate for Notum. (3) It is important to mention that Notum can reduce the amount of Dally when they are co-expressed in Drosophila S2 cells (Giraldez, 2002), suggesting that Notum can modify Dally as well. Although Notum can shed Dlp, whether shed Dlp acts as a Wg inhibitor needs to be further determined. Therefore, further experiments are necessary to define the mechanism(s) of Notum-mediated Wg inhibition (Han, 2005).

One important finding of this study is that removal of the Wg receptors (Fz and Fz2) and the co-receptor Arr does not lead to a loss of extracellular Wg. Fz2 has been proposed to play a major role in Wg gradient formation in the wing disc by ectopic expression studies. Although the high capacity of Fz2 in stabilizing Wg has been demonstrated, loss-of-function results clearly show that extracellular Wg levels are not reduced in clones mutant for fz2. This is apparently not due to the overlapping function of Fz, since the extracellular Wg level is enhanced, rather than reduced, in the absence of both Fz and Fz2 functions. The results argue that Fz2 is not essential for extracellular Wg gradient formation in vivo. It is important to note that in addition to Fz and Fz2, Fz3 is also expressed in the wing disc and its expression is upregulated by Wg signaling. Although Fz3 has lower affinity than Fz2 in Wg binding and acts as an attenuator of Wg signaling, its role in Wg distribution needs to be determined (Han, 2005).

It is further demonstrated that extracellular Wg is enhanced in cells mutant for fz-fz2 or arr, suggesting that Wg receptors (Fz and Fz2) and Arr shape extracellular Wg gradient by downregulating extracellular Wg levels. The data argue that this mainly results from upregulation of Dlp. Consistent with this view, the accumulated extracellular Wg can be eliminated by loss of HSPGs in sfl-fz-fz2 or arr-botv mutant clones. Importantly, it is shown that both extracellular Wg and Dlp levels are upregulated on the cell surface of clones mutant for dsh. These data provide compelling evidence that though a feedback mechanism, Wg signaling can control the Dlp levels to regulate the extracellular Wg gradient (Han, 2005).

Another alternative possibility is that enhanced Wg levels in fz-fz2 or arr clones may be caused by impaired Wg internalization. Although a significant amount of internalized Wg vesicles has been demonstrated in fz-fz2 or arr mutant clones, this possibility cannot be ruled out, since a quantitative comparison of Wg internalization between wild-type cells and fz-fz2 or arr mutant cells is difficult. Furthermore, as mentioned above, Fz3 is expressed in the wing disc and its expression is upregulated by Wg signaling. It is possible that Fz3 may mediate the internalization of Wg in the absence of Fz and Fz2 (Han, 2005).

Evidence has been presented that Wg morphogen movement is regulated by a diffusion mechanism(s) in the wing disc. Does Wg diffuse freely in the extracellular matrix/space? In this work, it is shown that Wg fails to move across a strip of cells mutant for the HSPGs Dally and Dlp. This result suggests that Wg cannot freely diffuse in the extracellular matrix. Instead, the findings are consistent with a model in which Wg movement is mediated by the HSPGs Dally and Dlp through a restricted diffusion along the cell surface. Similar mechanisms have been proposed for Hh and Dpp. In biological systems such as imaginal discs, the restraint of Wg spreading to the surface of the epithelial cell layer is important since the folding of imaginal discs, such as the leg disc, poses a problem if the Wg gradient formation were to occur out of the plane of the epithelial cell layer through free diffusion. In agreement with this view, the model proposes that Wg gradient formation depends on Wg movement through the cell surface of the disc epithelium (Han, 2005).

Glypicans shunt the Wingless signal between local signalling and further transport

The two glypicans Dally and Dally-like have been implicated in modulating the activity of Wingless, a member of the Wnt family of secreted glycoprotein. So far, the lack of null mutants has prevented a rigorous assessment of their roles. A small deletion was created in the two loci. Analysis of single and double mutant embryos suggests that both glypicans participate in normal Wingless function, although embryos lacking maternal and zygotic activity of both genes are still capable of transducing the signal from overexpressed Wingless. Genetic analysis of dally-like in wing imaginal discs leads to a model whereby, at the surface of any given cell of the epithelium, Dally-like captures Wingless but instead of presenting it to signalling receptors expressed in this cell, it passes it on to neighbouring cells, either for paracrine signalling or for further transport. In the absence of dally-like, short-range signalling is increased at the expense of long-range signalling (reported by the expression of the target gene distalless) while the reverse is caused by Dally-like overexpression. Thus, Dally-like acts as a gatekeeper, ensuring the sharing of Wingless among cells along the dorsoventral axis. This analysis suggests that the other glypican, Dally, could act as a classical co-receptor (Franch-Marro, 2005).

The fact that mutations in dally and dlp cause different phenotypes suggests that, although they both underpin Wingless function, these two glypicans could perform distinct activities. It is likely that both Dally and Dlp are able to capture Wingless at the surface of imaginal disc cells. From the point of view of a given cell in vivo, Wingless captured by Dally would be mostly destined for 'internal consumption', while Dlp-bound Wingless would be for export only. Subsequent long-range transport would occur by hopping from Dlp on one cell to Dlp on the next. Both glypicans would contribute to increasing the concentration of Wingless at the cell surface (Dally in cis and Dlp in trans). It is suggested that in the embryo too, Dlp and Dally help in the presentation and reception of Wingless, respectively. However, in this system, little Wingless transport takes place, maybe because release of Wingless from Dlp is not allowed. It is interesting that, in embryos, dlp is highly expressed in cells that secrete Wingless. Therefore, the role of Dlp would mainly be to ensure that plenty of Wingless is retained at the surface of Wingless-expressing cells thus allowing sustained short-range signalling. In both the embryonic and disc systems, the genetic redundancy between dally and dlp could be viewed as follows: reduction of capturing activity in dally mutants would be compensated by the 'presentation activity' of Dlp and vice versa. Further cell biological work will be needed to fully explore the specific activities of Dally and Dlp and also to discover how Wingless is transferred from one cell to another during transport, perhaps with the help of enzymes such as Notum/Wingful (Franch-Marro, 2005).

Drosophila heparan sulfate 6-O endosulfatase regulates Wingless morphogen gradient formation

Heparan sulfate proteoglycans (HSPGs) play critical roles in the distribution and signaling of growth factors, but the molecular mechanisms regulating HSPG function are poorly understood. This study characterized Sulf1, which is a Drosophila member of the HS 6-O endosulfatase class of HS modifying enzymes. Genetic and biochemical analyses show that Sulf1 acts as a novel regulator of the Wg morphogen gradient by modulating the sulfation status of HS on the cell surface in the developing wing. Sulf1 affects gradient formation by influencing the stability and distribution of Wg. It was also demonstrated that expression of Sulf1 is induced by Wg signaling itself. Thus, Sulf1 participates in a feedback loop, potentially stabilizing the shape of the Wg gradient. This study shows that the modification of HS fine structure provides a novel mechanism for the regulation of morphogen gradients (Kleinschmit, 2010).

Although roles for HSPGs in morphogen signaling and distribution have been well established, the molecular basis of these activities remains to be elucidated. A number of genetic and in vitro analyses have demonstrated the critical importance of HS moieties for HSPG function during developmental patterning. Recent reports using mutant forms of Dally and Dlp that lack all HS attachment sites have revealed the essential contribution of the core protein to regulating growth factor binding and signaling activity. Thus, the regulatory function of HSPGs is likely to be affected by a combination of both HS and core protein structures (Kleinschmit, 2010).

HS biosynthesis is a complex, multi-step process catalyzed by Golgi enzymes in a highly organized fashion. Recent studies have demonstrated that extracellular Sulfs further modify the HS fine structures in a post-synthetic manner (reviewed in Gorsi, 2007). Thus, Sulfs may contribute to generating structural diversity and modify the number of ligand binding sites on HS at the cell surface (Kleinschmit, 2010).

To better understand the importance of regulating HS sulfation during development, this study investigated the role of Drosophila Sulf1 in patterning and morphogenesis. Sulf1 mutant wings show specific phenotypes characteristic of abnormally high levels of Wg signaling near the Wg-expressing cells. Extracellular levels of Wg protein were elevated throughout the Sulf1 mutant wing discs, and decreased in cells overexpressing Sulf1. In addition, a Sulf1 mutation caused Wg to accumulate near its source, altering the shape of the gradient. Thus, Sulf1 is a novel regulator of Wg gradient formation. Disaccharide analysis of Sulf1 mutant HS showed abnormally high levels of tri-S disaccharide units, indicating that Sulf1 regulates Wg signaling by modulating HS fine structure. Given that Sulf1 decreases local levels of Wg protein in the extracellular space, it is likely that the domain structure of HS to which the Wg ligand preferentially binds includes tri-S disaccharide unit(s) as a major component. Thus, Sulf1 may redefine the shape of the Wg gradient by removing some of the Wg-binding sites from HS on the cell surface (Kleinschmit, 2010).

It has been shown in Drosophila embryos that a significant fraction of Wg protein is retained on the expressing cells in a HSPG-dependent manner. High levels of dally expression near the DV boundary of the wing disc suggest that Wg may be also trapped by HS in the developing wing. In such a situation, Sulf1 activity could reduce the trapping of Wg by cell surface HSPGs near the expressing cells. Wg protein, thus released from HS, could have two possible fates. First, Wg ligand could be quickly internalized by nearby cells for degradation. Second, released Wg ligand could escape degradation and migrate away from the trapped site. Therefore, theoretically, Sulf1 can affect the Wg gradient through two differential activities: (1) destabilization of Wg and (2) enhancement of Wg re-distribution by facilitating Wg release from the HSPGs. This study showed that Sulf1 reduces extracellular levels of Wg protein without affecting wg expression. In addition, Wg signal intensity plots for wild-type and Sulf1 mutant discs suggested that Sulf1 affects Wg distribution near the DV boundary. Thus, these observations are consistent with the idea that Sulf1 indeed modulates the Wg gradient by influencing both Wg stability and distribution (Kleinschmit, 2010).

How can Sulf1 contribute to lateral distribution of Wg? Gallet (2008) has proposed that Dally-like (Dlp) mediates apicobasal trafficking of Wg, which is required for its long-range gradient formation. A more recent study has shown that Dlp can act in a biphasic manner to potentiate Wg long-range signaling (Yan, 2009). In this model, Dlp either competes with the receptor or provides ligand to the receptor, dependent on its ratio to Wg and the receptor. In both models, however, since dlp expression is repressed at the DV compartment border, an additional mechanism by which Wg reaches the dlp-expressing cells appears to be required. Wg secreted from cells at the DV boundary is likely to be first trapped by Dally, a glypican expressed at high levels in this region. One possible function of Sulf1 is to facilitate the short-range movement of Wg from the expressing cells to Dlp. In this model, Sulf1, which is also abundant near the source of Wg, removes 6-O sulfate groups from Dally HS chains. This enzymatic cleavage would lower the efficiency of Wg trapping by Dally, allowing it to migrate away from the DV boundary. Released Wg would now have a better chance to reach Dlp, which recaptures and facilitates further diffusion of Wg (Gallet, 2008; Yan, 2009). Thus, this study demonstrates that modification of HS fine structure provides a novel mechanism to shape morphogen gradients (Kleinschmit, 2010).

Given that vertebrate Sulfs are known to positively regulate Wnt signaling, it is surprising that Drosophila Sulf1 has an opposite effect on the Wg pathway. The results suggest that Drosophila Sulf1 has a similar biochemical activity and it is expected that a direct consequence of the function of Sulf enzymes on Wnt/Wg protein is also similar between vertebrate and invertebrate models: Sulfs release Wnt/Wg ligands from HSPGs. It is proposed that the fate of the released Wnt/Wg could be different dependent on extracellular environment. In vertebrate systems where Sulfs enhance Wnt signaling, released Wnt appears to have better chance to bind and activate receptors. In contrast, a major fraction of Wg protein detached from HSPGs may be degraded in the Drosophila wing disc (Kleinschmit, 2010).

Although Sulfs are believed to function post-synthetically in the extracellular space, the effects of Sulf1 function were observed cell autonomously. In addition, experiments using Sulf1-Golgi showed that this modified form retains the ability to decrease extracellular levels of Wg protein, indicating that Sulf1 does not have to be secreted into the extracellular space to function. Thus, Sulf1 may act in the Golgi and/or on the cell surface. If Sulf1 acts extracellularly, Sulf1 is likely to adhere to the surface of the secreting cells as has been shown in vertebrate models: previous studies reported that Sulf enzymes associate with the cell fraction and not the medium fraction of transfected cultured cells. The binding of QSulf1 to the cell fraction in CHO cells has shown to be dependent on a large hydrophilic domain. Since a similar conserved hydrophilic domain is found in Drosophila Sulf1, it is hypothesized that Sulf1 may bind to a constituent of the ECM in close proximity to the expressing cells (Kleinschmit, 2010).

It was found that small Sulf1 clones show more severe phenotypes than large clones and Sulf1 homozygous mutant discs. This observation suggests that morphogen gradients are more severely disrupted in a developmental field with discontinuity of cell surface HS structures (e.g. discs with Sulf1 small clones) compared to one where HS sulfation is uniformly altered (e.g. Sulf1 homozygous mutant discs). However, the molecular mechanism behind this difference remains to be elucidated (Kleinschmit, 2010).

In situ hybridization showed that Sulf1 mRNA is expressed at high levels near both the AP and DV borders of the wing disc. Interestingly, this feature is similar to the expression pattern of dally in the wing disc. The DV boundary expression of dally is induced by Wg signaling. It ws shown that expression of Sulf1, like that of dally, is induced by Wg signaling. Thus, Sulf1, a negative regulator of the Wg pathway, participates in a negative feedback loop within this morphogen system. It has been previously shown that Dally is a component of the negative feedback loop for the Dpp signaling pathway, potentially stabilizing the shape of the Dpp gradient. In addition, Wg also induces Notum, which is a secreted antagonist of Wg and functions through the posttranslational cleavage of glypicans at the wing margin. Collectively, these results implicate HSPGs and HS biosynthetic machinery components as general constituents of morphogen feedback systems, supporting the stability and the robustness of morphogen signaling gradients (Kleinschmit, 2010).

