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

There is evidence the Dishevelled is directly downstream of the FZ2, the Wingless receptor. Dishevelled acts in both wingless and frizzled signaling. Many frizzled proteins, including FZ2, contain a Ser/Thr-X-Val motif at their C-terminal end: this motif has been shown to interact with PDZ (or DHR) domains in a variety of proteins (Gomperts, 1996). DSH contains a PDZ domain, suggesting a direct interaction of DSH with FZ2 and Frizzled (Klingensmith, 1994).

In Drosophila, planar cell polarity (PCP) signaling is mediated by the receptor Frizzled (Fz) and transduced by Dishevelled (Dsh). Wingless (Wg) signaling also requires Dsh and may utilize DFz2 as a receptor. A heterologous system was used to examine the interaction of Dsh with Frizzled. mRNAs encoding Fz or Frizzled2 and a fusion of Dsh to green fluorescent protein (Dsh-GFP) were synthesized in vitro and injected into Xenopus embryos at the four-cell stage. Animal caps from stage 9 embryos, dissected to reveal the blastocoelar cells, were then examined by confocal microscopy. Dsh is recruited selectively to the membrane by Fz but not Frizzled2, and this recruitment depends on the DEP domain but not the PDZ domain in Dsh. When Fz is expressed simultaneously with Dsh-GFP, Dsh-GFP shows a qualitative redistribution to the membrane or cell cortex. Under these conditions, localization of Dsh-GFP to filopodia present on the blastocoelar (free) surfaces of the animal cap cells was also noticed. Staining with phalloidin (and Dsh-GFP) revealed that the filopodia contain filamentous actin. It is interesting to note that although the filopodia stain with Dsh-GFP, little or no Fz localizes there; at the cell cortex, the Fz and Dsh-GFP show imperfect colocalization. Fz staining is localized predominantly to the plasma membrane, and to a lesser extent to intracellular membranes (probably ER and/or Golgi) in these cells. This suggests that while Fz may induce localization of Dsh-GFP to the membrane and filopodia, it may do so by a mechanism other than direct binding. Frizzled2, the Wingless receptor, fails to induce membrane localization of Dsh, even in the presence of a functional Fz2 ligand (Axelrod, 1998).

Drosophila Dsh is a modular protein of unknown function that is well conserved in relation to its vertebrate homologs. Alignment of family members reveals three conserved domains. The first, a DIX domain, is similar to a domain in murine Axin (see Drosophila Axin), a recently described modulator of the Wnt1 pathway. The second contains a PDZ domain; PDZ domains recognize and bind short motifs at the carboxyl termini of proteins (but may bind other motifs as well). PDZ domains can also form dimers. The third domain, called DEP, is conserved among a set of proteins that have in common the ability to regulate various GTPases, including both heterotrimeric G proteins and Ras-like small GTPases. A mutation in the DEP domain impairs both membrane localization and the function of Dsh in PCP signaling, indicating that translocation is important for function. A single amino acid substitution in the DEP domain of Dsh is shown to confer a loss of function for PCP signaling, yet the mutant protein is functional for Wg signaling. This single amino acid substitution, coded for by the dsh1 allele, allows for translocation to the membrane, but is thought to impair the ability of Dsh1 to associate with its target at the membrane. This altered membrane interaction diminishes the ability of Dsh1 to function in PCP signaling (Axelrod, 1998).

Further genetic and molecular analyses suggest that conserved domains in Dsh function differently during PCP and Wg signaling, and that divergent intracellular pathways are activated. For example, the individual domains Dsh(DIX) and Dsh deleted for the PDZ domain, are each dominant negative for Wg signaling but have no effect on PCP signaling. Overexpression of Zeste white3, or an activated Arm protein, both involved in Wingless signaling, also fail to produce any effect on PCP. It is proposed that Dsh has distinct roles in PCP and Wg signaling. The PCP signal may selectively result in focal Fz activation and asymmetric relocalization of Dsh to the membrane, where Dsh effects cytoskeletal reorganization to orient prehair initiation. This analysis suggests that Dsh has different interactions in PCP and Wg signaling and predicts an additional genetic behavior (Axelrod, 1998).

If Wg and planar polarity signaling utilize Dsh in a common fashion, then ectopic activation of one pathway should be able to cross-activate the other by promiscuously activating Dsh. In contrast, if each pathway utilizes Dsh in a distinct fashion, then ectopic activation might sequester Dsh in pathway-specific complexes, rendering it unavailable and therefore titrating the activity of the other pathway. These possibilities could best be tested under conditions in which Dsh is limiting. Overexpression of Fz causes a dominant gain-of-function PCP phenotype, and this phenotype is sensitive to the dose of dsh. Can Wg cross-activate Dsh activity for PCP signaling, or can it sequester Dsh? To investigate this question, it was first necessary to know if the Fz-overexpression phenotype would be either enhanced or suppressed by Wingless overexpression. Ectopic expression of Wg suppresses the Fz overexpression phenotype, suggesting that activation of Wg signaling may titrate the amount of Dsh available for PCP signaling. The reciprocal experiment was performed by asking if ectopic activation of the PCP pathway could interfere with Wg signaling. Because the ligand for PCP signaling is unknown, Fz was overexpressed during embryogenesis, and the cuticle phenotype analyzed. These embryos develop with lawns of denticles and are reminiscent of wg-mutant embryos, or those expressing dominant-negative Dsh constructs. The results suggest that titration can occur in this direction as well. The possibility cannot be ruled out that the titration observed in these experiments results from a promiscuous interaction between Wg and Fz, although this interaction may not occur in vivo. The observations are equally consistent with the possibility that under these conditions, activity of one pathway titrates the Dsh level available for the other (Axelrod, 1998).

The dishevelled gene interacts antagonistically with Notch and its ligand Delta. A direct physical interaction between Dishevelled and the Notch carboxyl terminus, distal to the cdc10/ankyrin repeats, suggests a mechanism for this interaction. It is proposed that Dishevelled, in addition to transducing the Wingless signal, blocks Notch signaling directly, thus providing a molecular mechanism for the inhibitory cross talk observed between these pathways (Axelrod, 1996). It therefore appears that Wingless and Notch talk to each other through Dishevelled.

There are similarities between how Wg signals to cells in the embryonic epidermis and in wing discs. Several studies suggest that Wg acts as a morphogen in both tissues. Moreover, it utilizes the same signal transduction pathway in responding cells. However, there are significant differences between how Wg works in the two tissues. In the wing disc, Wg specifies cell fate decisions but has no apparent role in controlling planar polarity of wing cells. In the embryonic epidermis, Wg specifies cell fate decisions and controls the planar polarity of cells. This planar polarity is manifested by the orientation of denticles along the anteroposterior axis, which is disrupted in wg mutants or can be redirected by wg misexpression. There is a second major difference between wing and embryo: in the wing, DFz2 and not Fz mediates the Wg signal. Misexpression of DFz2 increases the zone of Wg responsiveness in the wing, but Fz misexpression has no effect. Null fz mutants do not perturb cell fate decisions attributable to Wg. In the embryo, both Fz and DFz2 are required to mediate the Wg signal. Inhibition of both genes is sufficient to disrupt planar polarity and epidermal cell differentiation, whereas inhibition of each gene singly has no effect. These data are also consistent with experimental results in which Fz was overexpressed in embryos (Kennerdell, 1998 and references).

How can Wg, Fz, and DFz2 generate both polarity and cell fate responses in embryos and not in wing discs? One possibility is that in embryos they directly specify cell fates and indirectly affect cell polarity. For instance, they specify the diverse pattern of denticle types that might then determine overall denticle polarity. Another possibility is that distinct domains of Wg activate different cell responses by interacting with receptors in qualitatively different ways. Access to some Wg domains might be limiting in some tissues and not others. A third possibility is that Wg ligand-receptor interactions are quantitatively different in various tissues. A fourth possibility is that intrinsic factors couple Wg-bound Frizzled proteins to a particular cell response, and these factors are differentially active in various tissues (Kennerdell, 1998 and references).

Most ligands pair with specific receptors, and each pairing remains fixed for different tissues and different developmental stages. Wg appears to be an exception to this general rule. What is the significance behind Wg's diverse signaling properties? By adding greater flexibility in the competence of cells to respond to Wg, more diverse responses to a single ligand can be generated. Competence may be modified by changing the number of potential receptors and their ability to trigger more than one transduction pathway. Another reason for this diversity might be related to the function of Wg receptors in shaping the concentration gradient of Wg in a tissue. In the wing disc, high levels of Fz2 stabilize extracellular Wg and allow it to range farther from its source than in the absence of Fz2. Thus, if more than a single Frizzled protein can stabilize Wg, the combination of multiple receptor expression patterns might determine the Wg gradient. This simple combinatorial mechanism could potentially generate a broad range of gradient curves for a single ligand (Kennerdell, 1998).

Regulated endocytic routing of Wingless

Embryos have evolved various strategies to confine the action of secreted signals. Using an HRP-Wingless fusion protein to track the fate of endocytosed Wingless, it has been shown that degradation by targeting to lysosomes is one such strategy. Wingless protein is specifically degraded at the posterior of each stripe of wingless transcription, even under conditions of overexpression. If lysosomal degradation is compromised genetically or chemically, excess Wingless accumulates and ectopic signaling ensues. In the wild-type, Wingless degradation is slower at the anterior than at the posterior. This follows in part from the segmental activation of signaling by the Epidermal growth factor receptor, which accelerates Wingless degradation at the posterior, thus leading to asymmetrical Wingless signaling along the anterior-posterior axis (Dubois, 2001).

In Drosophila embryos, the distribution of Wingless appears to be regulated because it undergoes a transition from symmetrical to asymmetrical during stage 10, 5 hr after egg deposition. Most notably, before stage 10, Wingless spreads into the engrailed domain (at the posterior of each stripe of wingless expression), while shortly thereafter, it becomes barely detectable in the same cells. Thus, at early stages, Wingless spreads toward the posterior to specify the width of the engrailed domain (an early function of Wingless) and then recedes to allow distant engrailed-expressing cells to acquire a denticle fate (a decision taken after stage 11). How is this transition in Wingless distribution regulated and does it follow from a change in transport or stability (Dubois, 2001)?

During early Drosophila embryogenesis, the Wingless protein can be detected throughout the posterior compartment where it acts to sustain engrailed expression. Later, between stages 10 and 11, around 5 hr after egg deposition (AED), a transition occurs and Wingless becomes barely detectable within the engrailed domain even though it continues to be secreted by wingless-expressing cells. This early drop in Wingless staining follows from a decrease in the transcription of frizzled and frizzled2 within engrailed-expressing cells. Indeed, ectopic expression of frizzled or frizzled2 prolongs the presence of Wingless-containing vesicles in the engrailed domain. However, this is only a reprieve. In engrailed-gal4 UAS-Frizzled2 or engrailed-gal4 UAS-Frizzled embryos, the Wingless protein still decays within the engrailed domain (but later, around stage 11-12). This could be because transport into the engrailed domain is prevented or ineffective after this stage. However, the effect of a dominant-negative Frizzled2 (the N-terminal extracellular domain linked to a GPI anchor, Deltafrizzled2-GPI) suggests otherwise. If Deltafrizzled2-GPI is expressed in the engrailed domain, Wingless continues to be detected in receiving cells, even at late embryonic stages. This suggests that Wingless can be transported into the engrailed domain after stage 11-12. Importantly, Deltafrizzled2-GPI is likely to lack an endocytic signal since it is devoid of all intracellular residues. Since expression of Deltafrizzled2-GPI prevents the drop in Wingless staining within the engrailed domain, endocytosis of the Wingless/receptor complex could be responsible for downregulating Wingless levels (and function) there. This possibility was explored with an HRP-Wingless fusion protein (Dubois, 2001).

There are two main benefits from using an HRP fusion to study endocytic trafficking. One is that HRP activity is easily detected after reaction with 3, 3' diaminobenzidine (DAB), which produces an electron-dense deposit. Unlike many antigens, HRP activity is unaffected by fixation with gluteraldehyde and therefore, ultrastructural details are optimally preserved for electron microscopy (EM). A second benefit follows from the relative stability of HRP within the destructive environment of the late endosomal and lysosomal compartments. Importantly, this appears to be true even if HRP is part of a fusion protein. Tagging an endocytosed protein with HRP, enables it to 'be seen', even after it has been degraded in lysosomes (Dubois, 2001).

To track the degradation of Wingless in cells receiving the signal, transgenic flies were constructed expressing an HRP-Wingless fusion protein. Since modifications at the C terminus of Wingless are known to reduce signaling activity drastically, HRP was inserted at the N terminus of Wingless, just downstream of the signal peptide. Several genetic tests were performed to determine whether the fusion is secreted and activates the Wingless pathway. These show that driving UAS-HRP-Wingless with engrailed-gal4 substantially rescues the phenotype of a wingless null mutant. It is concluded that HRP-Wingless is secreted, acts at a distance, and is sufficiently active to replace endogenous Wingless in the embryonic epidermis (Dubois, 2001).

Having established that HRP-Wingless is active, it was used to track the fate of Wingless in cells receiving the signal. UAS-HRP-Wingless was expressed with wingless-gal4 and HRP activity was assayed with an in situ DAB reaction. At stage 12, the Wingless protein is no longer detectable within the engrailed domain of wild-type embryos. By contrast, vesicles containing HRP activity are plentiful in the same cells and at the same stage in the transgenic embryos. Importantly, in embryos of this genotype, vesicles containing Wingless immunoreactivity are very rarely detected, suggesting that the high number of HRP vesicles is not simply a consequence of overexpression from the wingless-gal4 driver. Thus, it appears that, within signal-receiving cells, the Wingless moiety of the fusion is partially degraded or denatured while HRP remains. This suggests that the fusion protein is targeted to the lysosomal compartment where only HRP is detectable (Dubois, 2001).

To confirm that much of the HRP activity detected in the above experiment is in degradative structure, embryos expressing HRP-Wingless were analyzed by EM. For such analysis, it is crucial to distinguish cells receiving the signal from those expressing it. Expressing cells are easily recognizable because their Golgi apparatus and endoplasmic reticulum and even the nuclear membrane are heavily labeled with DAB deposits. Staining in all these structures was used to recognize expressing cells. The subcellular location of HRP was then assessed in posterior, nonexpressing cells. There, HRP activity is mainly localized within degradative structures such as multivesicular bodies (MVBs) and lysosomes. It is concluded that, at stage 12, HRP-Wingless is rapidly endocytosed and targeted to lysosomes in cells posterior to the normal domain of wingless expression. Presumably, in the wild-type, Wingless is barely detectable in the same region because targeting to lysosomes is so rapid (Dubois, 2001).

Overexpression of Wingless in the engrailed domain has relatively minor phenotypic consequences. In particular, no bald cuticle is induced at the posterior of the source, suggesting a failure of Wingless to act in this direction. Indeed, in stage 12 engrailed-gal4 UAS-Wingless embryos, very little Wingless protein is detected at the posterior of engrailed cells, despite their massive overexpressing of Wingless. This could be due to a block of Wingless transport across the segment boundary at the posterior of the engrailed domain. However, the above results suggest an alternative, namely that Wingless is transported toward the posterior but is rapidly degraded there and is therefore undetectable. This possibility was tested by expressing UAS-HRP-Wingless with engrailed-gal4. Many HRP-containing vesicles can be seen at the posterior even at stages 12-13. For the most part, these vesicles are degradative structures (recognized by EM) and do not contain Wingless immunoreactivity. Thus the HRP-Wingless fusion (and presumably Wingless too) does cross the segment boundary toward the posterior but is rapidly forwarded to lysosomes. Taken together, these results show that the zone of Wingless degradation includes the engrailed domain and more posterior cells (Dubois, 2001).

The range of Wingless action is asymmetric in the embryonic epidermis. This is best shown in wingless mutant embryos that express Wingless in the engrailed domain because, after stage 11, the engrailed domain has well-defined boundaries both at the anterior and the posterior; hence, the fate of endocytic Wingless on both sides of the source can be compared. In such embryos (whose only source of Wingless is the engrailed domain), naked cuticle is made over 3-4 cell diameters at the anterior while no naked cuticle is made at the posterior. It was asked if this asymmetry is correlated with differential lysosomal targeting. UAS-HRP-Wingless was expressed with engrailed-gal4 and EM was used to identify HRP-containing vesicles in nonexpressing cells on both sides of the source. Labeled (HRP-containing) degradative structures are particularly easy to recognize because they are large. These were counted over a range of three cells on either side of the expression domain. Data was collected from several sections of stage 12 embryos. As expected, more vesicles are detected, on average, near the source. Superimposed on this graded distribution, a clear difference can be seen between anterior and posterior. Overall, four times more labeled MVBs and lysosomes were found at the posterior than at the anterior. Thus, HRP-Wingless is preferentially targeted to degradative structures at the posterior. Whether increased degradation follows from increased endocytosis or increased targeting of endocytic vesicles to degradative structures cannot be formally distinguished (Dubois, 2001).

A vesicle that contains both HRP and Wingless immunoreactivity is presumed to be an early endosome while, if only HRP is detectable, it is likely to be at a later stage of the endocytic pathway. Using double immunofluorescence, many vesicles were found containing both Wingless and HRP at the anterior of the expression domain. By contrast, most vesicles at the posterior are only stained with anti-HRP, confirming that they are degradative structures. It is suggested that, in posterior cells, Wingless is actively targeted to the lysosomal compartment while at the anterior, internalized Wingless lingers in early endosomes or is recycled (although some degradation takes place, too). Note that although confocal microscopy clearly reveals vesicles that contain HRP (and Wingless) at the anterior, such vesicles cannot be unambiguously identified by EM. Maybe they are too small or they contain too little fusion protein for the DAB reaction to generate enough contrast (Dubois, 2001).

So far, a clear inverse correlation has been shown between the rate of targeting to lysosomes and the ability of Wingless to signal. Would compromising lysosomal targeting affect signaling output? The endocytic pathway is affected by mutations in a variety of genes. One is clathrin, a gene required for endocytosis. Another is deep orange, which encodes a homolog of the yeast vacuolar protein-sorting protein, Vps18p. In Drosophila, the deep orange gene product is required for normal delivery of proteins to lysosomes. Unfortunately, one cannot generate embryos lacking the function of either gene because both genes have a strong maternal contribution that is required for oogenesis. Zygotic mutants of clathrin or deep orange proceed through development relatively normally and die only at the end of embryogenesis, without an obvious phenotype. In particular, the denticle pattern is essentially normal. It was reasoned that such zygotic mutants would have reduced activity of the corresponding gene (since they only have maternal products) and may therefore show a phenotype in a sensitized genetic background. Overexexpression of Wingless with the engrailed-gal4 driver leads to a relatively mild phenotype (in particular, row 2 denticles are present even though they are adjacent to the Wingless-misexpressing cells). Presumably, the degradation machinery acting at the posterior is able to cope with the excess Wingless. However, this is no longer the case when the activity of clathrin or deep orange is reduced. In the absence of zygotic contribution from clathrin or deep orange, Wingless originating from the engrailed domain produces excess naked cuticle within areas normally occupied by denticle belts, an indication of excess Wingless signaling at the time when cuticular fates are specified (Dubois, 2001).

This result suggests that in clathrin or deep orange mutants, excess Wingless is no longer degraded fast enough to allow denticle formation. This was confirmed by looking directly at Wingless protein in embryos that lack the zygotic contribution of clathrin or deep orange and express wingless under the control of engrailed-gal4. In either mutant, increased Wingless staining is seen both anterior and posterior to the source. In deep orange mutants, Wingless seems to accumulate in intracellular vesicles, some of which appear enormous. These are likely to be similar to the giant MVBs reported to accumulate in photoreceptors of deep orange animals. Note that in deep orange mutant embryos, staining is also strongly increased within the domain of expression. Most likely, expressing cells continue to endocytose Wingless at a high rate but cannot forward it to late endosomal compartments. In clathrin mutants, excess Wingless appears to localize at the cell membrane, consistent with the role of clathrin in endocytosis (Dubois, 2001).

In clathrin mutants, the spread of Wingless appears to broaden, in apparent contradiction with the report that a mutation in shibire (which encodes Dynamin) reduces the range of Wingless in Drosophila embryos. In fact, it has been suggested that recycling of internalized Wingless powers transport of the signal by a mechanism of planar transcytosis. However, in imaginal discs, shibire is required for both endocytosis and secretion. Therefore, the apparent reduction in the spread of Wingless in shibire mutant embryos could be due to reduced secretion. Clearly, the role of endocytosis (and planar transcytosis) in Wingless transport along the embryonic epidermis must be reexamined (Dubois, 2001).

Loss of clathrin or deep orange function leads to the accumulation of Wingless at the anterior (as well as the posterior) of the source. This suggests that Wingless is also degraded at the anterior. It is proposed that the difference between anterior and posterior cells is quantitative. Indeed, EM analysis shows that Wingless is targeted to lysosomes at the anterior but less frequently than at the posterior (Dubois, 2001).

Denticles are synthesized on a template of actin protrusions that form around stage 15. Therefore, the prospective denticle pattern can be recognized before cuticle deposition by staining with phalloidin. The area where Wingless accumulates correlates with the absence of actin bundles. This confirms the functional link between lack of Wingless degradation and the formation of naked cuticle (Dubois, 2001).