The role of glypicans in Wnt inhibitory factor-1 activity and the structural basis of Wif1's effects on Wnt and Hedgehog signaling

Proper assignment of cellular fates relies on correct interpretation of Wnt and Hedgehog (Hh) signals. Members of the Wnt Inhibitory Factor-1 (WIF1) family are secreted modulators of these extracellular signaling pathways. Vertebrate WIF1 binds Wnts and inhibits their signaling, but its Drosophila melanogaster ortholog Shifted (Shf) binds Hh and extends the range of Hh activity in the developing wing. Shf activity is thought to depend on reinforcing interactions between Hh and glypican HSPGs. Using zebrafish embryos and the heterologous system provided by D. melanogaster wing, this study reports on the contribution of glypican HSPGs to the Wnt-inhibiting activity of zebrafish Wif1 and on the protein domains responsible for the differences in Wif1 and Shf specificity. Wif1 strengthens interactions between Wnt and glypicans, modulating the biphasic action of glypicans towards Wnt inhibition; conversely, glypicans and the glypican-binding 'EGF-like' domains of Wif1 are required for Wif1's full Wnt-inhibiting activity. Chimeric constructs between Wif1 and Shf were used to investigate their specificities for Wnt and Hh signaling. Full Wnt inhibition required the 'WIF' domain of Wif1, and the HSPG-binding EGF-like domains of either Wif1 or Shf. Full promotion of Hh signaling requires both the EGF-like domains of Shf and the WIF domains of either Wif1 or Shf. That the Wif1 WIF domain can increase the Hh promoting activity of Shf's EGF domains suggests it is capable of interacting with Hh. In fact, full-length Wif1 affected distribution and signaling of Hh in D. melanogaster, albeit weakly, suggesting a possible role for Wif1 as a modulator of vertebrate Hh signaling (Avanesov, 2012; full text of article). The 'WIF' domain of WIF1 does not bind HS sidechains, but is sufficient for Wnt binding; the 'EGF-like' domains show only weak binding to Wnts on their own, but appear to strengthen Wnt binding to the 'WIF' domain. But while the Drososphila WIF1 homolog Shf contains both 'WIF' and 'EGF-like' domains, it does not inhibit Wg signaling; instead, it increases the levels or range of Hh signaling. This study found that a construct containing Shf's 'WIF' domain and the zebrafish Wif1's 'EGF-like' domains also cannot inhibit Wnt signaling, while the reciprocal construct with Wif1's 'WIF' domain and Shf's 'EGF-like' domain can. Similar results have been obtained with constructs made from Shifted and human WIF1. Thus, the ability to inhibit Wg activity, and likely to bind significant levels of Wg, resides in the different 'WIF' domains of Wif1 and Shf (Avanesov, 2012).

Surprisingly, Shf did show a weak ability to improve Wg signaling in sensitized backgrounds expressing either Wif1 or the dominant negative DFz2-GPI construct. While no obvious effect was ever detected of Shf on ex-Wg levels, it may weakly interact with Wg in a manner that reduces the levels bound to Wif1 or DFz2-GPI and increases the levels available for the Wg receptors. Consistent with this interpretation, UAS-shf did not alleviate margin defects caused by expression of UAS-wg RNAi, even though UAS-Dfz2-GPI and UAS-wg RNAi show a very comparable impact on Wg activity. Alternatively, Shf's effect on Wnt signaling might be due to interactions with the Wnt4 or Wnt6 expressed along the wing margin, which may have redundant roles in wing margin development that are only obvious in a sensitized background. Indirect effects via Hh signaling are unlikely, as Shf overexpression does not further increase Hh signaling (Avanesov, 2012).

The situation with Hh signaling is more complex. First, vertebrate WIF1's are not known to regulate vertebrate Hh signaling, but this study found that zebrafish Wif1 can weakly affect the reduced movement or accumulation of Hh normally observed in shf mutant wing discs. The Hh-GFP accumulation is abnormal, however, appearing more punctuate than in normal wing discs, perhaps accounting for its ability to reduce the expression of Hh targets (Avanesov, 2012).

Placing WIF domain of zebrafish Wif1 in the context of Shf's 'EGF-like' domains in a chimeric WIFWif1-EGFShf construct almost fully rescues loss of shf function, something not observed after expression of the Shf 'EGF-like' domains alone. Together, these data suggest that the 'WIF' domains of both Shf and zebrafish Wif1 are capable of interacting with Hh. Like Wnts, Hh is palmitoylated, and it has been suggested that these palmitates might bind a hydrophobic pocket found in the WIF domain, although this has been recently questioned. The activity of 'WIF' domains in Hh signaling may also vary between different vertebrates, since unlike the WIFWif1-EGFShf construct made using zebrafish 'WIF' domains, a similar construct made using the 'WIF' domain from human WIF1 does not rescue loss of shf function (Avanesov, 2012).

The Shf 'EGF-like' domains are necessary to confer a Shf-like level of Hh-promoting activity to the 'WIF' domains of zebrafish Wif1. The Hh-promoting activity of Wif1's 'WIF' domain is increased by placing it in the context of Shf's 'EGF-like' domains, and the low Hh-promoting activity of Shf's 'WIF' domain is not changed by placing it in the context of Wif1's 'EGF-like' domains. It is unlikely that the 'EGF-like' domains of Shf and Wif1 differ significantly in their HSPG-binding activities, since Wif1 and WIFWif1-EGFShf differ only slightly in their ability to inhibit Wnt signaling and interact genetically with Dlp. Therefore the alternative hypothesis is favored that Shf's 'EGF-like' domains contribute to Hh signaling through a mechanism independent of glypican binding. While the Shf 'EGF-like' domains alone (ShfδWIF) cannot increase Hh signaling, it was found that they can increase the levels of extracellular Hh, suggesting that they contribute to Hh binding, much as the 'EGF-like' domains of WIF1 do to Wnt binding (Avanesov, 2012).

Since Wif1 can alter Hh distribution and, more weakly, signaling in Drosophila, an important question is whether it can also do so in vertebrates. Because of its strong effects on Wnt signaling, vertebrate WIF1 family proteins have rarely been assayed for their effects on other pathways, so a weak modulation of one of the vertebrate Hhs remains a possibility (Avanesov, 2012).

The endocytic pathway and formation of the Wingless morphogen gradient

Controlling the spread of morphogens is crucial for pattern formation during development. In the Drosophila wing disc, Wingless secreted at the dorsal-ventral compartment boundary forms a concentration gradient in receiving tissue, where it activates short- and long-range target genes. The glypican Dally-like promotes Wingless spreading by unknown mechanisms, while Dynamin-dependent endocytosis is thought to restrict Wingless spread. Short-term expression of dominant negative Rab proteins was used to examine the polarity of endocytic trafficking of Wingless and its receptors and to determine the relative contributions of endocytosis, degradation and recycling to the establishment of the Wingless gradient. The results show that Wingless is internalized via two spatially distinct routes: one on the apical, and one on the basal, side of the disc. Both restrict the spread of Wingless, with little contribution from subsequent degradation or recycling. As has been shown for Frizzled receptors, depleting Arrow does not prevent Wingless from entering endosomes. Both Frizzled and Arrow are internalized mainly from the apical membrane. Thus, the basal Wingless internalization route must be independent of these proteins. Dally-like is not required for Wingless spread when endocytosis is blocked, and it is proposed that Dally-like promotes the spread of Wingless by directing it to lateral membranes, where its endocytosis is less efficient. Thus, subcellular localization of Wingless along the apical-basal axis of receiving cells may be instrumental in shaping the Wingless gradient (Marois, 2006).

The data suggest that the spread of Wg is controlled by restrictive clearance. Preventing Rab5-dependent internalization increases the spread of Wg through disc tissue. While this may also elevate extracellular Wg levels by indirect mechanisms (for example by modulating extracellular Wg proteolysis or release from disc tissue), the possibility is favored that the gradient is shaped by internalization of Wg itself, for several reasons. First, Wg is actually found in Rab5- and Rab7-positive endosomes. Second, treating living discs with protease inhibitors does not cause Wg to accumulate. Finally, differential centrifugation experiments suggest that only about 6% of Wg is not membrane-associated (Marois, 2006).

Some ligands signal from endocytic compartments after internalization. Therefore, it was initially of interest to discover whether the high levels of Wg accumulating after dominant negative Rab5SN expression could increase signal transduction. While Rab5SN-expressing cells both accumulate Armadillo and reduce Senseless expression, whether internalization of Wg is required for signaling was not further investigated because of the striking and unexpected transcriptional changes caused by blocking Rab5 activity. For example, transcription of both Fz2 and Dlp increases and that of Arrow plummets within a few hours of initiating Rab5SN expression - any of these changes by themselves could alter Wg signaling. Although this phenomenon are not yet understood, one might imagine that coupling transcriptional regulation of endocytic receptors to their actual endocytosis and/or degradation would be an effective homeostatic mechanism (Marois, 2006).

Studies in tissue culture cells have shown that inhibiting Rab7-dependent lysosomal degradation can increase recycling of some proteins to the cell surface. In imaginal discs, Rab7TN expression increases the abundance of Rab11 recycling endosomes, and inhibiting recycling via Rab11SN enlarges the Rab7-positive degradative compartment. Thus, the recycling and degradative pathways may compete for some cargo in discs as they do in cultured cells. This raises the possibility that changing the balance of degradation and recycling might affect the pool of extracellular Wingless available for spreading. Indeed, Wg appears to be recycled in embryos, and inhibiting lysosomal degradation in the embryonic ectoderm extends its range. However, the data suggest that the increase in the range of Wg caused by inhibiting degradation (at least in imaginal discs) is not the result of increased recycling and extracellular spread. Although Wg protein is detected in endosomes over a broader range in imaginal disc tissue expressing Rab7TN, there is no increase in the range of extracellular Wg. Furthermore, neither extracellular nor intracellular Wg distribution is affected by inhibiting the Rab4- or Rab11-dependent recycling pathways. Inhibiting degradation probably extends the apparent range of Wg by stabilizing internalized protein, raising its levels above the threshold of detectability in more distant cells (Marois, 2006).

The idea that apical-basal polarity of epithelial cells might play a role in regulating morphogen trafficking has been suggested by the observation that wg mRNA is enriched apically in the embryonic ectoderm. Changing mRNA localization alters the distribution of Wg protein in receiving tissue, raising the intriguing possibility that Wg might be trafficked differently depending on whether it is secreted apically or basal-laterally. In support of this idea, the data show that Wingless is internalized specifically from the apical and basal (but not lateral) surfaces of the disc epithelium. Indeed, the distribution of Rab5- and Rab7-positive endosomes in general suggests that the apical and basal surfaces are more endocytically active than other regions. The apical and basal internalization mechanisms may be distinct; the known receptors for Wingless, Fz2 and Arrow are internalized mainly from the apical surface (despite their steady-state basal-lateral localization), suggesting that basal Wingless endocytosis must be independent of these proteins. One possibility is that membrane association of Wg via Palmitate is sufficient to allow its endocytosis -- perhaps by mechanisms similar to those used by gpi-linked proteins. Alternatively, Wg bound to Lipoprotein particles might be internalized via Lipoprotein receptors (Marois, 2006).

It was surprising to observe that the Wg accumulating on the basal side of disc epithelial cells after Rab5SN expression does not spread onto the lateral membrane, since no barrier to diffusion has been identified between these domains. Three possible explanations are suggested. (1) There may indeed be a 'fence' separating the lateral and basal sides of disc epithelial cells. Neurons have a fence that prevents diffusion of lipid and lipid-linked proteins between the axon and the cell body, although it does not resemble a classical intercellular junction. (2) It may be that the receptor(s) that normally internalize Wg basally are linked to cytoskeletal components, or interact with extracellular matrix (ECM), and are not free to diffuse. If these Wg receptors were of sufficiently high affinity, they might trap Wg before it could move laterally. (3) Perhaps Wg itself interacts efficiently with basal ECM components (Marois, 2006).

While it seems that endocytosis restricts the spread of Wg on the apical and basal surfaces, it is not yet clear which receptors might be responsible. A simple model would predict that removing such a receptor should produce phenotypes similar to Rab5SN expression, i.e. increased and more extensive extracellular Wg, and less Wg in endosomes. Conversely, overexpression might be expected to compress the range of Wg distribution and decrease extracellular Wg. None of the known receptors behaves in this way. Previous studies showed that at least a fraction of Wg is still internalized in the absence of both Fz1 and Fz2. Furthermore, overexpression of Fz2 causes extracellular Wg accumulation over longer distances. This study has shown that loss of Arrow actually increases the amount of Wg present in Rab5- and Rab7-positive endosomes - more consistent with a role in Wg degradation after endocytosis. The complexity of these observations may reflect different mechanisms of Wg endocytosis on the apical and basal sides of the cell - understanding both these pathways and their interplay will be necessary to understanding how the Wg gradient forms (Marois, 2006).

While internalization limits the range of Wg accumulation, the glypican Dlp extends it. It has been proposed that Dlp allows Wg to interact with disc cells, increasing local Wg concentration and restricting its diffusion to the epithelium. The data, however, show that disc cells can accumulate high levels of Wg on their surface in the absence of Dlp as long as Rab5-dependent internalization is blocked. This observation is not consistent with a model in which Dlp traps Wg on the cell surface or helps it transfer from cell to cell. Instead, it suggests that Dlp normally stabilizes Wg at the cell surface by antagonizing the effects of Rab5-dependent internalization. While Wg is normally internalized from the apical- and basal-most surfaces of disc cells, Dlp overexpression recruits Wg to the lateral cell surface. This raises the possibility that Dlp stabilizes Wg and increases its range by changing its subcellular localization to protect it from endocytosis. Polarizing the distribution of morphogens within an epithelium may have a key regulatory role in the trafficking events leading to gradient formation (Marois, 2006).

Cellular trafficking of the glypican Dally-like is required for full-strength Hedgehog signaling and Wingless transcytosis

Hedgehog (Hh) and Wingless (Wg) morphogens specify cell fate in a concentration-dependent manner in the Drosophila wing imaginal disc. Proteoglycans, components of the extracellular matrix, are involved in Hh and Wg stability, spreading, and reception. This study demonstrates that the glycosyl-phosphatidyl-inositol (GPI) anchor of the glypican Dally-like (Dlp) is required for its apical internalization and its subsequent targeting to the basolateral compartment of the epithelium. Dlp endocytosis from the apical surface of Hh-receiving cells catalyzes the internalization of Hh bound to its receptor Patched (Ptc). The cointernalization of Dlp with the Hh/Ptc complex is dynamin dependent and necessary for full-strength Hh signaling. Wg is secreted apically in the disc epithelium and apicobasal trafficking of Dlp allows Wg transcytosis to favor Wg spreading along the basolateral compartment. Thus, Dlp endocytosis is a common regulatory mechanism of both Hh and Wg morphogen action (Gallet, 2008).