Clearly, clathrin and deep orange are needed to remove excess Wingless signal. It was asked if lysosomal degradation of Wingless is needed when normal quantities of Wingless are produced, such as in a wild-type embryo. Mutations in clathrin or deep orange are either too strong to permit cell viability or too weak to show a recognizable phenotype in embryogenesis. Using chloroquine, an antimalarial drug, lysosomal function can be inhibited in a subtler manner. Chloroquine is not membrane permeant and reaches late endosomal compartments through the endocytic pathway. There, it raises the pH and thus interferes with normal function. In order to preferentially block lysosomal function in epidermal cells, chloroquine (100 mg/ml) was injected into the perivitelline space around stage 9-10. Sixteen percent of the chloroquine-injected embryos go on to form large patches of ectopic naked cuticle. By comparison, ectopic naked cuticle was rarely seen in control embryos (3%) and covered much smaller areas. In a further 27% of chloroquine-injected embryos, row 1 denticles were lost in one or more segment. This was seen in only 3% of the control, buffer-injected, embryos. Overall then, excess naked cuticle was seen in 43% of the chloroquine-injected embryos, compared to 6% in the controls. This phenotype is most likely due to excess Wingless signaling following a failure to degrade endogenously produced Wingless at the time when Wingless is normally degraded to allow the specification of denticle fates. Indeed, many Wingless-containing vesicles were detected in the engrailed domain of chloroquine-injected embryos, even as late as stage 12, whereas, in the controls, very few such vesicles are seen. Chloroquine not only affects the overall distribution of Wingless but also its subcellular localization. Large Wingless-containing vesicles are seen, an observation that is consistent with a block of lysosomal degradation and the accumulation of Wingless in an endosomal compartment. In conclusion then, downregulation of the Wingless protein is required to terminate signaling and ensure that no ectopic naked cuticle forms (Dubois, 2001).

The spatial and temporal regulation of Wingless degradation implies the existence of one or several regulatory genes that are activated at the posterior and not at the anterior of each wingless stripe. The activity of hedgehog, as well as that of its downstream effector cubitus interruptus, are needed to prevent the spread of Wingless toward the posterior. In hedgehog (or cubitus interruptus), mutant embryos that overexpress wingless under the control of engrailed-gal4, many Wingless-containing vesicles are detected at the posterior and this correlates with increased Wingless signaling there. In light of the present results, it is presumed that a target of hedgehog is needed to accelerate Wingless degradation at the posterior. One candidate target gene is rhomboid, because it is only expressed at the posterior of each stripe of engrailed (and hedgehog) expression. Moreover, segmental expression of rhomboid commences around stage 11, roughly the stage when the second phase of Wingless degradation begins. By analogy with the experiments with hedgehog and cubitus interruptus, rhomboid mutants that overexpress Wingless under the control of engrailed-gal4 were examined and increased Wingless staining was found both within and at the posterior of each engrailed stripe. Thus, in the absence of rhomboid, Wingless degradation is impaired. This result implicates Egfr signaling, since rhomboid encodes a limiting factor needed for the activation of the Egfr ligand Spitz. A null mutation in Egfr leads to extensive morphological defects, thus making staging and analysis difficult. Nevertheless, in embryos that could be analyzed (of the genotype engrailed-gal4 UAS-Wingless EGFR-), excess Wingless is detected at the posterior of the expression domain. The role of Egfr could be mediated by a target gene of the MAP kinase pathway. Alternatively, or in addition, a nontranscriptional response to Egfr signaling could lead to Wingless degradation. The role of Pointed, a transcription factor that mediates many activities of the Egfr in Drosophila, was examined. In embryos of the genotype engrailed-gal4 UAS-Wingless pointed -, excess Wingless accumulates posterior to the source. Therefore, it appears that a transcriptional target of Egfr signaling is involved in modulating Wingless degradation (Dubois, 2001).

rhomboid mutants were used to assess the role of EGFR signaling when wild-type levels of Wingless are produced. This cannot be done by simply looking at rhomboid mutants because the first phase of Wingless clearance (following transcriptional repression of frizzled and frizzled2) already brings Wingless below detection level. Receptor expression was therefore artificially maintained in rhomboid mutants (engrailed-gal4 UAS-Frizzled2 rhomboid -). The general morphology of such embryos is again somewhat aberrant. Nevertheless, it could be clearly seen that Wingless-containing vesicles linger within the engrailed domain, even as late as stage 12, thus confirming the role of rhomboid in targeting Wingless to lysosomes. Although this is hard to prove formally, Wingless accumulation in rhomboid mutants appears to be in intracellular vesicles. In particular, the subcellular distribution of Wingless is clearly different from that seen in clathrin mutants or in embryos expressing Deltafrizzled2-GPI; two situations when Wingless accumulates at the cell surface. In conclusion, it is suggested that Rhomboid (and Egfr) regulate the transfer of Wingless from endosomes to degradative structures. However, it is unlikely to be the sole regulator since not all engrailed-expressing cells accumulate Wingless in the rhomboid mutant. Additional regulators might include a redundant homolog of rhomboid or a gene controlling a parallel degradation pathway (Dubois, 2001).

Wingless signaling at the Drosophila wing margin is SNARE-dependent

The wing of Drosophila has long been used as a model system to characterize intermolecular interactions important in development. Implicit in an understanding of developmental processes is the proper trafficking and sorting of signaling molecules, although the precise mechanisms that regulate membrane trafficking in a developmental context are not well studied. The Drosophila wing was used to assess the importance of SNARE-dependent membrane trafficking during development. N-Ethylmaleimide-sensitive fusion protein (NSF) is a key component of the membrane-trafficking machinery and a mutant form of NSF was constructed whose expression was directed to the developing wing margin. This resulted in a notched-wing phenotype, the severity of which was enhanced when combined with mutants of VAMP/Synaptobrevin or Syntaxin, indicating that it results from impaired membrane trafficking. Importantly, the phenotype is also enhanced by mutations in genes for wingless and components of the Notch signaling pathway, suggesting that these signaling pathways were disrupted. Finally, this phenotype was used to conduct a screen for interacting genes, uncovering two Notch pathway components that had not previously been linked to wing development. It is concluded that SNARE-mediated membrane trafficking is an important component of wing margin development and that dosage-sensitive developmental pathways can act as a sensitive reporter of partial membrane-trafficking disruption (Stewart, 2001).

The Syntaxin, VAMP, and SNAP-25 families of proteins are proposed to target and fuse transport vesicles with specific membrane compartments. The SNARE complex is a parallel four-helix bundle with one helix contributed by each of Syntaxin and VAMP and two contributed by SNAP-25 (Sutton, 1998). The formation of a trans-membrane complex, with VAMP on the transport vesicle and Syntaxin and SNAP-25 on the target membrane, is thought to lead to the fusion of the two membranes, resulting in a cis-membrane complex. It follows that the cis-residing protein complexes need to be broken apart to make those proteins available for further trans-complex formation. This complex breakdown occurs under the action of N-ethylmaleimide-sensitive fusion protein (NSF), an ATPase. NSF contains two nucleotide binding domains and demonstrable ATPase activity. Structural analyses have shown that NSF forms a hexamer in vivo. NSF is a homolog of the yeast gene SEC18 and analysis of SEC18 function also reveals its requirement for intracellular membrane transport. NSF-dependent ATP hydrolysis is required to disassemble SNARE complexes, although it is not required for the fusion step. Thus the role of NSF in vesicular transport appears to be primarily one of priming vesicles for fusion and dissociation of SNARE complexes to permit their recycling (Stewart, 2001).

In Drosophila there are two homologs of NSF: dNSF1 and dNSF2 (NEM-sensitive fusion protein 2). dNSF1 is the gene product of comatose and is primarily found in neurons, whereas dNSF2, in addition to being neuronally expressed, is broadly expressed within imaginal discs, salivary glands, and the ring gland (Boulianne, 1995). Thus, dNSF2 is the most likely isoform to contribute to intracellular trafficking in nonneuronal tissue. Despite their proposed role in most intracellular trafficking events, in vivo studies of SNARE proteins have concentrated on two main systems: the budding yeast and calcium-triggered exocytosis in neurons. Relatively little attention has been given to other in vivo contexts in which the SNARE proteins are likely to have important roles. For example, in signaling pathways it is self-evident that transmembrane receptors and ligands need to be delivered to the plasma membrane, although few studies have been devoted to specifically studying the role of SNARE proteins in this process and their potential influence on the strength of intracellular signaling (Stewart, 2001).

To investigate the role of SNARE proteins within a defined developmental process, advantage was taken of the key role of NSF in membrane-transport processes. Specifically, a dominant negative form of dNSF2 was expressed in wing imaginal discs and this was shown to disrupt proper wing margin formation. This phenotype is enhanced in trans-heterozygous combinations of mutant alleles of the SNARE proteins syntaxin or synaptobrevin, further supporting a role for SNAREs in this process. Using genetic and immunocytochemical analysis it has been shown that this phenotype can be attributed to a failure in the signaling pathways that normally govern wing margin development. Thus, SNARE-dependent transport mechanisms are critical to wing formation and their manipulation may provide new insights into the mechanisms controlling developmentally important signaling pathways (Stewart, 2001).

To investigate the function of SNARE-dependent transport mechanisms in Drosophila point mutants in the ATP-binding region of the D1 domain of dNSF2 were constructed. Each nucleotide-binding subdomain of NSF contains consensus ATP-binding domains known as the Walker A and Walker B motifs. The DEAD box of the Walker B motif is conserved in a large number of ATP-dependent enzymes and was first identified in RNA helicases that use ATP hydrolysis to unwind RNA prior to translation. This motif binds the Mg2+ ion that coordinates the phosphates of ATP for hydrolysis. In RNA helicases replacement of the glutamate residue within the modified DEAD box (DEID) eliminates ATP hydrolysis without affecting ATP binding). In mammalian NSF a similar substitution within that protein’s DEID box, E329Q, reduces ATPase activity and NSF-dependent Golgi transport activity. NSF has been shown to form hexamers and, when mixed with wild-type protein NSFE329Q, forms hexamers that also lack ATPase activity, leading to a dominant negative effect. Drosophila NSF2 shows 59% overall amino acid identity with CHO NSF and nearly 100% conservation within the ATP-binding p-loop and DEID box of the D1 domain. Thus the structural and functional properties of the dNSF2 ATPase domains are very likely to be identical to those previously defined in RNA helicases and mammalian NSF, and mutation of the glutamate residue with the Drosophila DEID box motif should also impair the ATPase activity of the protein (Stewart, 2001).

A NSFE/Q construct was created with a glutamate-to-glutamine substitution at position 326 of the dNSF2 D1 domain. In two separate ATPase assays it was found that the NEM-sensitive ATPase activity of NSFE/Q is 47.5% and 57.1% that of dNSF2WT. The mean ATPase activity is 15.2 nmol Pi/microg/h for the wild-type protein and 7.8 nmol Pi/microg/h for the mutant protein. The remaining ATPase activity in NSFE/Q may be attributable to the second ATPase site within the D2 domain of the protein (Stewart, 2001).

To express the mutant dNSF2 transgenic flies were created carrying UAS-NSFE/Q and UAS-NSFWT constructs for use in the Gal4-UAS expression system. C96-Gal4 is expressed in developing wing discs in a pattern that is similar to, though slightly broader than, wing margin proteins such as Wingless. When UAS-NSFE/Q is driven by C96-Gal4, loss of wing margin is observed. The expression of NSFWT does not cause any visible phenotype, indicating that simple overexpression of dNSF2 in the wing margin is not a cause of the phenotype (Stewart, 2001).

The observation that NSFE/Q causes loss of wing margin implies that SNARE-dependent transport is important for wing margin formation. To test this further mutant alleles of synaptobrevin and syntaxin, two well-characterized SNARE proteins, were used to determine whether they would enhance the wing phenotype. Indeed, all trans-heterozygous combinations of NSFE/QC96 with synaptobrevin or syntaxin loss-of-function alleles enhance the wing margin phenotype, thus providing further evidence of the involvement of SNARE proteins in wing margin development (Stewart, 2001).

The wing phenotype observed is similar to that observed with mutant alleles of Notch and Wingless signaling pathway genes. To determine whether components of these pathways could be contributing to the NSFE/QC96 wing phenotype the protein pattern of Wingless in third-instar imaginal wing discs was first examined and a striking effect on the distribution of Wingless was observed. In control discs Wingless appears as a three- to four-cell-wide stripe across the wing disc, whereas in discs expressing the mutant dNSF2 Wingless appears very narrow and patchy. Wg expression was then examined using a Wg-lacZ reporter construct and an incomplete pattern of Wingless expression was found, as was observed for the Wingless protein (Stewart, 2001).

Because Wg is a secreted protein Wg was examined under higher magnification using confocal microscopy to determine directly whether Wg secretion was impaired. In control discs there is punctate Wg staining, indicative of Wg secretion, in the tissue surrounding the narrow stripe of wing margin cells. In the regions of the mutant discs that are immunoreactive for Wg, punctate staining is seen surrounding the positive cells. However, the Wg signal is much stronger in those cells and confocal sectioning of the cells has revealed the accumulation of Wg at the apical region of the wing margin cells. These data indicate that mutant NSFE/Q impairs, but does not eliminate, Wingless secretion. Because Wingless expression is impaired and its activation is under the control of Notch signaling, the distribution patterns of other proteins involved in the Notch pathway were examined. Notch protein distribution was examined directly using a monoclonal antibody that recognizes the extracellular domain of Notch. At low magnification there is no major difference between mutant and control samples, with the antibody labeling the cell membranes in the wing pouch. However, at higher magnification, in addition to the membrane staining, immunoreactive puncta were also observed within the cells of the mutant wing disc that were not readily observed in the control discs. These puncta likely represent improperly sorted Notch proteins (Stewart, 2001).

The distribution of Cut, Delta, and Achaete, coded for by genes that are downstream of Notch activation in the wing margin signaling pathway, was examined -- all of these markers were disrupted in NSFE/QC96 larval wing discs. Cut is normally found in a pattern that overlaps with Wg along the presumptive wing margin, whereas in the mutant discs it appears in a broken pattern similar to that of Wg. Delta is normally expressed in two parallel bands along the D/V boundary and this pattern is thought to be the result of the downregulation of Delta in boundary cells by Cut and the upregulation of Delta in flanking cells by Wingless. In NSFE/QC96 wing discs the expression of Delta is reduced and the two parallel bands appear to be collapsed into a single band along the boundary. Achaete is normally expressed in two broad bands parallel to the D/V boundary in the anterior compartment of the wing disc defining a proneural cluster. In the NSFE/QC96 discs this pattern is severely disrupted: the number of Achaete-expressing cells is reduced and there is complete absence of Achaete in some areas (Stewart, 2001).

A similar pattern of disruption was found when lacZ reporter constructs were used to examine the expression of neuralized and vestigial, two other genes in the Notch pathway. neuA101-lacZ is normally detected in sensory organ precursors (SOPs) located in two rows of single cells parallel to the D/V boundary in the anterior compartment of late third-instar wing discs. In the mutant discs this pattern is disrupted and lacking in some areas along the wing margin, while SOPs elsewhere in the disc are unaffected. Similarly, vgBE-lacZ expression is disrupted. In wild-type discs vgBE-lacZ expression is seen in the D/V and anterior/posterior (A/P) boundaries, whereas in the mutant discs the expression in the D/V boundary is disrupted (Stewart, 2001).

Interestingly, expression in the A/P boundary remains, although the C96-Gal4 expression pattern overlaps this region. Taken together these results demonstrate that NSFE/Q affects the distribution and expression of several downstream components of the Notch signaling pathway. To confirm the effect of NSFE/Q on Notch signaling loss-of-function alleles of several genes in the Notch and Wingless pathways were examined for their ability to enhance the adult wing phenotype caused by NSFE/Q expression. In that Notch signaling is known to be highly sensitive to haploinsufficiency of interacting gene products, it was reasoned that these loss-of-function alleles should show genetic interaction. Two alleles of Notch and one each of Delta, Serrate, wingless, and fringe were examined and it was found that they all enhanced the wing phenotype in transheterozygous combination with NSFE/QC96. The severity of the phenotype produced by each allele was similar, although Df(1)N8, a null allele of Notch, did produce a more severe phenotype than did Nnd-3, a hypomorphic allele. With the exception of Df(1)N8, none of these mutants produces a wing-nicking phenotype when examined alone as heterozygotes. Thus, the enhancement of the adult wing phenotype by mutants in the Notch pathway supports the conclusion that NSFE/Q expression causes a defect in wing margin signaling pathways (Stewart, 2001).

Finally, the ability of UAS constructs of Notch, Delta, and Serrate to rescue the wing phenotype generated by NSFE/QC96 were tested. Complete rescue could be obtained with both Notch and Delta constructs. Serrate generally appears to rescue less well than do the other constructs because minor nicks in the distal wing persist. Furthermore, no rescue effect was seen when crosses were made to UAS-lacZ lines, indicating that competition for Gal4 protein is not responsible for rescue of the phenotype. The observation that UAS-Notch and UAS-Delta can completely rescue the NSFE/Q wing phenotype further indicates that the mutation affects intracellular transport and does not create a cell-lethal phenotype because cell lethality should not be rescued by Notch or Delta (Stewart, 2001).

Having established that NSFE/Q disrupts signaling at the wing margin in a SNARE-dependent manner, and that enhancement of the phenotype can be attributed to haploinsufficiency of known genes, it was asked whether the wings of the NSFE/QC96 flies could be used as a sensitized background to find novel genes involved in wing margin formation. To this end a small-scale screen was conducted for enhancers and suppressors of the phenotype. In the first set of experiments specific alleles of two genes were tested: big brain and porcupine. These have been shown to be important in Notch and Wingless signaling in other developmental contexts but have not previously been known to be important for wing margin development. In the NSFE/QC96 background it was found that both mutant alleles of these genes enhance the NSFE/QC96 wing margin phenotype. This result is the first report of the involvement of these two genes in wing margin development and suggests that NSFE/QC96 wings provide an ideal sensitized background for conducting forward genetic screens to identify novel genes involved in wing margin development (Stewart, 2001).

In the second set of experiments a test was performed for genetic interactions with deficiencies that uncover most of the Drosophila genome. Of the deficiencies tested, 33 interacting lines were identified that enhanced or suppressed the wing margin phenotype. The further characterization of these loci may reveal novel components of the SNARE or Notch and Wg signaling pathways (Stewart, 2001).

In view of current membrane-trafficking models, it is expected that expressing NSFE/Q impairs the ability of NSF to dissociate cis-SNARE complexes, making fewer SNARE proteins available for functional transmembrane complex formation and thus reducing intracellular transport. These results provide solid evidence that SNARE proteins are important in wing margin formation. This implies that the mutant NSF must suppress but not block all membrane traffic. The disruption of molecular markers, such as Wg, Delta, Achaete, Cut, Vestigial, and Neuralized, indicates that the NSFE/Q wing phenotype observed is the result of impaired signaling at the developing wing margin. This is consistent with data presented in other studies that manipulated the signaling pathway directly. For example, reduction of Notch activity with Nts alleles can lead to reduced and patchy Wingless expression. Wingless and Cut expression is also reduced and patchy in Notch mutant wing discs. Stripes of Delta and Serrate that normally flank the D/V boundary collapse into a single stripe along the margin in Nts alleles exposed to restrictive temperature. In NSFE/QC96 wing discs, changes in Wingless, Cut, and Delta patterns were observed that are similar to those that occur when Notch activity is directly manipulated; therefore, it seems that NSFE/Q expression phenocopies genetic mutants of Notch (Stewart, 2001).

Because the Notch and Wingless signaling pathways are so intertwined in controlling wing margin development it is difficult to determine whether the dNSF2 mutants cause a primary defect in one or the other of these proteins, although it seems likely that there are parallel effects on both. The experiments show not only a direct impairment of Wingless trafficking but also that Wg-lacZ expression is disrupted. The latter suggests that an upstream activator of Wingless expression is impaired (although this could be Wingless itself). It was found that Notch subcellular localization is disrupted and that a Wg-independent target of Notch signaling, the vestigial boundary enhancer, is also disrupted. Because this vestigial enhancer element is thought to be under the sole control of Notch this supports the idea that NSFE/Q has a direct effect on Notch signaling. Thus vgBE-lacZ expression data strongly suggest direct effects on both Wg and Notch. Moreover, because these molecules are at the top of the hierarchy controlling signaling at the wing margin this provides the likely explanation for the disruption of downstream targets of these genes (Stewart, 2001).

The molecular and genetic interactions that regulate developmentally important signaling pathways are important for defining the final outcome of the signaling cascade. For example, previous studies have identified several molecules, including Fringe, Big Brain, and Numb, that are proposed to influence Notch signals. Because the SNARE proteins interact with many protein partners, some of which are proposed to regulate their availability (e.g., Syntaxin’s interaction with rop/nsec-1), these data indicate that regulation of SNARE-dependent transport steps may represent an additional mechanism by which signal transduction pathways can be modulated during development (Stewart, 2001).

Functional genomic analysis of the Wnt-Wingless signaling pathway

The Wnt-Wingless (Wg) pathway is one of a core set of evolutionarily conserved signaling pathways that regulate many aspects of metazoan development. Aberrant Wnt signaling has been linked to human disease. In the present study, a genomewide RNA interference (RNAi) screen in Drosophila cells was used to screen for regulators of the Wnt pathway. 238 potential regulators were identified that include known pathway components, genes with functions not previously linked to this pathway, and genes with no previously assigned functions. Reciprocal-Best-Blast analyses reveal that 50% of the genes identified in the screen have human orthologs, of which 18% are associated with human disease. Functional assays of selected genes from the cell-based screen in Drosophila, mammalian cells, and zebrafish embryos have demonstrated that these genes have evolutionarily conserved functions in Wnt signaling. High-throughput RNAi screens in cultured cells, followed by functional analyses in model organisms, prove to be a rapid means of identifying regulators of signaling pathways implicated in development and disease (DasGupta, 2005).

Epistasis experiments allowed the placing of selected candidate genes in a hierarchy either upstream or downstream of known positive and negative regulators. Specific examples of three potential regulators that were identified in the screen include two known transcription factors, DP (dimerization partner) and Lilli (Lilliputian), and a novel gene, CG5402, as activators in the Wnt pathway in the primary screen. In vitro epistasis experiments in clone 8 cells placed each of the three candidate genes at three distinct steps in the pathway. CG5402 acts upstream of Axin but downstream of Wg, Fz, or Arr; DP functions downstream of Axin and Ck1alpha but upstream of ß-cat; and Lilli functions downstream of ß-cat. It is interesting that lilli encodes an HMG-box transcription factor. lilli has also been shown to interact genetically with arm, which further corroborates its role in the Wnt pathway. It is important to note that lilli interacts genetically with members of several signaling pathways, including the receptor tyrosine kinase (RTK)/Ras and the Decapentaplegic (Dpp) pathway, which underscores the power of the RNAi approach in assigning functions to genes with pleiotropic functions that may be critical factors involved in cross talk between multiple signaling pathways (DasGupta, 2005).