Previous studies failed to observe Dlp at the apical surface of the wing disc epithelium despite the fact that it has been extensively shown in vertebrates that GPI-linked proteins are mainly targeted to the apical surface of epithelial cells. By performing extracellular labeling and kinetic experiments, this study demonstrated that Dlp is targeted to the apical surface before being endocytosed and readdressed to the basolateral compartment. Blocking endocytosis allowed demonstration that apical surface accumulation takes place at the expense of its basolateral location, showing that Dlp is first targeted to the apical domain of the epithelium before being sent to the basolateral compartment. The GPI anchor of Dlp is essential for its internalization, as Dlp tethering by a transmembrane domain (e.g., GFP-Dlp-CD2) abolishes its capacity to be internalized. It is also concluded that Dlp apical targeting followed by its rapid internalization is essential for full Hh signal transduction and for shaping the Wg gradient because when Dlp internalization is blocked, both processes are impaired (Gallet, 2008).

Interestingly, a rescue was observed of the first row of dlp mutant cells by the wild-type surrounding cells. Two mechanisms could explain this: (1) GPI-linked proteins are inserted in the outer leaflet of the plasma membrane and are exposed to the extracellular space. Dlp could interact with extracellular proteins located in nearby cells that could aid in flipping Dlp from the outer leaflet of the plasma membrane of one cell to the next, and/or (2) Dlp might be carried by argosomes, which are large extracellular particles resembling low-density lipoproteins containing lipophorins, esterified cholesterol, and triglycerides surrounded by a phospholipid monolayer. These argosomes not only bear GPI-linked proteins such as HSPGs but also morphogens such as Wg or Hh and are able to travel over several rows of cells (Gallet, 2008).

Whereas numerous data have clearly demonstrated the role of the HSPG in Hh signaling through regulation of Hh stabilization, movement, and reception, the function of Dlp in Hh signaling remained unclear. This study clearly demonstrates that Dlp is exclusively necessary in Hh-receiving cells for full-strength Hh signaling. Strong colocalization of Ptc, Hh, and Dlp is observed in endocytic vesicles in Hh-receiving cells. Ptc is probably responsible for the Hh/Dlp internalization, because it was not possible to detect Dlp-Hh-containing endocytic vesicles in the absence of Ptc. Overexpressing Dlp increased the number of Ptc-Hh-internalized vesicles, whereas absence or tethering Dlp at the cell surface (e.g., GFP-Dlp-CD2) lowered the number of Ptc/Hh vesicles. Therefore, it can be imagined that the presence of Dlp increases Ptc-Hh internalization to elicit a high level of pathway activation. Dlp could also stabilize Ptc/Hh within the intracellular compartment, allowing a stronger and/or a longer signaling. The role of Dlp is clearly different from other Hh coreceptors. Indeed, already identified Hh coreceptors such as Ihog and Brother of Ihog (Boi) (Cdo and Boc, respectively, in vertebrates) stabilize Hh at the cell surface, whereas Dlp overexpression does not increase Hh binding on receiving cells in vivo or in vitro (Gallet, 2008).

It has been demonstrated that Hh is apically secreted by wing disc epithelial cells; however, the compartment by which target cells receive Hh signal (e.g., apical, lateral, or basal) remains controversial. Nevertheless, the data suggest that endocytosis from the apical surface is necessary to sustain full-strength Hh signaling because blocking endocytosis inhibits Dlp internalization from the apical surface and decreases Hh signaling activation. Moreover, a colocalization between Hh and extracellular Dlp is observed at the apical side of Hh-receiving cells but not at the basolateral part of the cell. It has also showed previously that in ptc mutant embryos, Hh accumulates at the apical side of receiving cells. Nevertheless, although blocking endocytosis strongly stabilized Dlp at the apical cell surface, it also stabilized Ptc at both poles of the epithelial cells, raising the possibility that part of the signal transduction occurs basally. Unfortunately, it was not possible to see any Hh accumulation at either pole of the receiving cells when endocytosis was blocked. Hence, a model is favored in which Hh can trigger its signal both basally and apically, where the apical signaling is amplified by Dlp and is necessary for a full-strength Hh signal (Gallet, 2008).

Blocking endocytosis impairs Hh signaling both in embryos and in wing discs. However, it has been previously published that inhibiting endocytosis in imaginal discs using a thermosensitive allele of shi (shi^ts) does not impaired Hh signaling: Ci and collier (an Hh target gene) are unaffected but Ptc is stabilized. How can the differences with these data be explained? It is important to note that the shi null allele is cell lethal; therefore, either a dominant-negative or a shi^ts allele, which may not fully inhibit endocytosis, must be used. Accordingly, Ptc stabilization is observed in only 30% of discs, but each time such a stabilization is observed, a decrease of dpp expression was observed. Increasing the level of Shi^DN expression increased the penetrance of the phenotype but triggered lethality, making analysis difficult (Gallet, 2008).

Wg is mainly secreted via the apical pole of producing cells. Strikingly, in those cells, Wg is strongly localized in endocytic vesicles that are abundant apically but also in multivesicular endosomes. Dlp overexpression decreases the level of Wg at the apical surface of cells while increasing Wg stability along their lateral compartment. Therefore, it is proposed that Wg is secreted apically and is then endocytosed with the help of Dlp. Once internalized, Dlp targets Wg by transcytosis to the lateral compartment, where it is stabilized and can spread farther away to activate long-range target genes (Gallet, 2008).

Intriguingly, Dlp seems to play antagonistic roles in Wg signaling. Although it inhibits Wg activity near the Wg source, it is also necessary for Wg pathway activation far from the Wg source. This functional duality is directly related to its pattern of expression. Indeed, Dlp is expressed at low levels along the Wg source (e.g., the D/V axis) owing to repression by the Wg pathway itself, whereas it is expressed at higher levels far from the Wg source. The results could explain how Dlp functions antagonistically in Wg signaling. The two Wg receptors DFrizzled2 (Dfz2) and Arrow are involved in both the signal transduction and the internalization of Wg, mainly through the apical surface to shape its gradient by targeting Wg to the lysosome. Therefore, it is proposed that low Dlp-dependent transcytosis of Wg in producing cells and neighboring ones allows a high level of Wg at the apical surface and hence a strong activation of the pathway. On the contrary, higher Dlp-dependent transcytosis of Wg in more distant cells reduces both Wg pathway activation and degradation, promoting Wg movement along the basolateral compartment. Accordingly, it is observed that, in absence of dlp, extracellular Wg is absent from the lateral compartment of distant cells (Gallet, 2008).

This model supposes that Dlp is able to internalize Wg independently of the receptors DFz2 and Arrow. Interestingly, several groups found some internalized Wg in the absence of the Wg receptors Dfz2 and Arrow. Moreover, Dlp overexpression stabilizes Wg at the cell surface at the expense of short-range signaling activity, in accord with the fact that Wg must be endocytosed by Dfz2/Arrow to promote strong signaling. The results further support the view that Wg may form two different complexes: on the one hand a Dlp-Wg complex involved in Wg transcytosis and stabilization, and on the other hand a DFz2-Arrow-Wg signaling complex that shapes the Wg morphogen gradient and signals. Interestingly, when GFP-Dlp-CD2 is overexpressed, although Wg is stabilized at the apical surface over a very long range, where it should activate its pathway, a much stronger inhibition of Wg signaling is observed and an absence of internalized Wg. Therefore, GFP-Dlp-CD2 may titrate Wg from its receptors and prevent internalization, giving rise to both the stabilization of Wg and the inhibition of the pathway. Under physiological conditions, Dlp targeting of Wg to the lateral compartment supports its stabilization and spreading at the expense of its internalization/degradation from the apical surface by its receptors (Gallet, 2008).

A targeted glycan-related gene screen reveals heparan sulfate proteoglycan sulfation regulates WNT and BMP trans-synaptic signaling

A Drosophila transgenic RNAi screen targeting the glycan genome, including all N/O/GAG-glycan biosynthesis/modification enzymes and glycan-binding lectins, was conducted to discover novel glycan functions in synaptogenesis. As proof-of-product,functionally paired heparan sulfate (HS) 6-O-sulfotransferase (hs6st) and sulfatase (sulf1), which bidirectionally control HS proteoglycan (HSPG) sulfation, were characterized. RNAi knockdown of hs6st and sulf1 causes opposite effects on functional synapse development, with decreased (hs6st) and increased (sulf1) neurotransmission strength confirmed in null mutants. HSPG co-receptors for WNT and BMP intercellular signaling, Dally-like Protein and Syndecan, are differentially misregulated in the synaptomatrix of these mutants. Consistently, hs6st and sulf1 nulls differentially elevate both WNT (Wingless; Wg) and BMP (Glass Bottom Boat; Gbb) ligand abundance in the synaptomatrix. Anterograde Wg signaling via Wg receptor dFrizzled2 C-terminus nuclear import and retrograde Gbb signaling via synaptic MAD phosphorylation and nuclear import are differentially activated in hs6st and sulf1 mutants. Consequently, transcriptional control of presynaptic glutamate release machinery and postsynaptic glutamate receptors is bidirectionally altered in hs6st and sulf1 mutants, explaining the bidirectional change in synaptic functional strength. Genetic correction of the altered WNT/BMP signaling restores normal synaptic development in both mutant conditions, proving that altered trans-synaptic signaling causes functional differentiation defects (Dani, 2012).

It is well known that synaptic interfaces harbor heavily-glycosylated membrane proteins, glycolipids and ECM molecules, but understanding of glycan-mediated mechanisms within this synaptomatrix is limited. A genomic screen aimed to systematically interrogate glycan roles in both structural and functional development in the genetically-tractable Drosophila NMJ synapse. 130 candidate genes were screened, classified into 8 functional families: N-glycan biosynthesis, O-glycan biosynthesis, GAG biosynthesis, glycoprotein/proteoglycan core proteins, glycan modifying/degrading enzymes, glycosyltransferases, sugar transporters and glycan-binding lectins. From this screen, 103 RNAi knockdown conditions were larval viable, whereas 27 others produced early developmental lethality. 35 genes had statistically significant effects on different measures of morphological development: 27 RNAi-mediated knockdowns increased synaptic bouton number, 9 affected synapse area (2 increased, 7 decreased) and 2 genes increased synaptic branch number. These data suggest that overall glycan mechanisms predominantly serve to limit synaptic morphogenesis. 13 genes had significant effects on the functional differentiation of the synapse, with 12 increasing transmission strength and only 1 decreasing function upon RNAi knockdown. Thus, glycan-mediated mechanisms also predominantly limit synaptic functional development. A very small fraction of tested genes (CG1597; pgant35A, CG7480; veg, CG6657; hs6st, CG4451; sulf1, CG6725 and CG11874) had effects on both morphology and function. A large percentage of genes (~30%) showed morphological defects with no corresponding effect on function, while only 7% of genes showed functional alterations without morphological defects, and <5% of all genes affect both. These results suggest that glycans have clearly separable roles in modulating morphological and functional development of the NMJ synapse (Dani, 2012).

A growing list of neurological disorders linked to the synapse are attributed to dysfunctional glycan mechanisms, including muscular dystrophies, cognitive impairment and autism spectrum disorders. Drosophila homologs of glycosylation genes implicated in neural disease states include ALG3 (CG4084), ALG6 (CG5091), DPM1 (CG10166), FUCT1 (CG9620), GCS1 (CG1597), MGAT2 (CG7921), MPDU1 (CG3792), PMI (CG33718) and PPM2 (CG12151). Two of these genes, Gfr (CG9620) and CG1597, showed synaptic morphology phenotypes in the RNAi screen. Given that connectivity defects are clearly implicated in cognitive impairment and autism spectrum disorders, it would be of interest to explore the glycan mechanism affecting synapse morphology in Drosophila models of these disease states. Glycans are well known to modulate extracellular signaling, including ligands of integrin receptors, to regulate intercellular communication. In the genetic screen, several O-glycosyltransferases mediating this mechanism were identified to show morphological (GalNAc-T2, CG6394; pgant35A, CG7480, O-fut2, CG14789; rumi, CG31152) and functional (pgant5, CG31651; pgant35A, CG7480) synaptic defects upon RNAi knockdown. These findings suggest that known integrin-mediated signaling pathways controlling NMJ synaptic structural and functional development are modulated by glycan mechanisms. The screen showed CG6657 RNAi knockdown affects functional differentiation, consistent with reports that this gene regulates peripheral nervous system development. The corroboration of the screen results with published reports underscores the utility of RNAi-mediated screening to identify glycan mechanisms, and supports use of the screen results for bioinformatic/meta-analysis to link observed phenotypes to neurophysiological/pathological disease states and to direct future glycan mechanism studies at the synapse (Dani, 2012).

From this screen, the two functionally-paired genes sulf1 and hs6st were selected for further characterization. As in the RNAi screen, null alleles of these two genes had opposite effects on synaptic functional differentiation but similar effects on synapse morphogenesis, validating the corresponding screen results. The two gene products have functionally-paired roles; Hs6st is a heparan sulfate (HS) 6-O-sulfotransferase, and Sulf1 is a HS 6-O-endosulfatase. These activities control sulfation of the same C₆ on the repeated glucosamine moiety in HS GAG chains found on heparan sulfate proteoglycans (HSPGs). At the Drosophila NMJ, two HSPGs are known to regulate synapse assembly; the GPI-anchored glypican Dally-like protein (Dlp), and the transmembrane Syndecan (Sdc). In contrast, the secreted HSPG Perlecan (Trol) is not detectably enriched at the NMJ, and indeed appears to be selectively excluded from the perisynaptic domain. In other developmental contexts, the membrane HSPGs Dlp and Sdc are known to act as co-receptors for WNT and BMP ligands, regulating ligand abundance, presentation to cognate receptors and therefore signaling. Importantly, the regulation of HSPG co-receptor abundance has been shown to be dependent on sulfation state mediated by extracellular sulfatases. Consistently, upregulation of Dlp and Sdc was observed in sulf1 null synapses, whereas Dlp was reduced in hs6st null synapses. In the developing Drosophila wing disc, HSPG co-receptors increase levels of the Wg ligand due to extracellular stabilization, and the primary function of Dlp in this developmental context is to retain Wg at the cell surface. Likewise, in developing Drosophila embryos, a significant fraction of Wg ligand is retained on the cell surfaces in a HSPG-dependent manner, with the HSPG acting as an extracellular co-receptor. Syndecan also modulates ligand-dependent activation of cell-surface receptors by acting as a co-receptor. At the NMJ, regulation of both these HSPG co-receptors occurs in the closely juxtaposed region between presynaptic bouton and muscle subsynaptic reticulum, in the exact same extracellular space traversed by the secreted trans-synaptic Wg and Gbb signals. It is therefore proposed that altered Dlp and Sdc HSPG co-receptors in sulf1 and hs6st mutants differentially trap/stabilize Wg and Gbb trans-synaptic signals at the interface between motor neuron and muscle, to modulate the extent and efficacy of intercellular signaling driving synaptic development (Dani, 2012).