Overall, this epistasis analysis of the potential positive regulators failed to place any new gene between Wg-Fz-Arr ligand-receptor complex and Dsh, even though known intermediates such as Arr and Fz were placed between Wg and Dsh by this method. Preliminary epistasis analysis of most genes encoding potential positive regulators revealed that they affect the pathway downstream of Dsh. These genes were further categorized into those that acted upstream or downstream of genes involved in phosphorylation or degradation of ß-cat (axin, ck1alpha, and slmb) and those that acted downstream of ß-cat. Altogether, the in vitro epistasis studies provide a starting point from which to investigate the mechanism of action of candidate genes identified in the screen (DasGupta, 2005).

The candidate genes that increased reporter activity when their expression was inhibited were further tested in order to categorize them into specific functional groups. First, it was determined whether RNAi of potential negative regulators could ectopically activate the TOP-Flash, Pangolin responsive reporter in the absence of Wg stimulus. Of the 129 negative regulators tested, 63% (83 out of 129) activated reporter activity after dsRNA-mediated knockdown, which suggests a potential role in the regulation of basal Wg activity in a cell. Genes in this category could be either directly or indirectly acting at the level of regulation of Arm/ß-cat stability and/or phosphorylation or at the level of target gene regulation. RNAi knockdown of the remaining 47 genes promoted expression of the TOP-Flash reporter only in the presence of Wg, which suggests a role specifically in Wg-stimulated cells. This second class of genes could be functioning either at the level of ligand-receptor regulation or receptor-mediated endocytosis, or they may be involved in the regulation of the stable pool Arm/ß-cat that is present only in a stimulated cell. This class includes regulators, such as nkd and Dlp, that have been shown to regulate the intracellular and extracellular trafficking of Wg, respectively (DasGupta, 2005).

Whether decreased expression of candidate negative regulators required downstream effectors such as Arm and Pan to activate the Wnt-ß-cat-responsive reporter gene was tested. Cells with arm or pan dsRNA were tested together with individual dsRNAs specific for selected negative regulators. With the exception of two genes, CR31616 and CG4699, Arm and Pan were indeed required for activation of the TOP-Flash reporter (in the absence of Wg stimulus), which placed them epistatically downstream of most negative regulators (DasGupta, 2005).

To further test the relevance of the genes identified as potential regulators of the Wnt pathway, selected candidates were overexpressed in cells in culture and in Drosophila wing imaginal discs in vivo. One of the candidate genes encoded the small GTPase Rab5. Rab5 has a central role in early endocytic trafficking by directing the budding of endocytic vesicles from the plasma membrane, their movement along microtubules, and their fusion with sorting endosomes. Rab5 has been implicated in controlling the shape of the long-range gradient of the transforming growth factor superfamily member, Dpp, in the Drosophila wing by regulating the endocytosis of ligand-receptor complex. Rab5-interacting proteins, such as APPL1 and APPL2, as well as other proteins involved in the formation of clathrin-coated vesicles (CCVs) (such as Eps15, epsin, and ß-arrestin 2), can undergo nucleocytoplasmic shuttling and can interact with nuclear transcription factors to regulate expression of target genes. These studies indicate that the endocytic machinery may be directly involved in nuclear signaling functions as well (DasGupta, 2005).

In the screen, RNAi-mediated depletion of Rab5 only promoted reporter activity if cells were also stimulated with Wg. Conversely, cotransfection of increasing amounts of Rab5 cDNA together with the Wg cDNA in Drosophila cells displayed a dose-dependent repression of Wg-mediated TOP-flash reporter activity. The effect on STF reporter activity in mammalian 293T cells upon Rab5 overexpression and small interfering RNA (siRNA)-mediated knockdown was similar to the effects obtained in fly cells in culture (DasGupta, 2005).

To assess whether Rab5 could similarly affect Wg signaling in vivo, the GAL4-UAS (upstream activation sequence) system was used to drive the expression of wild-type rab5 in the Drosophila wing imaginal disc with a specific wing-margin driver, C96-GAL4. The expression of senseless, a proneural gene that is a target of the Wg signaling pathway at the wing margin (straddling the dorsal-ventral boundary) was used as a readout for pathway activity. Overexpression of Rab5 (C96GAL4-UASRab5WT) resulted in a partial to complete loss of senseless expression at the wing margin compared with that in control discs (C96GAL4). Expression of wg itself was not affected. Nor was expression of senseless in the proneural clusters at the distal regions of the wing pouch. Because Rab5 has been implicated in receptor-mediated endocytosis and degradation of morphogenetic signals, it was thought that overexpression of Rab5 might influence endocytosis of the endogenous Wg protein and thus might alter signaling activity at the plasma membrane, but antibody staining against extracellular Wg revealed no difference in the levels of secreted Wg protein between regions that displayed high and low levels of senseless expression. Thus, Rab5 appears to have a role in the control of Wg signaling activity in which it acts to inhibit Wg-dependent activation of target genes (DasGupta, 2005).

The observations suggest that overexpression of Rab5 does not affect the extracellular distribution of Wg protein per se. It is possible that Rab5 could be perturbing the distribution of the receptors and coreceptors Fzd2 and Arrow (Lrp6). However, any significant change in the distribution of receptors is unlikely based on this analysis of extracellular Wg and previous studies that have demonstrated the role of Frizzled-2 receptor in regulating extracellular distribution of Wingless and shaping the Wg gradient in the wing imaginal disc. Nonetheless, subtle changes at the level of receptors and/or coreceptors cannot be ruled out. Alternatively, Rab5 could be regulating trafficking of the stabilized pool of Arm/ß-cat, which is present only in a Wg-induced cell and thus affecting the downstream Wingless readout as judged by antibody staining for Senseless (DasGupta, 2005).

Functional antagonism of Wingless and DWnt-4

In Drosophila, wingless and DWnt-4 are two physically clustered Wnt genes, that are transcribed in overlapping patterns during embryogenesis and, in several instances, are controlled by the same regulatory molecules. To address the question of the functional relationship of wingless and DWnt-4, how embryonic cells respond when they are exposed, simultaneously or not, to the encoded Wnt signals was examined. DWnt-4 has the capacity to antagonize Wingless signaling both in the Drosophila ventral epidermis and in a heterologous system, the Xenopus embryo. Evidence indicates that DWnt-4 inhibits the Wingless/Wnt-1 signaling pathway upstream of the activation of transcriptional targets (Gieseler, 1999).

The functional relationship between the two Wnt products was examined by comparing their ability to induce embryonic axis duplication in Xenopus. To do this, different concentrations of mRNA, encoding either Wg or DWnt-4, were injected into the vegetal region of ventral blastomeres of four-cell stage Xenopus embryos. As expected, providing WG mRNA results in the development of a secondary axis. This effect occurs in a dose-dependent manner. Five picograms of the WG message are sufficient to induce in 65% of injected embryos a fully developed supernumerary axis including cement glands, a marker of most dorso-anterior structures. In contrast, injection of as much as 6 ng of DWnt4 mRNA fails to induce embryonic axis duplication. To determine whether DWnt-4 influences the axis-inducing activity of Wg, mRNAs encoding the two Wnt molecules were co-injected into the ventral marginal zone of cleavage-stage embryos. One ng of DWnt-4 message decreases the frequency at which Wg induces a complete ectopic axis, since embryos generally develop an incomplete axis devoid of cement gland. Embryos injected with 2.4 ng of DWnt-4 mRNA never show complete axis duplication. Further increases in the amount of DWnt-4 further inhibits the phenotype induced by Wg; with 6 ng of DWnt-4 mRNA, the formation of a secondary axis is blocked in most embryos. Taken together these results indicate that DWnt-4 antagonizes in a dose-dependent manner the ability of Wg to cause axis duplication. DWnt-4 also counteracts Wnt-1 signaling upstream of the activation of target genes in the Spemann Organizer (Gieseler, 1999).

In contrast with ventral injections, directing DWnt-4 mRNA into the dorsal marginal zone of a four-cell stage Xenopus embryo inhibits formation of the endogenous axis, resulting in the disappearance of the anterior-most dorsal structures. Of the embryos injected with 6 ng of DWnt-4 mRNA, 90% fail to develop head structures including eyes and cement glands. In Xenopus, the formation of axial structures, including the head, requires beta-catenin activity, which results in the activation of Spemann Organizer genes in the dorsal territories of the early gastrula. To investigate the mechanism by which DWnt-4 inhibits dorsal patterning, the expression of chd, gsc, otx2 and Xnr3 were examined. Whole mount in situ hybridization on embryos dorsally injected at the four-cell stage and left to develop until stage 10, show that DWnt-4 does not alter the transcription patterns of the Organizer genes, nor does injection into the ventral marginal zone induce expression at ectopic sites. Thus, unlike products of the Wnt-1 class of genes, DWnt-4 does not have the capacity to influence the transcription of Organizer genes. These results indicate that the repression of dorsal axis formation by DWnt-4 presumably does not result from deficient formation of the Spemann Organizer, but from changes occurring later in dorsal development (Gieseler, 1999).

In the process of patterning the embryonic ventral ectoderm, Wg has at least two temporally distinct roles. A paracrine function of Wg is first required for the maintenance of engrailed (en) transcription in the adjacent posterior two rows of cells. This activity results in the formation and stabilization of parasegmental boundaries (cell-stabilization phase). During this time, Wg is also required for establishment of the diverse array of denticle types within the anterior half of the parasegment. Later wg function is essential for maintaining its own transcription and for specifying the cell fate that secretes naked cuticle within the posterior half of the parasegment (cell-specification phase). To determine whether DWnt-4 can modulate Wg signaling in these processes, the ventral cuticle defects resulting from DWnt-4 overexpression using the UAS/Gal4 system were examined. Several UAS-DWnt-4 transgenic lines were established and crossed with the daughterless (da) Gal4 driver that provides ubiquitous expression of UAS transgenes beginning early in embryogenesis. Embryos bearing one copy of UAS-DWnt-4 transgene show mild but consistent cuticular defects consisting of ectopic denticles within the region of naked cuticle. Posterior segments are always more affected than anterior ones and segmentation and diversity in denticle morphology are not entirely lost. Stronger phenotypes were obtained by increasing DWnt-4 expression through two copies of the UAS-DWnt-4 transgene. Under these conditions, the naked cuticle is often completely lost and replaced with denticle-decorated cuticle, especially in the posterior-most segments. The phenotype produced by ubiquitous DWnt-4 expression is opposite that of the naked cuticle mutant phenotype, in that it is produced by ectopic Wg. Providing DWnt-4 ubiquitously thus results in a phenotype that resembles that induced by loss of wg activity. When wg function is removed throughout embryogenesis, naked cuticle does not form; compartment boundaries are lost and all denticles display similar morphology. However, when wg function is removed at stage 10 in development, naked cuticle is still affected, but not completely lost, while denticle diversity is maintained. The phenotypes observed in da-Gal4-UAS-DWnt-4 embryos resemble the latter, suggesting that DWnt-4 antagonizes Wg activity during the late cell-specification phase reproducibly observed (Gieseler, 1999).