HS sulfation modification is linked to modulating the intercellular signaling driving neuronal differentiation . In particular, WNT and BMP ligands are both regulated via HS sulfation of their extracellular co-receptors, and both signals have multiple functions directing neuronal differentiation, including synaptogenesis. In the Drosophila wing disc, extracellular WNT (Wg) ligand abundance and distribution was recently shown to be strongly elevated in sulf1 null mutants. Moreover, sulf1 has also recently been shown to modulate BMP signaling in other cellular contexts. Consistently, this study has shown increased WNT Wg and the BMP Gbb abundance and distribution in sulf1 null NMJ synapses. The hs6st null also exhibits elevated Wg and Gbb at the synaptic interface, albeit the increase is lower and results in differential signaling consequences. In support of this contrasting effect, extracellular signaling ligands are known to bind HSPG HS chains differentially dependent on specific sulfation patterns. It is important to note that the sulf1 and hs6st modulation of trans-synaptic signals is not universal, as Jelly Belly (Jeb) ligand abundance and distribution was not altered in the sulf1 and hs6st null conditions. This indicates that discrete classes of secreted trans-synaptic molecules are modulated by distinct glycan mechanisms to control NMJ structure and function (Dani, 2012).

At the Drosophila NMJ, Wg is very well characterized as an anterograde trans-synaptic signal and Gbb is very well characterized as a retrograde trans-synaptic signal. In Wg signaling, the dFz2 receptor is internalized upon Wg binding and then cleaved so that the dFz2-C fragment is imported into muscle nuclei. In hs6st nulls, increased Wg ligand abundance at the synaptic terminal corresponds to an increase in dFz2C punctae in muscle nuclei as expected. In contrast, the increase in Wg at the sulf1 null synapse did not correspond to an increase in the dFz2C-terminus nuclear internalization, but rather a significant decrease. One explanation for this apparent discrepancy is the 'exchange factor' model based on the biphasic ability of the HSPG co-receptor Dlp to modulate Wg signaling. In the Drosophila wing disc, this model suggests that the transition of Dlp co-receptor from an activator to repressor of signaling depends on Wg cognate receptor dFz2 levels, such that a low ratio of Dlp:dFz2 potentiates Wg-dFz2 interaction, whereas a high ratio of Dlp:dFz2 prevents dFz2 from capturing Wg. In sulf1 null synapses, a very great increase was observed in Dlp abundance (~40% elevated) with no significant change in the dFz2 receptor. In contrast, at hs6st null synapses there is a decrease in Dlp abundance (15% decreased) together with a significant increase in dFz2 receptor abundance (~25% elevated). Thus, the higher Dlp:dFz2 ratio in sulf1 nulls could explain the decrease in Wg signal activation, evidenced by decreased dFz2-C terminus import into the muscle nucleus. In contrast, the Dlp:Fz2 ratio in hs6st is much lower, supporting activation of the dFz2-C terminus nuclear internalization pathway. This previously proposed competitive binding mechanism dependent on Dlp co-receptor and dFz2 receptor ratios predicts the observed synaptic Wg signaling pathway modulation in sulf1 and hs6st dependent manner (Dani, 2012).

At the Drosophila NMJ, Gbb is very well characterized as a retrograde trans-synaptic signal, with muscle-derived Gbb causing the receptor complex Wishful thinking (Wit), Thickveins (Tkv) and Saxaphone (Sax) to induce phosphorylation of the transcription factor mothers against Mothers against decapentaplegic (P-Mad). Mutation of Gbb ligand, receptors or regulators of this pathway have shown that Gbb-mediated retrograde signaling is required for proper synaptic differentiation and functional development. Further, loss of Gbb signaling results in significantly decreased levels of P-Mad in the motor neurons. This study shows that accumulation of Gbb in sulf1 and hs6st null synapses causes elevated P-Mad signaling at the synapse and P-Mad accumulation in motor neuron nuclei. Importantly, sulf1 null synapses show a significantly higher level of P-Mad signaling compared to hs6st null synapses, and this same change is proportionally found in P-Mad accumulation within the motor neuron nuclei. These findings indicate differential activation of Gbb trans-synaptic signaling dependent on the HS sulfation state is controlled by the sulf1 and hs6st mechanism, similar to the differential effect observed on Wg trans-synaptic signaling. Genetic interaction studies show that these differential effects on trans-synaptic signaling have functional consequences, and exert a causative action on the observed bi-directional functional differentiation phenotypes in sulf1 and hs6st nulls. Genetic correction of Wg and Gbb defects in the sulf1 null background restores elevated transmission back to control levels. Similarly, genetic correction of Wg and Gbb in hs6st nulls restores the decreased transmission strength back to control levels. These results demonstrate that the Wg and Gbb trans-synaptic signaling pathways are differentially regulated and, in combination, induce opposite effects on synaptic differentiation (Dani, 2012).

Both wg and gbb pathway mutants display disorganized and mislocalized presynaptic components at the active zone (e.g. Bruchpilot; Brp) and postsynaptic components including glutamate receptors (e.g. Bad reception; Brec/GluRIID). Consistently, the bi-directional effects on neurotransmission strength in sulf1 and hs6st mutants are paralleled by dysregulation of these same synaptic components. Changes in presynaptic Brp and postsynaptic GluR abundance/distribution causally explain the bi-directional effects on synaptic functional strength between sulf1 and hs6st null mutant states. Alterations in active zone Brp and postsynaptic GluRs also agree with assessment of spontaneous synaptic activity. Null sulf1 and hs6st synapses showed opposite effects on miniature evoked junctional current (mEJC) frequency (presynaptic component) and amplitude (postsynaptic component). Further, quantal content measurements also support the observation of bidirectional synaptic function in the two functionally paired nulls. Genetic correction of Wg and Gbb defects in both sulf1 and hs6st nulls restores the molecular composition of the pre- and postsynaptic compartments back to wildtype levels. When both trans-synaptic signaling pathways are considered together, these data suggest that HSPG sulfate modification under the control of functionally-paired sulf1 and hs6st jointly regulates both WNT and BMP trans-synaptic signaling pathways in a differential manner to modulate synaptic functional development on both sides of the cleft (Dani, 2012).

This paper has presented the first systematic investigation of glycan roles in the modulation of synaptic structural and functional development. A host of glycan-related genes were identified that are important for modulating neuromuscular synaptogenesis, and these genes are now available for future investigations, to determine mechanistic requirements at the synapse, and to explore links to neurological disorders. As proof for the utilization of these screen results, this study has identified extracellular heparan sulfate modification as a critical platform of the intersection for two secreted trans-synaptic signals, and differential control of their downstream signaling pathways that drive synaptic development. Other trans-synaptic signaling pathways are independent and unaffected by this mechanism, although it is of course possible that a larger assortment of signals could be modulated by this or similar mechanisms. This study supports the core hypothesis that the extracellular space of the synaptic interface, the heavily-glycosylated synaptomatrix, forms a domain where glycans coordinately mediate regulation of trans-synaptic pathways to modulate synaptogenesis and subsequent functional maturation (Dani, 2012).

Two matrix metalloproteinase classes reciprocally regulate synaptogenesis

Synaptogenesis requires orchestrated intercellular communication between synaptic partners, with trans-synaptic signals necessarily traversing the extracellular synaptomatrix separating presynaptic and postsynaptic cells. Extracellular matrix metalloproteinases (Mmps) regulated by secreted tissue inhibitors of metalloproteinases (Timps), cleave secreted and membrane-associated targets to sculpt the extracellular environment and modulate intercellular signaling. This study tested Mmp roles at the neuromuscular junction (NMJ) model synapse in the reductionist Drosophila system, which contains just two Mmps (secreted Mmp1 and GPI-anchored Mmp2) and one secreted Timp. All three matrix metalloproteome components co-dependently localize in the synaptomatrix. Both Mmp1 and Mmp2 independently restrict synapse morphogenesis and functional differentiation. Surprisingly, either dual knockdown or simultaneous inhibition of the two Mmp classes together restores normal synapse development, identifying a novel reciprocal suppression mechanism. The two Mmp classes co-regulate a Wnt trans-synaptic signaling pathway modulating structural and functional synaptogenesis, including the GPI-anchored heparan sulfate proteoglycan (HSPG) Wnt co-receptor Dally-like Protein (Dlp), cognate receptor Frizzled-2 and Wingless ligand. Loss of either Mmp1 or Mmp2 reciprocally misregulates Dlp at the synapse, with normal signaling restored by co-removal of both Mmp classes. Correcting Wnt co-receptor Dlp levels in both mmp mutants prevents structural and functional synaptogenic defects. Taken together, these results identify a novel Mmp mechanism that fine-tunes HSPG co-receptor function to modulate Wnt signaling to coordinate synapse structural and functional development (Dear, 2015).

A large number of Mmps are expressed in the mammalian nervous system, with roles in neurodevelopment, plasticity and neurological disease. Understanding how each Mmp individually and combinatorially functions is hindered by genetic redundancy and compensatory mechanisms. This study exploited the Drosophila system to analyze a matrix metalloproteome containing just one member of each conserved component: one secreted Mmp, one membrane-tethered Mmp and one Timp. Both Mmp classes were found to attenuate structural and functional synaptic development, with electrophysiological, ultrastructural and molecular roles in both presynaptic and postsynaptic cells. A surprising discovery is that the Mmp classes suppress each other's requirements at the synapse. From discrete activities to redundancy, cooperation and now reciprocal suppression, studies continue to reveal how Mmps interact to regulate developmental processes. This study shows that the two Mmp classes play separable yet interactive roles in sculpting NMJ development. During the writing of this manuscript, a genomic Mmp2 rescue line was produced (Wang, 2014), which will be critical in further testing this interactive mechanism. It will be interesting to determine whether the Mmp suppressive mechanism is used in other developmental contexts, other intercellular signaling pathways and in mammalian models. Mammalian Mmp9 regulates synapse architecture and also postsynaptic glutamate receptor expression and/or localization. Likewise, mammalian Mmp7 regulates both presynaptic properties and postsynaptic glutamate receptor subunits. Thus, the dual roles of Mmps in pre- and postsynaptic compartments appear to be evolutionarily conserved (Dear, 2015).

Previous work demonstrated that Mmp1 and Mmp2 both regulate motor axon pathfinding in Drosophila embryos, albeit to different degrees and in this study, double Mmp mutants still exhibited defasciculated nerve bundles that separate prematurely. Consistently, both Mmp single mutants display excessive terminal axon branching at the postembryonic NMJ, but here the defect is fully alleviated by the removal of both Mmps. Other studies have either not identified, or not tested, a similar Mmp interaction, suggesting that reciprocal suppression might be specific to synaptogenesis. However, there are numerous reports that highlight the importance of Mmp and Timp balance. Mmp:Timp ratios can influence protease activation, localization, substrate specificity and Timp signaling and are commonly used as predictive clinical correlates in disease pathology. At the Drosophila NMJ, a similar reciprocal suppression interaction between pgant glycosyltransferases involved in O-linked glycosylation regulates synaptogenesis via integrin-tenascin trans-synaptic signaling. A recent study reported that pgant activity protects substrates from Furin-mediated proteolysis, which is a protease responsible for processing or activating Drosophila Mmp1 and Mmp2. Thus, Mmp proteolytic and glycan mechanisms could converge within the NMJ synaptomatrix to regulate trans-synaptic signaling (Dear, 2015).

New antibody tools produced in this study provide the means to interrogate an entire matrix metalloproteome, and will be important for testing Mmp and Timp functions throughout Drosophila. Many Mmps are both developmentally and activity regulated, with highly context-dependent functions. Future work will temporally dissect this mechanism at the developing NMJ and investigate how activity might regulate Mmp localization and function. It will be informative to correlate synaptogenic Mmp requirements with Mmp enzymatic activity by using in situ zymography assays, although non-enzymatic roles are certainly also possible. Lack of ultrastructure defects in Mmp mutant NMJs suggests that Drosophila Mmps have primarily instructive functions at the synapse, rather than broad proteolytic roles in ECM degradation. Consistently, Drosophila Mmp2 instructs motor axon pathfinding via a BMP intercellular signaling mechanism. Conversely, Mmp2 functions permissively in basement membrane degradation while shaping dendritic arbors. Because synaptic bouton size is reduced in mmp1 mutants, Mmp1 activity might degrade a prohibitive physical barrier at the NMJ. However, the results indicate a primary Mmp role in regulating intercellular signaling during synaptic development (Dear, 2015).

HSPG co-receptors of trans-synaptic ligands are key modulators of NMJ synaptogenesis and HSPGs are also established substrates of both mammalian and Drosophila Mmps. Mmp1 and Mmp2 differentially regulate the HSPG Dlp co-receptor to restrict the Wnt Wg trans-synaptic signaling driving structural and functional NMJ development. How might both increased and decreased levels of the Dlp co-receptor yield increased FNI pathway signal transduction? Regulation of Wnt signaling interactions ligands, co-receptors and receptors is managed at many levels. The 'Wg exchange factor model' provides a mechanistic framework for understanding the suppressive interactions of Mmp. In this mechanism, a low Dlp:Frz2 ratio helps the Frz2 receptor obtain more Wg, whereas a high Dlp:Frz2 ratio prevents Frz2 from capturing Wg as Dlp competes and sequesters Wg away from Frz2. Importantly, however, Dlp exhibits a context-dependent, bimodal role as both activator and repressor. Indeed, previous studies show these mechanisms are a key driving force in Wg signal transduction at the Drosophila NMJ (Dani, 2012; Friedman, 2013). In mmp1 mutants, Wg and Dlp are both reduced, resulting in a low Dlp:Frz2 ratio and elevated FNI. In mmp2 mutants, Dlp is spatially diffuse and Frz2 is increased, similarly resulting in a low Dlp:Frz2 ratio and elevated FNI. Balance is reset with Mmp co-removal because neither form of Mmp-induced HSPG tuning occurs. In this regard, it might be predicted that Dlp reduction in mmp2 mutants would only further increase FNI and therefore structural and functional defects. It is likely that absolute Dlp levels are the important driving factor in synaptogenesis and/or that Dlp exhibits bimodal functions in synaptic development (Dear, 2015).

Interestingly, a recent mouse study showed the Mmp3 hemopexin domain promotes Wnt signaling by inhibiting a negative Wnt regulator, raising the possibility that Mmps can act as molecular switches (or in feedback loops) dictating Wnt transduction. Another study suggests that Wnt signaling can directly mediate co-regulation of heparanase and Mmps. Indeed, both neural activity and intercellular signaling can stimulate Mmp-dependent ectodomain shedding of plasma membrane target proteins, thereby directly regulating the surface abundance of HSPGs and receptors, as well as other Mmps, which thus reciprocally modulate intra- and extracellular organization. From this model, the spatial arrangement of Dlp could be affected by co-regulated sheddase activity that is differentially altered in mmp1 and mmp2 mutants. Specifically, Mmp2 could shed Dlp, resulting in an increased area of Dlp expression in mmp2 mutants and loss of Mmp2 regulation by Mmp1 could result in aberrant Dlp restriction in mmp1 mutants, with Mmp co-removal remediating the Dlp domain thereby restoring normal Wnt trans-synaptic signaling. Future work will test the reciprocal impacts of Wnt signaling on Mmp expression and/or function in the context of synaptic development (Dear, 2015).

Emerging evidence suggests HSPG glycosaminoglycan (GAG) chains function as allosteric regulators of Mmps, with GAG content or composition influencing the localization and substrate specificity of Mmp. Indeed, Wg signaling is sensitive to perturbations in HSPG chain biosynthesis and HS modifying enzymes, which modulate both NMJ structure and function. It is easy to envision how tissue- and development-stage-specific HS modifications could coordinate HSPG/Mmp-dependent functions, thereby differentially regulating diverse signaling events, which enable context-specific responses instructed by the extracellular environment. Future work will examine how dual inputs of the HSPG co-receptor function and how Mmp proteolytic cleavage coordinates Wnt trans-synaptic signaling during synaptogenesis, particularly in the context of the Fragile X syndrome (FXS) disease model. Given that both loss or inhibition Mmp and correction of HSPG elevation independently alleviate synaptic defects in the FXS disease state, the overlapping mechanism provides an exciting avenue to therapeutic interventions for FXS and, potentially, related intellectual disability and autism spectrum disorders (Dear, 2015).

Internalization is required for proper Wingless signaling in Drosophila melanogaster

The Wingless pathway regulates development through precisely controlled signaling. This study shows that intracellular trafficking in the Wg target cell regulates Wg signaling levels. In Drosophila cells stimulated with Wg media, dynamin or Rab5 knockdown causes reduced reporter (Super8XTOPflash) activity, suggesting that internalization and endosomal transport facilitate Wg signaling. In the wing, impaired dynamin function reduces Wg transcription. However, when Wg production is unaffected, extracellular Wg levels are increased. Despite this, target gene expression is reduced, indicating that internalization is also required for efficient Wg signaling in vivo. When endosomal transport is impaired, Wg signaling is similarly reduced. Conversely, the expression of Wg targets is enhanced by increased transport to endosomes or decreased hepatocyte growth factor-regulated tyrosine kinase substrate- mediated transport from endosomes. This increased signaling correlates with greater colocalized Wg, Arrow, and Dishevelled on endosomes. Since these data indicate that endosomal transport promotes Wg signaling, these findings suggest that the regulation of endocytosis is a novel mechanism through which Wg signaling levels are determined (Seto, 2006).

This analysis has revealed the surprising finding that intracellular transport affects the efficiency of Wg signaling. In cell culture, knockdown of dynamin, a protein essential for clathrin-mediated internalization, reduces the TOPFlash/RL ratio (RL is Renilla luciferase), which is suggestive of decreased Wg signaling. Similarly, Rab5 knockdown causes reduced TOPFlash/RL ratios under most conditions, suggesting that internalization and endosomal transport are important for Wg signaling. Interestingly, transfection with polIII-RL, a control vector used in a recent screen for modifiers of Wg signaling, produces conflicting results for Rab5 compared with other RL controls, indicating that cell culture-based Wg signaling assays are very sensitive to experimental conditions. Thus, although the cell culture results indicate an endocytic regulation of Wg signaling, in vivo validation is critically important (Seto, 2006).

In the wing, further evidence was found that Wg signaling levels are highly dependent on intracellular transport. When endocytosis is altered, ligand levels and signaling levels are uncoupled such that high Wg levels do not necessarily enhance signaling. Therefore, limited usage has been made of the term morphogen gradient, which could refer to either ligand or signaling levels. Instead describe Wg distribution and signaling readouts are described. When internalization is inhibited in a domain that does not affect Wg production, high levels of Wg(ex) were found, likely as a result of reduced degradation. However, Wg target gene expression is diminished, indicating that impaired internalization decreases Wg signaling in vivo as well as in cell culture. When early endosomal transport is impaired, Senseless (Sens) and Distal-less (Dll) expression are also reduced despite abundant Wg levels. In both cases, markers of high signaling levels are especially affected, indicating that intracellular signaling is important to achieve robust Wg signaling levels. The differential decrease also argues that changes in Sens and Dll expression are not merely the result of cell death or global changes in transcription. Further supporting this, normal expression of other genes was found in the wing pouch. Additionally, when endosomal transport is enhanced or when transport from the endosome is impaired, Wg signaling is increased. These data suggest that protein localization to the endosome facilitates Wg signaling. Conversely, increased transport to MVBs decreases the expression of Wg readouts. This causes an adult wing phenotype that can be suppressed by Wg signaling components. Thus, it is proposed that in addition to low levels of cell surface signaling, intracellular Wg signaling is critical for proper signaling levels (Seto, 2006).

Because endocytosis is tightly regulated, intracellular Wg signaling may allow for the rapid modulation of signaling levels. For example, endosomal transport can be regulated merely by changing the GDP/GTP state of Rab5. This work indicates that impaired endosomal transport by GDP-bound Rab5 reduces Wg signaling, whereas enhanced endosomal fusion by GTP-bound Rab5 increases signaling. Because the GDP/GTP-binding state of Rab5 is controlled posttranslationally by GTPase-activating proteins and guanine nucleotide exchange factors, endocytic regulation likely allows more of a rapid adjustment of signaling than regulatory mechanisms requiring transcription and translation. Furthermore, because endocytic rates vary between cell types, this regulation may allow signaling to be adjusted in particular parts of the body or cells of a tissue. Thus, regulated endocytosis allows for precise temporal and spatial control of Wg signaling (Seto, 2006).

Endocytosis is hypothesized to regulate signaling through several mechanisms. For example, lysosomal degradation of internalized active receptor tyrosine kinases serves to attenuate signaling. However, the data suggest that Wg signaling is enhanced by endocytosis. One theory by which intracellular transport facilitates signaling is that the internalization of ligand-receptor complexes promotes interactions with other signaling members recruited to or already present on endosomes. In MAPK signaling, ERK1 receptors form protein complexes with endosomal MP1 and p14, leading to greater activation of signaling. Similarly, TGFβ signaling may be enhanced by receptor internalization to endosomes where the Smad2 anchor protein SARA is enriched. Although this work and that of others suggests that Wg undergoes receptor-mediated internalization in the wing, these data alone cannot explain the enhanced Wg signaling observed. However, not only are Wg and Arrow colocalized in large endosomal accumulations in hrs mutants, but they also colocalize with the cytoplasmic signaling component Dsh. The colocalization of Wg, Arr, and Dsh correlates with the increased expression of Wg readouts. These data suggest that internalization and endosomal transport may promote Wg signaling by facilitating associations between the Wg-receptor complex and downstream signaling components like Dsh. Interestingly, Dsh is reportedly present on intracellular vesicles, and mutations that impair vesicular localization do disrupt canonical Wg signaling (Seto, 2006).

Axin, a protein that inhibits Wg signaling by down-regulating Arm levels, has also been shown to colocalize with Dsh on intracellular vesicles. Upon Wg signaling, Axin relocalizes from intracellular puncta to the plasma membrane. This correlates with Arm stabilization and increased Wg signaling. Because Axin associates with Dsh and the cytoplasmic tail of Arr, it is proposed that internalized Wg forms an endosomal signaling complex that may relocalize Axin, thereby stabilizing Arm and facilitating signaling (Seto, 2006).

A model of intracellular Wg signaling is presented. Based on the data obtained from altering endocytosis, Wg at the cell surface produces only low levels of Wg signaling in the wing. Wg associates with its receptors and is internalized. When endocytic vesicles fuse with the early endosome, the cytoplasmic domains of the Wg receptors Frizzled and Arr are able to associate with downstream signaling components like Dsh, thereby facilitating Wg signaling. Subsequent endosomal sorting into MVB inner vesicles sequesters the Wg-receptor complex from other signaling components, and the activation of signaling transduction is halted (Seto, 2006).

Cellular retention and recycling of Wingless

There is considerable interest in the mechanisms that drive and control the spread of morphogens in developing animals. Although much attention is given to events occurring after release from expressing cells, release itself could be an important modulator of range. Indeed, a dedicated protein, Dispatched, is needed to release Hedgehog from the surface of expressing cells. In Drosophila embryos, much Wingless (as well as a GFP-Wingless fusion protein) remains tightly associated with secreting cells. Retention occurs both within the secretory pathway and at the cell surface and requires functional heparan sulfate proteoglycans. As a further means of retention, secreting cells readily endocytose Wingless protein that does reach the cell surface. Such endocytosed Wingless can in turn be sent back to the cell surface (the first direct observation of ligand recycling in live embryos). Recycling may serve to sustain high-level signaling in this region of the epidermis (Pfeifer, 2002).

In order to explore Wingless trafficking, transgenic Drosophila embryos that express biologically active GFP-Wingless were made. GFP was inserted at the amino terminus of Wingless, just downstream of the signal peptide. To first assess the activity of GFP-Wingless, it was asked whether it could replace endogenous Wingless during embryonic development. UAS-GFP-wingless was expressed in a wingless null mutant with wingless-GAL4. In most embryos, the mutant phenotype is rescued to a wild-type pattern, showing that GFP-Wingless is active. However, this does not necessarily mean that GFP-Wingless spreads normally along the epithelium, because the signal can be delivered by the progeny of expressing cells. Indeed, a membrane-tethered form of Wingless also rescues a wingless null mutant in this assay. To eliminate the contribution of cell spreading, UAS-GFP-wingless was expressed at the posterior side of the parasegment boundary (with engrailed-GAL4) in a wingless mutant (wingless^-; engrailed-GAL4 UAS-GFP-wingless). Since the parasegment boundary is a clonal boundary, it prevents the anteriorward spread of engrailed-expressing cells, and any action of GFP-Wingless toward the anterior in this assay must follow from the spread of the signal itself (all evidence suggests that there is no relay via a secondary signal. In embryos whose sole source of Wingless is GFP-Wingless expressed in the engrailed domain, large expanses of naked cuticle (an indication of Wingless signaling) form, and a near normal cuticle pattern develops, implying normal spread of GFP-Wingless. Because the cuticle pattern is a rather late readout of Wingless signaling, a more immediate assay, the embryonic expression of serrate, a gene that is repressed by Wingless signaling, was examined. The posterior edge of each serrate stripe marks the anterior limit of the range of Wingless at stages 11 and 12. GFP-Wingless driven with engrailed-GAL4 represses serrate across several cell diameters at the anterior of the engrailed-GAL4-expressing cells. On the basis of the above functional assay, it is concluded that GFP-Wingless spreads along the embryonic epidermis like the wild-type protein (Pfeifer, 2002).

With the aim of directly tracking the spread of Wingless independently of cell inheritance, GFP-Wingless was expressed in the engrailed domain and live embryos were imaged. As expected, expressing cells are very bright. However, little or no fluorescence is detectable outside the expression domain. Likewise, in wing imaginal discs, GFP-Wingless is retained by expressing cells, but, in this case, more GFP fluorescence is detected outside the expression domain. This could be because expressing imaginal disk cells retain Wingless to a lesser extent or because Wingless is not degraded as rapidly outside the expression domain. In any case, retention is specific to Wingless since a form of GFP engineered to be secreted (GFP^secr; GFP with the signal peptide of Wingless fills the perivitelline space even when it is expressed locally, either with engrailed-GAL4 or wingless-GAL4. Thus, GFP itself diffuses readily in the extracellular space, while Wingless appears to be specifically retained by expressing cells (Pfeifer, 2002).

Wingless transits for an extended time in the secretory pathway, and this probably contributes to retention by expressing cells. Retention also occurs at the cell surface. In the embryo, Wingless protein is most prominently found in intracellular vesicles located on the apical side. GFP-Wingless, too, is mostly found in apical vesicles, which are often seen to disappear at the apical surface as if fusing there. Thus, it seems likely that most Wingless is secreted apically (at the outer-facing surface of the embryo). Therefore, it was possible to adapt to the embryo a procedure designed to detect extracellular Wingless in imaginal discs. Live wingless^- embryos carrying UAS-GFP-wingless driven by engrailed-GAL4 were hand devitellinized, stained with anti-Wingless, and then fixed. After staining for extracellular Wingless, embryos were permeabilized and stained with anti-Engrailed to identify the domain of expression. All detectable extracellular Wingless is confined to expressing cells even though some Wingless must be present in anterior nonexpressing cells because they respond to Wingless (they repress serrate and make naked cuticle). Thus, in the embryonic epidermis, much extracellular Wingless remains associated with the surface of cells that secrete it (Pfeifer, 2002).

Using the same staining procedure, it was found that, in wild-type embryos, extracellular endogenous Wingless is detectable in 3- to 4-cell-wide stripes corresponding roughly to the range of Wingless action. In particular, Wingless is seen at the surface of cells that transcribe wingless (adjacent to the engrailed domain) and of cells located 2–3 cell diameters at the anterior. Since, as shown in the previous experiment, extracellular Wingless is only detectable at the surface of cells that secrete it, surface staining at the anterior of the domain of transcription likely reveals Wingless retained by the progeny of expressing cells. This means that most of the cells that respond to Wingless also secrete it and, therefore, the distinction between sending and receiving cells is blurred (Pfeifer, 2002).