Two series of experiments were carried out to determine whether the phenotypes produced by DWnt-4 in the ventral ectoderm result from a reduction in Wg signal transduction. First, the transcription patterns of Wg target genes were examined. The expression of en marks the activity of Wg during the early phase, while the expression of wg itself marks the late phase of Wg activity through its autoregulation. In stage 9 da-Gal4-UAS-DWnt-4 embryos, expression of the two target genes is not significantly altered. Thus, wg and en presumably still function in the definition and stabilization of parasegmental boundaries; this is consistent with the previous observation that ubiquitously provided DWnt-4 does not abolish the segmental organization of the embryo. In stage 11 embryos, since the expression of wg and en is no longer interdependent and Wg is required for the maintenance of its own transcription, wg expression in the ventral ectoderm becomes severely reduced. Consistent with the stronger defects observed in posterior cuticle, a more pronounced inhibition of wg in posteriormost segments has been reproducibly observed. At the same time, the En protein distribution remains unaffected. Second, it was asked whether DWnt-4 has the ability to influence intracellular accumulation of the Armadillo (Arm)/beta-catenin protein, a key mediator of the Wg/Wnt-1 signaling. Cytoplasmic Arm is stabilized in response to signaling by Wg, allowing its translocation into the nucleus where it acts in association with the Pangolin/dTCF factor, a transcriptional effector of the pathway. Loss of wg function results in a reduction of the amount of Arm, when compared to wild type. DWnt-4 overexpression also induces a clear, though less pronounced, reduction in the levels of Arm. Thus, signaling by DWnt-4 negatively interferes with the Wg signaling pathway upstream of the Arm stabilization step (Gieseler, 1999).

Wingless undergoes a hydrophobic modification and is targeted to lipid rafts, a process that requires porcupine

Wnt signaling pathways regulate many developmental responses; however, little is known about how Wnt ligands function on a biochemical level. Recent studies have shown that Wnt-3a is palmitoylated before secretion. Drosophila Wingless also undergoes a lipid modification. Lipidation occurs in the endoplasmic reticulum and is dependent on Porcupine, a putative O-acyltransferase. After modification, Wingless partitions as a membrane-anchored protein and is sorted into lipid raft detergent-insoluble microdomains. Lipidation, raft targeting, and secretion can be blocked by the addition of 2-bromopalmitate, a competitive inhibitor of O-acyltransferase activity. Based on these results a model is proposed whereby lipidation targets Wingless to secretory vesicles that deliver the ligand to specialized microdomains at the cell surface where it can be packaged for secretion (Zhai, 2004; full text of article).

Wntless, a conserved membrane protein dedicated to the secretion of Wnt proteins from signaling cells

Cell-cell communication via Wnt signals represents a fundamental means by which animal development and homeostasis are controlled. The identification of components of the Wnt pathway is reaching saturation for the transduction process in receiving cells but is incomplete concerning the events occurring in Wnt-secreting cells. This paper describes the discovery of a novel Wnt pathway component, Wntless (Wls/Evi) and shows that it is required for Wingless-dependent patterning processes in Drosophila, for MOM-2-governed polarization of blastomeres in C. elegans, and for Wnt3a-mediated communication between cultured human cells. In each of these cases, Wls is acting in the Wnt-sending cells to promote the secretion of Wnt proteins. Since loss of Wls function has no effect on other signaling pathways yet appears to impede all the Wnt signals analyzed, it is proposed that Wls represents an ancient partner for Wnts dedicated to promoting their secretion into the extracellular milieu (Bänziger, 2006).

The development of complex multicellular organisms relies to large extent on the ability of cells to communicate via extracellular signals. One family of such signaling molecules comprises the Wnt proteins. They direct growth and cell fates in processes as diverse as embryonic segmentation, limb development, and CNS patterning. Recent evidence also implicates Wnt signaling in the postembryonic regulation of stem cell number and stem cell differentiation in mammalian skin and colon epithelia as well as in the hematopoietic system. Perturbations in Wnt signaling promote both human degenerative diseases and cancer (Bänziger, 2006).

Due to the immense importance of this class of signaling proteins, there is a profound interest to understand how Wnt proteins are synthesized, transported, received, and their signal transduced. A powerful way to reach this goal and identify the components involved is to study Wnt signaling in genetic model systems, such as Drosophila melanogaster or Caenorhabditis elegans (Bänziger, 2006).

A defining member of the Wnt family is the product of the Drosophila gene wingless (wg). In the embryo, Wg acts as a short-range inducer to pattern the epidermis: it is secreted by subpopulations of cells within each segment and received by their neighbors, which respond by expressing particular target genes. Loss of Wg function causes a segment-polarity phenotype, in which the larval epidermis forms a lawn of ventral denticles and lacks naked cuticle between the segmental denticle belts. At later stages, during larval development, Wg also acts at longer distances and in a concentration-dependent manner. In the wing imaginal disc, for example, Wg is produced by cells of the prospective wing margin and spreads symmetrically along the dorsoventral axis to define nested patterns of target-gene expression (Bänziger, 2006).

Many of the components of the cellular machinery responsible for transducing Wnt signals from the cell surface to the nucleus have been identified: receptors of the Frizzled and LRP families, the cytosolic proteins Dishevelled, Glycogen Synthase Kinase 3, Axin, APC, and β-catenin, the transcription factors TCF/Lef as well as many of their nuclear cofactors. Less is known about the processes controlling the secretion of Wnt and its movement through the territory between source and target cells. Wnts are secreted as glycoproteins and contain as a hallmark 22 cysteine residues at conserved positions in their amino acid sequence. Some of these cysteine residues might be engaged in the folding of Wnt proteins by disulfide bond formation. Purification and precise biochemical characterization of a Wnt protein succeeded only recently and led to the discovery that the first conserved cysteine residue (C77 in Wnt3a) is palmitoylated. A candidate acyltransferase to mediate this lipid addition is encoded by the Drosophila porcupine (porc) and C. elegans mom-1 genes. This acyltransferase is the only molecularly known pathway component that acts in Wnt-sending cells only. Glypican members of heparan sulfate proteoglycans (Dally and Dally-like in Drosophila) also act in Wnt-sending cells, but they mainly function in transporting Wnt proteins as well as other signaling molecules toward target cells (Bänziger, 2006).

To complete the identification of components involved in Wg signaling, an in vivo screening system was established that permits the isolation of Wg pathway genes that have so far resisted isolation by genetic means. This study describes the discovery of wntless (wls), a novel segment-polarity gene in Drosophila, and shows that it is required for Wg signaling throughout development. In an accompanying report, Bartscherer (2006) refers to this gene as evenness interrupted (evi). This characterization of wls/evi revealed that it encodes a novel, yet evolutionarily and functionally conserved, transmembrane protein that, intriguingly, is only required in Wg-sending cells. Analogous to Dispatched (Disp), a protein functioning in the release of Hedgehog (Hh) proteins, Wls is necessary for the secretion of Wnt proteins. However, unlike Disp, this function of Wls is not limited to the lipid-modified form of its target signal. Wls represents only the second component dedicated to the Wnt pathway that acts exclusively in signal-sending cells. Its discovery adds an unanticipated level of complexity that must be factored into concepts for Wnt-mediated cell-cell communication (Bänziger, 2006).

Genetic experiments showing that Wls is only required in Wg-sending cells, together with the finding that Wnt3a activity requires Wls function in cultured producer but not responder cells, placed Wls upstream of the Wnt/Wg receptors LRP/Arrow and Frizzled and the downstream signal-transduction cascade. These results on their own could not discriminate between the possibilities that in the absence of Wls Wnt ligands are sent out in an inactive form or are not sent out at all. However, the accumulation of Wg protein in wls mutant cells, together with the severe reduction of Wg/Wnt levels in tissue or medium surrounding mutant cells, indicated that the prime cause of the wls mutant phenotype must be a shortage of Wg/Wnt ligands in their target field. In the absence of Wls, Wg/Wnt proteins are predominantly retained in the producing cells and are thus unable to move in adequate quantities to nearby cells (Bänziger, 2006).

Some residual Wg-dependent patterning still occurs in wls mutants, such as reduced expression of the low-threshold Wg target gene Distal-less. Hence, a small fraction of Wg protein must be escaping, below the limits of immunohistochemical detection. Consistent with this assumption, it was found that wls homozygous mutant larvae did not survive a 50% reduction of Wg expression (by removal of one wg gene copy). All these observations suggest that the function of Wls is to ensure the efficient secretion of Wnts at physiologically effective levels. It is important to emphasize that these results do not exclude an additional role of Wls in conferring activity to Wnt ligands, a role that could be masked by the secretion phenotype (Bänziger, 2006).

The identification of a multitransmembrane protein required for the secretion of Wnt signaling proteins is reminiscent of the analogous discovery of Dispatched. In the absence of Disp, Hh signals are not secreted from cells and are unable to reach neighboring cells. In both pathways, the ligands are lipid-modified and tightly associated with cell membranes, yet are able to move over considerable distances in vivo. The predominant view of how Disp functions is that it releases Hh ligands from their membrane association. In apparent contrast to Disp, Wls seems to function earlier in the pathway, as it is already required for the transport of Wnt ligands to the cell surface, not just for their release from the membrane (Bänziger, 2006).

Other proteins implicated in the release of signaling proteins are the glypicans Dally-like (Dlp) and Dally, which belong to the family of heparan sulfate proteoglycans (HSPGs). Both Dlp and Dally bind Wg protein, and if their function is perturbed (by removing Sulfateless, an enzyme required for their synthesis) Wg protein no longer accumulates at the surface of Wg-secreting cells and animals exhibit wg loss-of-function phenotypes. These phenotypes are superficially similar to those caused by the lack of Wls function, raising the possibility that Wls might function like Dlp and Dally or play an essential role in their biosynthesis. However, in embryos with perturbed glypican function, Wg protein is undetectable at the surface of secreting cells because it is not stabilized there and released too readily. By contrast, when Wls function is perturbed, Wg/Wnt protein is poorly released, and its extracellular concentration is diminished rather than increased. Moreover, glypicans play a role in the movement of Wg from cell to cell, a role that was excluded for Wls by analysis of genetic mosaics. Together, the results indicate that Wls is required for proper intracellular transport of Wg/Wnt proteins to the cell surface (Bänziger, 2006).

If Wnt proteins use the general mode of intracellular conveyance, like other proteins destined for secretion, they may need special assistance to do so. For example, it is possible that Wnt polypeptides require a molecular chaperone function to reach a particular structural state for proceeding in vesicular transport from the ER to the Golgi apparatus. Wls might provide such a chaperone function, and in its absence Wnt ligands might not be admitted to public transportation. An intriguing parallel to such a chaperone function for Wnt signals is the recently identified chaperone for Wnt receptors: Boca and its mammalian homolog Mesd (Culi, 2003; Hsieh, 2003) are specifically required for the folding and stability of a subset of the low-density lipoprotein receptor (LDLR) family, which also comprises the Wg and Wnt receptors Arrow and LRP5/6. A high degree of client specificity is not unprecedented for molecular chaperone systems, but several observations argue against the view that Wls is a Wnt chaperone. First, for most chaperone systems analyzed, unfolded client proteins are targeted for degradation, rather than accumulation, in the absence of chaperone function. Second, the movement of such improperly folded polypeptides is blocked early along the secretory pathway, at the level of the ER, a phenotype not observed in cells lacking Wls function. And finally, the predominant location of Wls in the Golgi apparatus does not support a function in the ER for this protein (Bänziger, 2006).

Wnt proteins are posttranslationally modified by the addition of a palmitate group at a conserved cysteine residue. A candidate enzyme for catalyzing this lipid modification is encoded by the Drosophila porc gene. It is possible that lipid-modified Wnt proteins require a dedicated mechanism for their secretion and that Wls is part of such a mechanism. In principle, it is also possible that Wls is required together with Porc to palmitoylate Wg protein. However, Wls and Porc localize to different compartments of the secretory pathway, and the findings in mammalian cells indicate that Wnt3aC77A, which lacks the palmitoylation site, is still subject to regulation by Wls. Therefore, although a functional overlap between Porc and Wls function cannot be ruled out, the view is favored that Wls is acting subsequent to Porc in an aspect of the secretion of Wnt proteins that does not center on their palmitate modification (Bänziger, 2006).

Two recent observations that may relate to the mechanism of Wls action are particularly interesting. One concerns the discovery that a significant fraction of Wg protein inside secreting cells of the embryo is present in endocytic vesicles and sent back to the surface of such cells (Pfeiffer, 2002). The precise function of this ligand recycling is not yet known, but it may allow Wg protein to gain access to cellular compartments from which standard secretory proteins are excluded. A role that could be envisaged for Wls is to relieve Wg from this 'holding pattern' and allow it to ride public transportation out of the cell. Hence, loss of wls function would cause an abnormally high amount of Wg to cycle in this holding pattern, accompanied by a corresponding deficit in the extracellular milieu (Bänziger, 2006).

The other observation concerns lipophorin particles in Drosophila and the notion that such particles may serve as vehicles for the movement of lipid-modified ligands such as Wg or Hh (Panakova, 2005). An intriguing possibility would be if Wls is required for loading Wg protein onto such particles in a certain cellular compartment to which lipophorin particles and Wg have common access (Bänziger, 2006).

In a variation of these themes, Wnt proteins may have to cycle between different secretory or endocytic compartments before they can be secreted as active proteins. This cycling may endow them with posttranslational modifications or permit their charging onto lipophorin particles. Wls might be required to liberate Wnt proteins from this cycle and permit their subsequent secretion as mature signals (Bänziger, 2006).

Whatever the precise molecular mechanism by which Wls allows Wnt proteins to be secreted, it is remarkable that Wls functions so specifically on Wnt proteins. In a diverse set of experiments in which Wls function was reduced, spatially restricted, or abolished, no effects of Wls on other secreted proteins or cellular processes were detected. Equally astonishing is the observation that Wls is required for all members of the Wnt family that have been analyzed in this respect. In nematode and vertebrate systems, a subset of Wnt ligands has been found to signal via 'noncanonical' routes to control the polarities of individual cells; examples for such ligands are MOM-2 and Wnt5a, respectively. The current findings suggest that both polarity Wnts, as well as Wnt signals activating canonical transduction pathways, depend on Wls function for their respective activities. This apparent 'pan-Wnt promiscuity' argues for a fundamental relationship between Wls and structural aspects, as opposed to functional properties, of Wnt proteins. The presence of a single Wls ortholog in all metazoan genomes, and its absence in unicellular phyla and plants, indicates that Wnt signaling always depended on, and hence coevolved with, Wls function (Bänziger, 2006).

WLS retrograde transport to the endoplasmic reticulum during Wnt secretion

Wnts are transported to the cell surface by the integral membrane protein WLS (also known as Wntless, Evi, and GPR177). Previous studies of WLS trafficking have emphasized WLS movement from the Golgi to the plasma membrane (PM) and then back to the Golgi via retromer-mediated endocytic recycling. This study found that endogenous WLS binds Wnts in the endoplasmic reticulum (ER), cycles to the PM, and then returns to the ER through the Golgi. An ER-targeting sequence was identified at the carboxyl terminus of native WLS that is critical for ER retrograde recycling and contributes to Wnt secretory function. Golgi-to-ER recycling of WLS requires the COPI regulator ARF as well as ERGIC2, an ER-Golgi intermediate compartment protein that is also required for the retrograde trafficking of the KDEL receptor and certain toxins. ERGIC2 is required for efficient Wnt secretion. ER retrieval is an integral part of the WLS transport cycle (Yu, 2014).

Secretion of Wnt ligands requires Evi, a conserved transmembrane protein

Wnt signaling pathways are important for multiple biological processes during development and disease. Wnt proteins are secreted factors that activate target-gene expression in both a short- and long-range manner. Currently, little is known about how Wnts are released from cells and which factors facilitate their secretion. This study has identified a conserved multipass transmembrane protein, Evenness interrupted (Evi/Wls), through an RNAi survey for transmembrane proteins involved in Drosophila Wingless (Wg) signaling. During development, evi mutants have patterning defects that phenocopy wg loss-of-function alleles and fail to express Wg target genes. evi's function is evolutionarily conserved; depletion of its human homolog disrupts Wnt signaling in human cells. Epistasis experiments and clonal analysis place evi in the Wg-producing cell. These results show that Wg is retained by evi mutant cells and suggest that evi is the founding member of a gene family specifically required for Wg/Wnt secretion (Bartscherer, 2006).

How organisms control the spatial and temporal activation of signaling pathways remains an important question in biology. With the availability of whole genome sequences, one of the key challenges is the functional annotation of the encoded gene products. RNAi approaches have become a powerful genetic tool to systematically dissect cellular pathways. This study analyzed the consequences of depleting almost all known Drosophila transmembrane proteins for Wg signaling. The survey in cultured cells identified Fz, Fz4, and LRP6/Arr, transmembrane proteins known to positively regulate Wg signaling. It is interesting to note that both Fz and Fz4, but not Fz2, appear to be required in cells for Wg signal transduction in a nonredundant manner. Moreover, a previously uncharacterized locus, evi/CG6210, which is a member of an evolutionarily conserved gene family, has been identified as a novel regulator of Wnt/Wg signaling both during Drosophila development and in human cells. evi has not been found in other RNAi screens for Wg signaling components. In imaginal discs and embryos, evi is specifically required for Wg secretion, and loss of evi leads to accumulation of Wg in the producing cell (Bartscherer, 2006).

While significant progress has been made in understanding of how signals are transduced from the ligand-activated receptor to the nucleus in the signal-receiving cell, little is known about the events that lead to the secretion of a functional ligand from the producing cell. Recent studies investigating Hh signaling have identified several factors required in secreting cells, which posttranslationally modify the ligand prior to secretion (Bartscherer, 2006).

Like Hh, Wg has been demonstrated to be lipid modified, a modification that accounts for its hydrophobicity and its targeting to lipid rafts. This posttranslational modification is probably mediated by the ER-resident protein Porcupine (Por). To date, Por remains the only factor known to be specifically required for Wg secretion. Another factor implicated in the secretion of Wg is the dynamin Shibire (Shi). However, Shi has been shown to have a general function in vesicular transport rather than to be dedicated exclusively to Wg secretion (Bartscherer, 2006).

The multipass transmembrane protein Evi is specifically required for Wg/Wnt secretion across species. Three independent lines of evidence suggest that Evi has a specific role in Wnt signaling. (1) Loss-of-function alleles phenocopy wg mutations and do not show other obvious phenotypes. During early embryonic development, germline clones of evi phenocopy mutations in other core Wg pathway components. The phenotype of evi is similar to wg during wing imaginal-disc development. (2) evi knockdown does not prevent the JAK/STAT ligand Upd from activating a cell-based reporter. (3) In vivo analysis demonstrates that evi loss of function leads to neither abrogation of Hh target-gene expression in receiving cells nor Hh accumulation in producing cells. This demonstrates that Evi is not required for the secretion of either Upd or Hh ligands and supports the model that Evi is not part of a general secretory machinery. Interestingly, Evi does not seem to be involved in the PCP signaling pathway, which might indicate that Fz-dependent PCP signaling in Drosophila might not rely on a Wnt factor (Bartscherer, 2006).

While there is no evidence that evi acts in other signaling pathways, the results indicate that evi is a core Wg/Wnt signaling-pathway component in both invertebrates and vertebrates. Cell-based experiments in Drosophila and human cells show that Wg/Wnt signaling is impaired when evi is depleted by RNAi. In vivo, germline and somatic clones of evi phenocopy wg loss-of-function mutations (Bartscherer, 2006).

These results show that Evi is necessary for efficient secretion of Wg. Cell culture experiments and clonal analyses in imaginal-disc tissues demonstrate that evi is required in the Wg-producing cell. Like wg, evi shows cell-nonautonomous phenotypes on target-gene expression in somatic cell clones. It was observed that Wg accumulates in Wg-producing cells in an evi mutant background. Wg is not efficiently released into the supernatant of cultured cells treated with evi dsRNA. In addition to the embryonic phenotype, these results demonstrate an impaired Wg secretion when evi is absent. However, target genes that require only low levels of Wg, such as dll, are expressed in evi2 mutant discs. This phenotype can be explained by maternally contributed evi transcript, although it cannot be excluded that Wg is still secreted in low amounts by evi mutant cells (Bartscherer, 2006).

Sequence analysis of evi predicts that the gene encodes a transmembrane protein. Consistently, an Evi-EGFP fusion protein localizes to the plasma membrane, suggesting that Evi, unlike the ER-resident protein Por, might regulate Wg secretion at the plasma membrane. Sequence analysis does not indicate any obvious enzymatic activity, such as glycosyl- or acyltransferase activity. However, in polarized cells, such as wing imaginal-disc cells, loss of evi leads to the delocalization of Wg protein from the apical side. In mammalian cells, apical secretion of proteins has been attributed to specialized secretory pathways that include protein sorting into cholesterol/sphingolipid-rich microdomains (lipid rafts) within the trans-Golgi network. It has been shown that acylation of Wg by Por is required for incorporation of Wg in polarized vesicles that transport Wg to specific sites at the plasma membrane. Since loss of evi leads to the delocalization of Wg protein from the apical side, it is proposed that Evi might be involved in the proper localization of Wg prior to its secretion. However, in evi mutant cells, accumulation of Wg alone may account for its delocalization and may not be due to a specific involvement of Evi in apical sorting. Interestingly, apical localization of Wg-containing vesicles has been shown to be involved in Wg signaling in the Drosophila embryo. A possible function of Evi may lie in the packaging, localization, or fusion of such Wg-containing vesicles (Bartscherer, 2006).

This stuyd shows that Evi, a conserved transmembrane protein, is required for efficient secretion of Wg in embryos and from imaginal-disc cells in vivo. As Evi appears to localize to the plasma membrane, it is tempting to speculate that the biological role of Evi is downstream of the ER-resident protein Por in the secretion of Wnts. The results support a model in which the cellular release of growth factors is performed by dedicated machineries that provide an additional level of regulation for the secretion of ligands. Evi might therefore belong to a growing family of specialized factors that regulate the cellular release of specific families of growth factors (Bartscherer, 2006).