One obvious class of molecules that could affect Wingless retention in expressing cells are heparan sulfate proteoglycans (HSPGs). The distribution of Wingless was examined in embryos lacking sugarless, a gene encoding an enzyme required for the biosynthesis of heparan sulfate. In embryos lacking maternal and zygotic sugarless, endogenous transcription of wingless decays (an indirect consequence of decreased Wingless signaling). The distribution was assayed of exogenous, HA-tagged, Wingless protein expressed from a UAS transgene controlled by paired-GAL4, a driver that is unaffected by the lack of Wingless signaling. As a control, HA-Wingless was driven by paired-GAL4 in otherwise wild-type embryos. In the control embryos, exogenous Wingless is easily detected in odd-numbered segments, within the paired domain. By contrast, in the absence of sugarless, very little exogenous Wingless is detectable despite continuous transcription from the paired-GAL4 driver. Both secretory and surface-associated Wingless are expected to contribute to signal upon traditional staining. Indeed, in control embryos, surface HA-Wingless alone is easily detectable, suggesting that if it were not affected by the sugarless mutation, it should contribute to signaling. It is concluded, therefore, that both surface and intracellular Wingless protein disappear in sugarless mutants, even if transcription is maintained at a high rate. This is in contrast with the effect of removing sulfateless in imaginal discs cells (where only extracellular Wingless is lost but is consistent with the effect of heparinase treatment, also in imaginal discs. sulfateless is required downstream of sugarless for further addition of sugar chains. It is conceivable, therefore, that depending on the level of sugar modification, proteoglycans differentially affect Wingless protein in a different subcellular compartment (Pfeifer, 2002).

Loss of detectable Wingless protein in sugarless mutants could follow from decreased stability or lack of retention by expressing cells. The latter alternative implies a concomitant increase in soluble extracellular Wingless, which is not apparent. However, sparse uniformly distributed protein might be hard to detect. Indeed, a functional assessment of Wingless signaling suggests that exogenous Wingless is present in the perivitelline space of sugarless embryos. In embryos lacking maternal and zygotic sugarless activity, engrailed expression decays for lack of Wingless signaling. This is rescued by paired-GAL4-driven exogenous Wingless, indicating that Wingless must be present in the extracellular space at a sufficient level to activate the pathway. Importantly, engrailed expression is rescued in all segments, even in even-numbered segments in which paired-GAL4 is not expressed. In otherwise wild-type embryos, paired-GAL4-driven Wingless does not reproducibly affect engrailed expression in even-numbered segments (their width is normal). Therefore, removing sugarless activity increases the range of Wingless (possibly allowing it to spread uniformly). This is consistent with a previous observation that engrailed stripes widen temporarily in sugarless mutants. Why then would engrailed expression decay in sugarless mutants? It is likely that, without retention, Wingless becomes diluted below levels sufficient to sustain engrailed expression. As a result, expression of hedgehog (a target of engrailed that is essential for sustained wingless expression) would decay, and the complete extinction of wingless expression would ensue, hence the terminal phenotype (Pfeifer, 2002).

It is concluded that HSPGs are needed for the retention of Wingless both at the surface and within the secretory pathway of expressing cells (an additional effect on stability, although unlikely, cannot be excluded). Retention could provide a regulatory step in the control of range. For example, a specific membrane protein (analogous to Dispached) could be involved in releasing GPI-anchored complexes comprising Wingless and Dally or Dally-like from the secretory pathway and cell surface, as and when required. In any case, these results highlight the role of HSPGs in cells that secrete a signal. A requirement in receiving cells is unlikely since engrailed expression is maintained by exogenous Wingless in the absence of sugarless activity (Pfeifer, 2002).

Even though no extracellular Wingless can be detected at the surface of identifiable nonexpressing cells, a few Wingless-containing vesicles can be seen in such cells, confirming that they were reached by Wingless. The expectation that these vesicles are endocytic was tested by injecting 10 kDa tetramethylrhodamine dextran (Rho-Dx) in the perivitelline space of engrailed-GAL4 UAS-GFP-wingless embryos. Rho-Dx is readily taken up by epidermal cells and thus marks the endocytic pathway. GFP-Wingless vesicles seen in nonexpressing cells also contain Rho-Dx, thus confirming their endocytic nature (Pfeifer, 2002).

Clearly, cells at the anterior of the parasegment boundary are capable of internalizing Wingless. These are the cells that, in the wild-type, manufacture and secrete Wingless. Therefore, it was expected that, at the anterior of the parasegment border of wild-type embryos, secretory and endocytic Wingless-containing vesicles coexist. To assess the relative importance of these two populations, GFP-Wingless was expressed under the control of wingless-GAL4. As with engrailed-GAL4, most of the fluorescence is confined to expressing cells (or their progeny), confirming that the Wingless signal remains tightly associated with expressing cells. To recognize endocytic vesicles, the perivitelline space of wingless-GAL4 UAS-GFP-wingless embryos was injected with Rho-Dx around stage 9, and embryos were subsequently imaged live, around stages 11 and 12. This procedure showed that GFP-Wingless-positive vesicles (about 50%) are endocytic. Therefore, wingless-expressing cells internalize Wingless at a high rate, perhaps reflecting the availability of Wingless at the cell surface. By contrast, very few endocytic vesicles are present when the same cells receive Wingless from a neighbor (as assayed by expressing GFP-Wingless in the engrailed domain). It is concluded that, although much intracellular Wingless is endocytic in wild-type embryos, most endocytosis of Wingless occurs in secreting cells, which have plenty of Wingless available at their surface (Pfeifer, 2002).

What happens to Wingless after it has been endocytosed? Is Wingless trafficking specifically regulated after endocytosis? To approach this question, the behavior of GFP-Wingless vesicles was investigated in live embryos. The movement of fluorescent vesicles was tracked in two dimensions in stage-11 and -12 embryos expressing GFP-Wingless with wingless-GAL4. This is the stage at which Wingless activity specifies epidermal cell fate. It is also the stage at which epidermal cells undergo an extensive rearrangement (germband retraction) that complicates the tracking of intracellular vesicles. A computational tool was developed to adjust for this offset movement. After removing the offset movement, the mean square displacement (MSD) was calculated for increasing time intervals, e.g., Dt. An MSD versus D plot was used as a quantitative characteristic of vesicle motion and also to classify various types of movement. Three classes of movement were recognized: random walks, directed diffusion, and diffusion in a cage. The MSD was calculated for endoGFP-Wingless vesicles (containing both GFP-Wingless and Rho-Dx), generic endocytic vesicles (only Rho-Dx positive), and secretory vesicles (fluorescent vesicles in UAS-GFP^secr wingless-GAL4 embryos). Importantly, differences were found between the behaviors of endoGFP-Wingless and generic endocytic vesicles. This is an indication that endoGFP-Wingless vesicles are handled differently from generic endocytic vesicles, suggesting that endocytic trafficking of Wingless is under distinct control (Pfeifer, 2002).

Endocytosed material can either be sent to lysosomes for degradation or be recycled to the plasma membrane. Often, ligand-receptor complexes dissociate in the endocytic pathway, with the ligand being forwarded to lysosomes and the receptor being recycled. It was specifically asked whether endoGFP-Wingless (i.e., the ligand) is ever recycled. This was done in live wingless^- embryos expressing GFP-Wingless with engrailed-GAL4. As argued above, vesicles found at the anterior of the expression domain are necessarily endocytic, and no Rho-Dx injection is required. Fluorescence time-lapse microscopy shows that such vesicles can return to the cell surface, providing the first direct observation of ligand recycling in a living embryo. Note that only the apical surface, which is optically accessible, was examined and therefore whether transepithelial transcytosis takes place cannot be ascertained. Ligand recycling is a key feature of the planar transcytosis model of transport; according to this model, the ligand spreads by repeated cycles of internalization, recycling, and presentation to further cells. The observation of Wingless recycling shows that planar transcytosis is plausible. However, one should note that, in the above experimental situation, very few endoGFP-Wingless vesicles are present in nonexpressing cells. Therefore, either a small number of vesicles are sufficient for transport or transcytosis is not an important contributor to transport. Since increased ligand recycling has been documented to enhance the potency of a signal in tissue culture cells, an alternative is that cells at the anterior of the parasegment boundary (which require sustained Wingless signaling) recycle Wingless to ensure maximal signaling. Indeed, if the contribution of cell movement is removed and, as a result, Wingless levels are drastically reduced in this region, signaling is nevertheless sustained, maybe as a result of ligand recycling (Pfeifer, 2002).

Fab1 phosphatidylinositol 3-phosphate 5-kinase controls trafficking but not silencing of endocytosed receptors

The trafficking of endocytosed receptors through phosphatidylinositol 3-phosphate [PtdIns(3)P]-containing endosomes is thought to attenuate their signaling. This study shows that the PtdIns(3)P 5-kinase Fab1/PIKfyve controls trafficking but not silencing of endocytosed receptors. Drosophila fab1 mutants contain undetectable phosphatidylinositol 3,5-bisphosphate levels, show profound increases in cell and organ size, and die at the pupal stage. Mutant larvae contain highly enlarged multivesicular bodies and late endosomes that are inefficiently acidified. Clones of fab1 mutant cells accumulate Wingless and Notch, similarly to cells lacking Hrs, Vps25, and Tsg101, components of the endosomal sorting machinery for ubiquitinated membrane proteins. However, whereas hrs, vps25, and tsg101 mutant cell clones accumulate ubiquitinated cargo, this is not the case with fab1 mutants. Even though endocytic receptor trafficking is impaired in fab1 mutants, Notch, Wingless, and Dpp signaling is unaffected. It is concluded that Fab1, despite its importance for endosomal functions, is not required for receptor silencing. This is consistent with the possibility that Fab1 functions at a late stage in endocytic receptor trafficking, at a point when signal termination has occurred (Rusten, 2006).

Cell growth, survival, proliferation, and differentiation are controlled by signals that activate their cognate receptors on the cell surface. Important examples include the soluble ligand Wnt (and its Drosophila homologue Wingless) and the membrane bound ligand Delta, which bind to G protein-coupled receptors and Notch receptors, respectively, on receiving cells. During development and normal physiology, the levels of the ligands and their receptors are tightly controlled in time and space (Rusten, 2006).

Receptor density at the cell surface is an important determinant of signaling responses, and there are both slow and fast mechanisms attenuating receptor levels. Transcriptional down-regulation is a slow and long-lasting mechanism, whereas posttranslational modification and/or internalization represent fast ways to reduce the amounts of functional receptors on the cell surface. Internalization of many receptors, including Notch and Wnt receptors, is followed by their transport from endosomes to lysosomes, where they become degraded, resulting in a transient reduction in the ability of cells to receive signals. Adding to the complexity of signaling regulation is the fact that ligand-bound receptors may also signal from endosomal membranes, and their signaling output from endosomes may differ from the output triggered from the plasma membrane (Rusten, 2006).

The key roles of the endocytic pathway in cell signaling are highlighted by the analyses of mutants interfering with endocytic trafficking. Such an example is provided by Hrs, a protein that sorts ubiquitinated receptors into intraluminal vesicles of multivesicular bodies (MVBs), destined for degradation in lysosomes. Drosophila hrs mutants show impaired sorting of receptors into MVBs, causing their accumulation in early endosomes. In hrs mutants, Dpp (a transforming growth factor-β homologue) and epidermal growth factor receptor signaling is enhanced, presumably because the activated receptors have a prolonged residence time in the limiting membrane of endosomes. Likewise, mutations of two subunits of the endosomal sorting complex required for transport (ESCRT)-I and -II, Tsg101 and Vps25, which are thought to function immediately downstream of Hrs, cause endosomal accumulation of receptors and tumor-like overproliferation in a cell nonautonomous manner due to increased Notch signaling. This supports the view that proper endocytic traffic has an important antitumorigenic function (Rusten, 2006).

Hrs is recruited to endosome membranes by binding the phosphoinositide (PI) phosphatidylinositol (PtdIns) 3-phosphate [PtdIns(3)P], formed by phosphorylation of PtdIns by a class III PI 3-kinase. PtdIns(3)P is specifically localized to endosomal membranes and not only recruits Hrs but also several other proteins containing FYVE or PX domains (Ellson, 2002; Stenmark, 2002). Class III PI 3-kinase and PtdIns(3)P are thus crucial regulators of endocytic trafficking, mediating endosome fusion as well as degradative sorting, recycling, and retrograde trafficking to the biosynthetic pathway (Lindmo, 2006a). PtdIns(3)P is metabolized by dephosphorylation and by lysosomal lipases. In addition, this PI can be phosphorylated in the 5-position of the inositol headgroup, giving rise to phosphatidylinositol 3,5-bisphosphate [PtdIns(3,5)P₂]. The kinase catalyzing this phosphorylation, Fab1, was first characterized in yeast. Saccharomyces cerevisiae fab1 mutants have abnormally enlarged vacuoles and show impaired trafficking of the ubiquinated cargo carboxypeptidase S to the vacuole lumen (Odorizzi, 1998). Fab1 is evolutionarily conserved, and overexpression of a kinase-dead mutant of the mammalian Fab1 homologue PIKfyve in cultured cells has been reported to inhibit fluid-phase transport of endocytic markers but not recycling/degradation of endocytosed receptors or sorting of procathepsin D (Ikonomov, 2001; Ikonomov, 2003). Moreover, PIKfyve has been found to be phosphorylated by the PI 3-kinase–regulated protein kinase, PKB, after insulin stimulation (Berwick, 2004), and PIKfyve colocalizes with a highly motile subpopulation of vesicles containing insulin-responsive aminopeptidase (Rusten, 2006).

These findings indicate that Fab1/PIKfyve plays a role in controlling specific membrane trafficking processes, but its functions in signal termination and in the physiology of a multicellular organism are not known. To address this, Drosophila fab1 mutants were generated and their phenotype was studied with respect to survival, growth, membrane trafficking and cell signaling. It was found that the activity of Drosophila Fab1 is essential for development and cell volume control and that its inactivation leads to endosomal accumulation of Wingless and Notch. Remarkably, this accumulation is not accompanied by increased signaling, indicating that Fab1, unlike Hrs and ESCRT-I and -II, is not involved in receptor silencing (Rusten, 2006).

That endosomal sorting of ubiquitinated cargoes is of great physiological importance is illustrated by studies of Drosophila mutants of the two ESCRT subunits, Tsg101 and Vps25. Loss of these proteins yields endosomal accumulation of receptors and ubiquitin, similarly to hrs mutants. Importantly, loss of Tsg101 and Vps25 in clones of cells causes a tumor-like overproliferation of adjacent tissue due to increased Notch-mediated signaling. No such effects were observed with fab1 mutant clones consistent with the finding that Notch signaling (as well as Wg and Dpp signaling) is unaffected in fab1 mutants. Thus, Fab1, unlike Hrs and ESCRT-I and -II proteins, does not seem to play any role in receptor silencing, even though it is important for receptor degradation. This is reminiscent of the ESCRT-III subunit hVps24, which mediates degradation but not silencing of the epidermal growth factor receptor. Moreover, it is interesting to note that impaired Hrs, Tsg101, or Vps25 function causes a strong accumulation of ubiquitinated proteins in endosomes, whereas this was not observed in fab1 mutant clones. These results, together with the fact that Fab1 mainly localizes to later endocytic structures than Hrs, suggest that Fab1 functions later than Hrs and ESCRT-I/-II in endocytic trafficking, at a point beyond receptor deubiquitination and signal termination (Rusten, 2006).

Studies in yeast and mammalian cells have suggested a role for Fab1 in endocytic membrane homeostasis, although its exact functions are not known. Indeed, confocal and EM revealed the accumulation of larger late endosomes in fab1 mutant Drosophila cells, consistent with previous studies in fab1 yeast and overexpression of kinase-dead PIKfyve in mammalian cells. The findings that the enlarged vacuoles in fab1 yeast mutants and late endosomes in kinase-dead PIKfyve-overexpressing cells contain few internal vesicles have suggested the possibility that Fab1 could mediate formation of such vesicles (Odorizzi, 1998; Ikonomov, 2001). In agreement with this, in fab1 mutant Drosophila cells enlarged endosomes were frequently observed with few or no intraluminal vesicles. However, in the fab1 mutants highly enlarged MVBs were frequently observed that were filled with numerous normal-sized intraluminal vesicles. This indicates that the increased endosome size in the absence of Fab1 cannot be explained by an inhibited formation of intraluminal vesicles, in contrast to what has been reported for Hrs. A more likely explanation is that late endosomes expand in fab1 mutants because of inhibited retrograde membrane flux to the biosynthetic and early endocytic pathways (Rusten, 2006).

Cell and organ size is controlled by genetic, hormonal, and environmental inputs. In particular, insulin signaling is important for growth, and the functions of the downstream class I PI 3-kinases in growth signaling are well characterized. The striking growth phenotypes observed in fab1 mutants indicate that PtdIns(3)P 5-kinase also regulates cell size. Interestingly, however, whereas PI 3-kinases promote growth, the current findings indicate that Fab1 has an inhibitory effect on cell size. Garland cells were strongly enlarged in fab1 mutants, suggesting a function of Fab1 in negative cell size regulation. In addition, fab1 deficiency led to a thickening of legs and enlargement of wings and heads, demonstrating a role for Fab1 in attenuating organ size. Overexpression of Drosophila Fab1 did not cause any overt growth-inhibitory effects, consistent with the finding that overexpression of Fab1 in yeast does not yield any increase in PtdIns(3,5)P₂ levels, presumably because regulatory components are limiting. No strong genetic interactions were detected between fab1 and mutants in components of the insulin signaling pathway, suggesting that the increased cell size in fab1 mutants may not be due to up-regulation of this pathway. Instead, there was a striking correlation between cell size and endosome overgrowth in fab1 mutant larvae. Thus, the increased cell and organ size in fab1 mutants may be due to the volume expansion of endosomes. It is therefore proposed that Fab1, through its effects on endosome morphology, functions in negative regulation of cell volume. Further work will reveal whether Fab1 also regulates cell size by additional mechanisms (Rusten, 2006).

Notch interaction with Wingless

The cell surface receptor Notch is required during Drosophila embryogenesis for production of epidermal precursor cells. The secreted factor Wingless is required for specifying different types of cells during differentiation of tissues from these epidermal precursor cells. The results reported here show that the full-length Notch and a form of Notch truncated in the amino terminus associate with Wingless in S2 cells and in embryos. In S2 cells, Wingless and the two different forms of Notch regulate expression of Frizzled 2, a receptor of Wg; hairy, a negative regulator of achaete expression; shaggy, a negative regulator of engrailed expression, and patched, a negative regulator of wingless expression. Analyses of expression of the same genes in mutant N embryos indicate that the pattern of gene regulations observed in vitro reflects regulations in vivo. These results suggest that the strong genetic interactions observed between Notch and wingless genes during Drosophila development is at least partly due to regulation of expression of cuticle patterning genes by Wingless and the two forms of Notch (Wesley, 1999).

Notch, a cell surface receptor, is required for the production of different types of cells during Drosophila development. Notch activates expression of one set of genes in response to ligand Delta and another set of genes in response to the ligand Wingless. Just how Notch initiates these different intracellular activities has been the focus of this study. Cultured cells expressing Notch were treated with Delta or Wingless, and the effect on Notch was examined by Western blotting. Treatment of cells with Delta results in accumulation of ~120-kDa Notch intracellular domain molecules in the cytoplasmic fraction. This form of Notch does not accumulate in cells treated with Wingless; rather, the ~350-kDa full-length Notch molecules accumulate. These results indicate that N responds differently to binding by Delta and by Wingless, and suggest that although the Delta signal is transduced by the Notch intracellular domain released from the plasma membrane, the Wingless signal is transduced by the Notch intracellular domain associated with the plasma membrane. It is proposed that the N receptor is a 'switch' for activation of different signaling pathways during development. Dl binds the EGF-like repeats 11-12 region to shunt the N120-Su(H) complex into the nucleus for turning on the expression of Dl-related genes. Wg binds the EGF-like repeats 19-36 region to send a transcriptional activator to the nucleus for turning on the expression of Wg-related genes (Wesley, 2000).

Schneider cells expressing Notch (S2-N cells) treated with Dl for 1h accumulate ~120-kDa N molecules (N120). Dl binds N in the extracellular region, including EGF-like repeats 11 and 12). S2 cells expressing N molecules lacking this region, N^DeltaEGF1-18, do not accumulate N120 molecules in response to treatment with Dl. This indicates that N120 accumulates in response to Dl binding N. N120 is the complete intracellular domain and is similar to the ~120-kDa N intracellular domain molecule shown to accumulate in vivo in response to D (Wesley, 2000).

N120 molecules do not accumulate in S2-N cells treated with Wg for 1, 2, or 5 hours. However, S2-N cells treated with Wg for 5h accumulates ~350-kDa N molecules (N350) but not S2-N cells treated with Dl. N350 is the full-length co-linear N molecule containing both the intracellular and extracellular domains. Wg binds N in the EGF-like repeats 19-36 region. S2 cells expressing N molecules lacking this region, N^{DeltaEGF19-36}, do not accumulate co-linear molecules when treated with Wg for 5h. In contrast, truncated, co-linear N^{DeltaEGF19-36} molecules containing the Wg binding sites accumulate upon treatment with Wg. These results indicate that accumulation of N350 in S2-N cells is in response to Wg binding N. Accumulation of N350 molecules is also discernible in cells treated with Wg for 2h when the blots are exposed to film for shorter periods. In contrast to Wg-treated cells, Dl-treated cells in the same blots always have lower levels of N350 compared with the levels in untreated cells (Wesley, 2000).

Accumulation of N350 molecules in Wg-treated cells is not due to activity of the endogenous Notch gene, which is rearranged in S2 cells. It is not due to a general increase or stabilization of all proteins in the cells: all N molecules do not accumulate, and the total protein levels in the three lanes are comparable. It is also not due to a Wg effect that is unrelated to N binding but retards N processing for cell surface presentation. Otherwise, co-linear N^{DeltaEGF19-36} would have also accumulated, but it did not. Thus, whereas Dl binding full-length N results in accumulation of N120, Wg binding results in accumulation of the co-linear N350 (Wesley, 2000).

Treatment of S2-N cells with Dl or Wg for 2 h also results in accumulation of ~55-kDa N molecules (N55). N55 contains only the amino terminus half of the intracellular domain, requires about 2 h to accumulate, and is variably recovered after about 3 h of treatment (Wesley, 2000).

To determine whether the responses observed in S2 cells are general N responses to treatments with Dl and Wg, the experiments were repeated with clone-8 cells that express N endogenously. The results show that N in clone-8 cells responds similarly to N in S2 cells. Treatment with Dl results in accumulation of N120 and not N350, whereas treatment with Wg results in accumulation of N350 and not N120; both Dl and Wg treatments result in accumulation of N55 molecules. The difference in levels of N350 between Dl-treated and Wg-treated cells is obvious after just 2 h of treatment. Clone-8 cells express a higher level of N55 molecules in the absence of any treatment, presumably because they also express Dl endogenously (Wesley, 2000).

When Dl binds N in vivo, the ~120-kDa N intracellular domain is released into the cytoplasm. To determine whether the N120 in these in vitro experiments with Dl also accumulates in the cytoplasm, S2-N cells were fractionated and analyzed following treatments with Dl and Wg. Following treatment with Dl, N120 molecules accumulate in the cytoplasmic fraction. In contrast, N350 molecules accumulate in the membrane fraction following treatment with Wg. N55 molecules are not consistently detected in these experiments as they are very unstable in this fractionation and extraction procedure (Wesley, 2000).

It is not known whether the N120 molecules that accumulate in the cytoplasm in response to Dl are the same as those present in the membranes or whether they are different molecules migrating in the same region of the gel. Membrane-tethered N intracellular domain (N^intra), untethered Nintra, and N120 migrate alongside each other in these gels. N120 molecules associated with the membranes or with the cytoplasm are probably the membrane-tethered or released N intracellular domain, respectively. Accumulation of N350 molecules in response to Wg is likely to be in the intracellular membranes associated with production of the heterodimeric cell surface receptor. N55 is derived from N350 upon activation of Notch signaling by a ligand (Wesley, 2000).

The development and patterning of the wing in Drosophila relies on a sequence of cell interactions molecularly driven by a number of ligands and receptors. Genetic analysis indicates that a receptor encoded by the Notch gene and a signal encoded by the wingless gene play a number of interdependent roles in this process and display very strong functional interactions. At certain times and places, during wing development, the expression of wingless requires Notch activity and that of its ligands Delta and Serrate. This has led to the proposal that all the interactions between Notch and wingless can be understood in terms of this regulatory relationship. This proposal has been tested by analyzing interactions between Delta- and Serrate-activated Notch signaling and Wingless signaling during wing development and patterning. Cell death caused by expressing dominant negative Notch molecules during wing development cannot be rescued by coexpressing Nintra. This suggests that the dominant negative Notch molecules cannot only disrupt Delta and Serrate signaling but can also disrupt signaling through another pathway. One possibility is the Wingless signaling pathway, since the cell death caused by expressing dominant negative Notch molecules can be rescued by activating Wingless signaling. Furthermore, the outcome of the interactions between Notch and Wingless signaling differs when Wingless signaling is activated by expressing either Wingless itself or an activated form of the Armadillo. For example, the effect of expressing the activated form of Armadillo with a dominant negative Notch on the patterning of sense organ precursors in the wing resembles the effects of expressing Wingless alone. This result suggests that signaling activated by Wingless leads to two effects: a reduction of Notch signaling and an activation of Armadillo (Brennan, 1999).

Expression of a dominant negative Notch molecule (Extracellular Notch or ECN) throughout the developing wing mimics the effects of loss of Notch function. However, Nintra cannot rescue the cell death caused by overexpressing ECN. Since Nintra provides constitutive signaling for Delta and Serrate during wing development and the effects of ECN are mediated by the sequestration of extracellular molecules that can interact with Notch, this suggests that the ECN molecule is sequestering extracellular molecules other than Delta and Serrate and attenuating signaling through another pathway. One candidate pathway is the Wingless signaling pathway, since the cell death caused by expressing the ECN can be rescued by activating Wingless signaling. Therefore, it is possible that the ECN molecule is sequestering the Wingless protein. The possibility that Wingless can bind the extracellular domain of Notch is supported by the results that are presented here, in particular, by two observations: first, that some of the deleterious effects of ECN can be suppressed by Wingless, but not Wingless signaling in the form of a constitutively active Armadillo molecule; and second, that this interaction requires specific EGF-like repeats of Notch, namely repeats 17-19 and 24-26 but not 10-12. Evidence for a physical interaction between Notch and Wingless has also been provided recently by Wesley (1999) who finds that the Wingless protein is enriched in a biopanning assay designed to identify proteins that interact with the extracellular domain of the Notch protein and that Wingless can be immunoprecipitated with Notch from embryo extracts and cultured cells. These experiments also show that the association of Wingless with Notch requires the integrity of a region of Notch centered around EGF-like repeats 24-26 (Wesley, 1999) which these experiments indicate are essential for the interactions that are described between Wingless and ECN during wing development and patterning (Brennan, 1999).

High levels of Wingless throughout the developing wing induce widespread development of sensory organs, an observation that correlates with the requirement for Wingless in this process during normal development. However, it is consistently observed that an activated form of Armadillo has a much weaker effect than Wingless on neural development. However, the difference is unlikely to be due to a weak UASarm* insert used in these experiments since in other instances where only a Wingless signal is required, such as the induction of the wing primordium during the early events of wing development, overexpressing Arm* or Wingless has very similar effects. A possible insight into the differences that the expression of Wingless and Arm* has on neurogenesis comes from the experiments where these two proteins are coexpressed with the ECN molecule. In these experiments the phenotypes generated by expressing UASECN with UASwg or UASarm* are very similar; namely, disrupting Notch signaling by expressing the ECN protein makes UASarm* and UASwg functionally equivalent. This suggests that the difference between the phenotypes generated by expressing Wingless and Arm* on their own might arise from the ability of Wingless to inhibit Notch signaling, which Arm* is unable to do; attenuating Notch signaling blocks lateral inhibition, which leads to increased numbers of sense organs. Since Wingless can activate Armadillo, overexpression of Wingless can achieve both effects simultaneously (Brennan, 1999).

When Arm* is coexpressed with ECN, the dominant negative molecule reduces Notch signaling, providing the function of Wingless that is missing in Arm* and thus making this molecule functionally equivalent to Wingless. These results raise the question of how Wingless signaling inhibits Notch signaling and where in the Wingless signaling pathway the cross-talk between the two pathways occurs. The inability of Arm* to inhibit Notch signaling indicates that the cross-talk must occur upstream of Armadillo. One possibility is that the inhibition occurs through Wingless interacting with the extracellular portion of Notch, preventing the Notch protein from interacting with its ligands. However, it is more likely to occur through the interaction of Dishevelled with the intracellular domain of the Notch protein, which has been shown previously to inhibit Notch signaling (Axelrod, 1996). In keeping with this, it has been found that overexpressing the Dishevelled protein can induce sense organ development as effectively as overexpressing Wingless; this suggests that Dishevelled can also disrupt Notch signaling as effectively as Wingless. Finally, it is possible that the interaction of Notch with both Dishevelled and Wingless is required to inhibit Delta signaling through Notch, since it has been shown previously that the ability to overexpress Dishevelled, which induces supernumerary sense organs, requires Wingless function (Axelrod, 1996). The interference of Wingless signaling with Notch signaling can also provide an explanation for the effects of ectopic expression of Wingless on the patterning of the veins and its sensitivity to the concentration of Delta. Overexpression of Wingless would reduce the availability of Notch for lateral inhibition by causing Dishevelled to sequester Notch into complexes that are unable to transduce the Delta signal. This would reduce the effectiveness of lateral inhibition signaling, an effect which would be exaggerated in situations of limiting signaling, as is observe in Dl heterozygotes or when Wingless is coexpressed with ECN (Brennan, 1999).

The interaction of Wingless and Notch signaling that has been observed might also be important during normal neural development. Wingless and Delta have opposite effects during neurogenesis; Wingless promotes while Delta suppresses the development of sense organs. Various experiments suggest that during the segregation of neural precursors a reduction of Notch signaling in the precursors themselves is as important as the Delta-mediated activation of Notch signaling in the surrounding cells. It is possible that, like the activation of Notch by Delta, the suppression of Notch signaling is an active process mediated by the interaction of Wingless and Dishevelled with Notch. If this were the case, since both Delta and Wingless have spatially and temporally regulated patterns of gene expression, their interactions with Notch could contribute to the well-documented bias in the appearance of precursors from clusters of cells with neural potential. This competitive interaction could also account for the observed increases in Wingless signaling associated with reductions in Notch signaling during lateral inhibition (Brennan, 1999).

Indirect interactions of Wingless

PTC is found associated with WG in discrete regions of the lateral plasma membrane of the embryonic epidermal cells. Preferential sites of accumulation resemble the described localization of the cell-cell adhesive junctions of the epidermal cells. PTC partially co-localizes with the WG protein in the wg-expressing and nearby cells, in structures that seem to be endocytic vesicles. These data suggest the interaction of PTC protein with elements of the reception complex of WG, as a means of controlling wg expression (Capdevila, 1994).

The segment polarity gene porcupine encodes a multitransmembrane protein involved in Wingless processing. porc mutant embryos, that is, embryos devoid of both maternal and zygotic porc gene activity, die before hatching and manifest a segment polarity phenotype indistinguishable from that of wingless mutants. Two genes known to require wg for expression in imaginal discs, neuralized and cut, are not seen in porc mutants. porc acts nonautonomously in wg patterning, meaning that porc mutant areas, in the background of a wild type epidermis, display a wild type configuration of denticle belts. Thus porc mutant cells can respond to Wingless from adjacent normal cells (Kadowaki, 1996).

PORC protein is concentrated at the reticular perinuclear endoplasmic reticulum. The protein has eight transmembrane segments suggesting that PORC may form a pore that facilitates transport of macromolecules, including Wingless protein, across membranes. Because the protein lacks the ATP-binding cassette that is characteristic of ABC transporter proteins, PORC may function in association with other transporter proteins. PORC modifies Wingless processing by changing the ratio of three forms of Wingless detected in lysates of cells, but does not significantly increase the total level of secreted WG. It is thought that form I is a precursor while forms II-IV are N-linked glycosylated products. All forms, with the exception of form I, contain high mannose-type N-glycans. Forms III and IV are synthesized in the presence of PORC, suggesting that PORC stimulates the processing of WG intermediates (Kadowaki, 1996).

Drosophila segment polarity gene product porcupine stimulates the posttranslational N-glycosylation of wingless in the endoplasmic reticulum

Wnt is a family of cysteine-rich secreted glycoproteins, which controls the fate and behavior of the cells in multicellular organisms. In the absence of Drosophila segment polarity gene porcupine (porc), which encodes an endoplasmic reticulum (ER) multispanning transmembrane protein, the N-glycosylation of Wingless (Wg), one of Drosophila Wnt family, is impaired. In contrast, the ectopic expression of porc stimulates the N-glycosylation of both endogenously and exogenously expressed Wg. The N-glycosylation of Wg in the ER occurs posttranslationally, while in the presence of dithiothreitol, it efficiently occurs cotranslationally. Thus, the cotranslational disulfide bond formation of Wg competes with the N-glycosylation by an oligosaccharyl transferase complex. Porc binds the N-terminal 24-amino acid domain (residues 83-106) of Wg, which is highly conserved in the Wnt family and stimulates the N-glycosylation at surrounding sites. Porc is also necessary for the processing of Drosophila Wnt-3/5 in both embryos and cultured cells. Thus, Porc binds the N-terminal specific domain of the Wnt family and stimulates its posttranslational N-glycosylation by anchoring them at the ER membrane possibly through acylation (Tanaka, 2002).

Porc functions on the N-terminal domain of Wg, which is conserved among Wnt family members. It is therefore possible that Porc functions on the processing of other Drosophila Wnt proteins in addition to Wg. To address this possibility, focus was placed on DWnt-3/5, because its specific antibody was available. In wild type embryos at stage 13, DWnt-3/5 is mainly localized on the commissural axon tracts of the central nervous system. In contrast, DWnt-3/5 appears to be confined in the cell bodies of neurons at the ventral nerve cord of porc embryos, which corresponds to DWnt-3/5 RNA expression domain. The shape and number of axon tracts is somewhat disorganized in porc embryos, but they are clearly present based on the staining pattern by monoclonal antibody BP102. Thus, in porc embryos, both Wg and DWnt-3/5 are not secreted from the synthesizing cells. In addition, Porc binds DWnt-3/5 and stimulates its N-glycosylation in S2 cells. These results therefore demonstrate that Porc can function on the N-glycosylation of multiple Drosophila Wnt proteins (Tanaka, 2002).

Sol narae (Sona) is a Drosophila ADAMTS involved in Wg signaling

ADAMTS (a disintegrin and metalloproteases with thrombospondin motif) family consists of secreted proteases, and is shown to cleave extracellular matrix proteins. Their malfunctions result in cancers and disorders in connective tissues. This paper reports that a Drosophila ADAMTS named Sol narae (Sona; CG9850) promotes Wnt/Wingless (Wg) signaling. sona loss-of-function mutants are lethal and rare escapers had malformed appendages, indicating that sona is essential for fly development and survival. sona exhibited positive genetic interaction with wntless (wls) that encodes a cargo protein for Wg. Loss of sona decreased the level of extracellular Wg, and also reduced the expression level of Wg effector proteins such as Senseless (Sens), Distalless (Dll) and Vestigial (Vg). Sona and Wg colocalized in Golgi and endosomal vesicles, and were in the same protein complex. Furthermore, co-expression of Wg and Sona generated ectopic wing margin bristles. This study suggests that Sona is involved in Wg signaling by regulating the level of extracellular Wg (Kim, 2016).

Proteases were originally started out as simple destructive enzymes in order to digest proteins and to provide amino acids to ancient organisms, but many proteases evolved in later times are specialized to change activity, localization, and binding properties of proteins and thereby affect many cellular functions. More than four hundred thousand proteases in all organisms can be classified into 9 categories and numerous subfamilies. Among these proteases, ADAMTS family as a subclass of ADAM (a disintegrin and metalloproteases) family constitutes a group of zinc-dependent secreted proteases widely expanded during metazoan evolution, including 6 members in flies, 5 members in nematodes, and 19 members in mammals (Kim, 2016).

These ADAMTSs are involved in many biological actions by processing mostly ECM and some non-ECM substrates. For example, ADAMTS-1 cleaves versican and aggrecan, and plays a key role in the ovulation process. ADAMTS-2, 3, and 14 cleave procollagen I , and mutations in ADAMTS-2 cause Ehlers-Danlos syndrome, a connective tissue disorder. ADAMTS-7 and 12 are significantly upregulated in arthritic patients. Besides ECM proteins, ADAMTS-13 cleaves von Willebrand factor (vWF) in blood, and mutations in ADAMTS-13 result in thrombotic thrombocytopenic purpura (TTP). In addition, ADAMTSs either enhance or inhibit cancer development. The level of ADAMTS-7 is upregulated in carcinoma and ADAMTS-1 promotes tumor development through the induction of stromal reaction. In contrast, ADAMTS-9 suppresses the formation of carcinoma by inhibiting angiogenesis, and stable expression of ADAMTS-16 decreases proliferation of cancer cells. Loss of ADAMTS-12 in mammals also increased tumor growth and progression (Kim, 2016).

ADAMTS is synthesized as a zymogen and has a relatively long prodomain. The physical interaction between the prodomain and the metalloprotease domain is essential for the latency of enzyme activity. Removal of prodomain in most ADAMTSs is mediated by furin, a proprotein convertase, in the secretory pathway. However, prodomains of ADAMTS-9, -10, -15 are processed by furin in ECM. In case of MIG-17 that is involved in male gonadal formation in C. elegans, the prodomain is cleaved autocatalytically. Thus, the activation mechanism of ADAMTS family appears to be diverse and may be tightly controlled in order to ensure the generation of active forms at the right time and place (Kim, 2016).

This study reports that an ADAMTS encoded by the CG9850 gene in Drosophila melanogaster is capable of promoting Wnt/Wg signaling. Wnt family proteins are conserved morphogens for growth, development and adult homeostasis in all metazoans. CG9850 was named sol narae (sona) meaning 'small wing' in Korean, based on the small wing phenotype of mutant escapers. Fly Wg, a homolog of mammalian Wnt-1, is a prototype of Wnt family proteins essential for the development of all fly appendages, and the wing imaginal disc has been an excellent system to study Wg signaling. Wg is known to be secreted from Wg-producing cells at the dorsal-ventral (DV) midline in the wing pouch and forms a concentration gradient in extracellular matrix (ECM). Wg binding to Frizzled receptors on the plasma membrane of Wg-responding cells activates Wg signaling cascade, and Wg effector proteins including Sens, Dll and Vg are expressed in different regions of wing pouch (Kim, 2016).

This study has focused on answering the following questions. Where and when is sona transcribed and translated? Where is the active form of Sona present? Which gene shows genetic interaction with sona? What are the in vivo roles of sona? sona exhibits a positive genetic interaction with wntless (wls) whose function is essential for secretion of Wg, and sona is positively involved in Wg signaling. Based on data provided in this report, it is proposed that Sona may modify proteins involved in Wg signaling (Kim, 2016).

ADAMTSs are secreted metalloproteases that are known to be involved in mainly ECM remodeling. Among six ADAMTSs in the fly, Papilin is essential for the formation of basement membrane and fly development, Stall functions in ovarian follicle formation and exhibits positive genetic interaction with Delta and ADAMTS-A is important for cell migration, especially in detaching cells from the apical ECM in salivary gland. This report has shown that Sona is a fly ADAMTS essential for fly development and survival. Transient coexpression of Sona and Wg increased the number of wing margin bristles, indicating that Sona is positively involved in Wg signaling. Accordingly, loss of sona decreased the level of Wg effector proteins as well as the level of extracellular Wg. Based on these results, it is proposed that Sona, as an ADAMTS, modifies yet unidentified protein(s) essential for Wg signaling (Kim, 2016).

During fly development, sona was transcribed at a high level in discrete regions in imaginal discs, which corresponded to the malformed regions in adult appendages of sona escapers. For instance, dorsal eye disc, the center of antenna disc, and outer ring of leg disc expressed the high level of sona transcripts, and sona escapers accordingly had disoriented ommatidial bristles in the dorsal eye, malformed arista, and kinked femur. Wing disc also exhibited the complicated mosaic pattern of sona transcription, and adult wings of sona escapers were small and abnormally shaped. Involvement of Sona in modulating the level of extracellular Wg may explain why these malformed adult structures are generated in sona escapers because Wg is specifically expressed in eye, wing and leg discs and determines the fate of organs (Kim, 2016).

The genetic link between Sona and Wg signaling was identified in a genetic screen in which a wls allele could rescue the lethal phenotype caused by the overexpression of Sona. Likewise, wing notching by the loss of wls was rescued by overexpression of sona. Furthermore, the loss of sona decreased the level of extracellular Wg. Taken together, these results raised a possibility that Sona may be involved in Wg signaling by affecting Wg secretion. How may Sona positively regulate Wg secretion? To act on Wg secretion, Sona has to be activated intracellularly, and function in secretory pathways. It has been shown that the prodomains of most ADAMTSs are cleaved in trans-Golgi network to become active. Thus, activated intracellular Sona may cleave unidentified proteins involved in Wg secretion and thereby promote the secretion of Wg. Indeed, intracellular Sona was enriched in the apical region while extracellular Sona are more enriched in the basolateral region. Similarly, intracellular and extracellular Wg are enriched in the apical and basolateral regions, respectively. It has been recently shown that Wg is secreted to the apical side and then reentered cells by endocytosis, and then moves to the basal side and secreted by transcytosis. It will be interesting to figure out whether Sona and Wg may be secreted together by transcytosis (Kim, 2016).

Besides the function of intracellular Sona for Wg secretion, presence of active Sona in conditioned medium of S2 cell culture suggests that extracellular active Sona may be involved in Wg signaling by modifying unknown ECM components. Immunocytochemical analysis of Sona confirmed that the active form of Sona devoid of the prodomain is present in basal ECM of wing discs. Therefore, active Sona may cleave ECM proteins that affect stability or activity of Wg. Well-studied ECM proteins essential for Wg signaling and formation of Wg gradient are Heparan sulfate proteoglycans (HSPG) such as Division abnormally delayed (Dally) and Dally-like (Dlp). These HSPGs can be modified by proteins such as Notum and Matrix metalloprotease 2 (Mmp2). Notum blocks Wg activity as α/β-hydrolase by modifying Dally and Dlp, and Mmp2 cleaves Dlp to inhibit the interaction between Dlp and Wg. Thus, Sona may act on these HSPGs or related ECM proteins to affect the stability or activity of extracellular Wg (Kim, 2016). Involvement of Sona in Wg signaling raises a possibility that some mammalian ADAMTSs may also be involved in Wnt signaling. Some mammalian ADAMTSs are known to function as positive factors for tumor invasion and progression. Overexpression of Wnts or downstream components of Wnt signaling also induces various tumors such as colon cancer, breast cancer, and leukemia. Wnt signaling is also essential for the growth and remodeling of bones and connective tissues. Overlapping functions of ADAMTSs and Wnt signaling supports the view that some mammalian ADAMTSs may be linked to Wnt signaling. Further work on identifying the intracellular or extracellular substrate(s) of Sona is required to fully understand how Sona is positively involved in Wg signaling (Kim, 2016).

Interaction of Wingless with its receptors: Dishevelled and the transduction of the Wingless signal

Continued wingless Protein Interactions part 3/3 | back to part 1/3

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.