Sprinter/Wntless: a novel transmembrane protein required for Wg secretion and signaling

Wingless (Wg) is a secreted ligand that differentially activates gene expression in target tissues. It belongs to the Wnt family of secreted signaling molecules that regulate cell-to-cell interactions during development. Activation of Wg targets is dependent on the ligand concentration in the extracellular milieu; cellular mechanisms that govern the synthesis, delivery and receipt of Wg are elaborate and complex. sprinter (srt) encodes a novel, evolutionarily conserved transmembrane protein required for the transmission of the Wg signal. Mutations in srt cause the accumulation of Wg in cells that express it, and retention of the ligand prevents activation of its target genes in signal-receiving cells. In the absence of Srt activity, levels of Wg targets (including Engrailed in embryos lacking maternal and zygotic srt, and Senseless and Achaete in wing discs) are reduced. Activation of Wg targets in the receiving cells does not require srt. Hence, the function of Srt is restricted to events occurring within the Wg-producing cells. srt is not required for any aspect of Hedgehog (Hh) signal transduction, suggesting specificity of srt for the Wg pathway. It is proposed that srt encodes a protein required for Wg secretion that regulates maturation, membrane targeting or delivery of Wg. Loss of srt function in turn diminishes Wg-pathway activation in receiving cells (Goodman, 2007; full text of article).

The srt/CG6210/Wntless genomic locus is composed of three exons with two possible splice variants to encode novel proteins of 594 (isoform A) and 562 (isoform B) amino acids that include or exclude exon 2. Both splice variants are expressed in Drosophila, since the EK288129 and CK00022 ESTs exclude the second intron, whereas the srt GH01813 cDNA used in the rescue experiment includes it. Indeed, both splice variants are expressed in S2R+ cells. Analysis of the amino acid sequence suggests that Srt is composed of four to eight transmembrane domains. The signal sequence constitutes the first transmembrane domain because it does not have a good consensus-signal peptidase-cleavage site. The next four hydrophobic sequence elements all represent potential transmembrane domains, but are either too short to traverse the membrane or are weakly hydrophobic, reducing the likelihood that they are within the membrane. The next three hydrophobic regions of the Srt protein are probably transmembrane domains. Based on these observations, it is hypothesized that Srt has four transmembrane domains with a large N-terminal globular extracellular/luminal domain that has two potential N-linked glycosylation sites, although several other topologies are clearly possible. This predicted structure places the Trp492 srt7E4 nonsense mutation within the last transmembrane domain to yield either a truncated protein or one that is earmarked for degradation through nonsense-mediated decay of the message or through the breakdown of the misfolded protein. It was also noticed that, although Flybase has srt/CG6210 annotated as a multi-drug-resistance-related protein (MRP), this analysis of the Srt amino acid sequence indicates that the only commonality between these proteins is that they are multi-transmembrane-spanning proteins. Hence, the current annotation of srt/CG6210 in Flybase as MRP is incorrect (Goodman, 2007).

Sequence comparison of Srt to all protein databases reveals that its closest known relative is found in Drosophila pseudoobscura sharing 87% identity and 91% similarity along its length. In Drosophila melanogaster, the closest relative of Srt is encoded by CG13409, located at cytological region 94A, and has only 22% identity and 42% similarity. Srt shows much stronger homology to protein sequences from its evolutionarily distant relatives, suggesting that Srt is unique in Drosophila. The alignment of the Drosophila Srt isoform B relative to nematode, frog and human reveals that Drosophila Srt isoform B shares 43% identity and 62% similarity with human Srt (hSrt). Whereas some regions in the N-terminus and the majority of C-terminal regions of Srt diverge from its vertebrate relatives, there is a high level of conservation that extends throughout the central region of Srt. The most N-terminal amino acids, including the signal sequence/first putative transmembrane domain, are fairly well conserved across species, even in the absence of a good consensus peptidase cleavage site, supporting the hypothesis that this constitutes a transmembrane domain (Goodman, 2007).

It is believed that the primary function of Srt in the Wg pathway is to support the maturation of activate Wg ligand. In this capacity, it is possible that Srt acts in post-translationally processing Wg, in the targeting of Wg to the plasma membrane or in the release of active Wg from the membrane. Since porcupine mutants, as well as point mutations in the Wg protein itself, yield similar Wg-retention phenotypes, and because porc is required for the post-translational processing of Wg, a role for Srt in the post-translational processing of Wg is one possible function of Srt. In this role, it would be predicted that Srt might act as an enzyme that either participates in known post-translational changes to the Wg protein, such as glycosylation or palmitoylation, or identifies a new post-translational alteration in Wg that is required for its maturation. In addition to catalyzing the palmitoylation of Wg, the action of Porc is required to target Wg to lipid rafts in the plasma membrane (Zhai, 2004). This observation suggests that membrane targeting might occur by an active process mediated by specific protein(s). Another possible function of Srt could be as a Wg-specific chaperone protein that promotes proper folding and shuttles Wg through the secretory pathway to the plasma membrane once posttranslational processing is complete. Indeed, there is precedent for the need of protein-specific chaperones in the Wg pathway. In order for functional Arrow -- the Wg low-density lipoprotein co-receptor -- to reach the plasma membrane, it requires the activity of a specific chaperone protein, Boca. Recent studies suggest that at least some Wg protein is loaded into lipoprotein particles during larval development, which may be required for the movement of lipid-modified Wg in the extracellular space to establish its morphogenetic gradient in the wing. These lipoprotein particles are exogenously synthesized in the fat body and must be loaded with their lipid-modified cargo in the cells that express the ligand. Hence, there must be protein(s) present at the plasma membrane that catalyzes this process. Sprinter may be localized within membrane rafts, at the ready to load palmitoylated Wg into arriving lipoprotein particles for dissemination to the Wg target cells. Another potential role for Sprinter could be to act indirectly on Wg by supporting the posttranslational maturation or subcellular targeting of the proteins that directly regulate these processes, although a physical interaction between Wg and Srt (also known as Wls or Evi) has been reported (Bänziger, 2006; Goodman, 2007).

Localization of Srt within the secretory pathway could be predictive of its function. Srt localization in lipid rafts at the plasma membrane would suggest involvement in generating Wg-loaded lipoprotein particles. However, ER or Golgi localization could indicate a role in Wg maturation as it moves through the secretory pathway. Although determination of the subcellular localization of Srt awaits specific antibodies, it was observed that, in srt7E4-mutant tissues, there is shift in the cellular distribution of Wg toward the basolateral surface of wing-disc cells - the surface of Wg extracellular gradient formation in target cells - without disruption of Golgi localization. This would suggest that the srt block to Wg maturation occurs within the ER or a post-Golgi compartment of the secretory pathway (Goodman, 2007).

Wingless secretion requires endosome-to-Golgi retrieval of Wntless/Evi/Sprinter by the retromer complex

The glycolipoproteins of the Wnt family raise interesting trafficking issues, especially with respect to spreading within tissues. Recently, the retromer complex has been suggested to participate in packaging Wnts into long-range transport vehicles. Analysis of a Drosophila mutant in Vps35 show that, instead, the retromer complex is required for efficient progression of Wingless (a Drosophila Wnt) through the secretory pathway. Indeed expression of senseless, a short-range target gene, is lost in Vps35-deficient imaginal discs. In contrast, Vps35 is not required for Hedgehog secretion, suggesting specificity. Overexpression of Wntless, a transmembrane protein known to be specifically required for Wingless secretion overcomes the secretion block of Vps35-mutant cells. Furthermore, biochemical evidence confirms that Wntless engages with the retromer complex. It is proposed that Wntless accompanies Wingless to the plasma membrane where the two proteins dissociate. Following dissociation from Wingless, Wntless is internalized and returns to the Golgi apparatus in a retromer-dependent manner. Without the retromer-dependent recycling route, Wingless secretion is impaired and, as electron microscopy suggests, Wntless is diverted to a degradative compartment (Franch-Marro, 2008).

The retromer complex influences Wnt secretion by recycling Wntless from endosomes to the trans-Golgi network

Secreted Wnt proteins play essential roles in many biological processes during development and diseases. However, little is known about the mechanism(s) controlling Wnt secretion. Recent studies have identified Wntless (Wls) and the retromer complex as essential components involved in Wnt signaling. While Wls has been shown to be essential for Wnt secretion, the function(s) of the retromer complex in Wnt signaling is unknown. This study examined a role of Vps35, an essential retromer subunit, in Wnt signaling in Drosophila and mammalian cells. Compelling evidence is provided that the retromer complex is required for Wnt secretion. Importantly, Vps35 colocalizes in endosomes and interacts with Wls. Wls becomes unstable in the absence of retromer activity. These findings link Wls and retromer functions in the same conserved Wnt secretion pathway. It is proposed that retromer influences Wnt secretion by recycling Wntless from endosomes to the trans-Golgi network (TGN) (Belenkaya, 2008).

This study provided convincing evidence for mechanistic roles of retromer and Wls in Wnt secretion. The retromer complex is involved in Wnt secretion in both Drosophila and mammalian cells. Wls stability is regulated by Dynamin-mediated endocytosis, and retromer plays an essential role in maintaining Wls protein levels, possibly via the retrieval of Wls from endosomes to the Golgi. Together, these findings have linked retromer and Wls into the same conserved Wnt secretion pathway. It is proposed that retromer influences Wnt secretion by recycling Wntless from endosomes to the TGN (Belenkaya, 2008).

One main finding of this work is the demonstration of a retromer requirement for Wnt secretion. First, it was shown that Wg protein accumulated inside the Wg-producing cells, whereas Wg levels in the receiving cells were reduced in the absence of the retromer activity. Second, secretion of Wg, as well as of Wnt3a and Wnt5a proteins, is inhibited in cultured cells, in which retromer activity is depleted by RNAi. Finally, a role of retromer in controlling the levels of Wls protein, which is essential for Wnt secretion, is demonstrated (Belenkaya, 2008).

Previously, Coudreuse (2006) found that vps-35 mutant in C. elegans did not affect the level of the Wnt, EGL-20, within producing cells when examined using an EGL-20-GFP fusion protein. Coudreuse also failed to detect Wnt3a secretion defects in Vps35-depleted mammalian cells. These differences could be due to variations in experimental procedures. In the current experiment, mosaic clones were used to compare Wg accumulation in Vps35 mutant cells with surrounding wild-type cells, whereas Coudreuse used homozygous Vps35 mutant embryos to compare with wild-type embryos. Furthermore, Coudreuse used a relatively short time period (maximum 5 hr) to collect Wnt3a protein in Vps35-depleted cells, which perhaps was not long enough to detect a decrease in Wnt3a secretion. Finally, EGL-20-GFP and Wnt3a-proteinA chimeras used previously might be less dependent on retromer activity for secretion compared with the native forms of these proteins. In the current experiments, nontagged Wg, Wnt3a, and Wnt5a proteins were used to avoid this potential complication (Belenkaya, 2008).

Consistent with a role of retromer in Wg secretion, reduced Wg short-range and long-range signaling activities were observed in the absence of retromer activity in the wing discs. The data argue strongly that diminished levels of Wg protein contribute to the reduced short- and long-range signaling activities of Wg. In this regard, it is important to mention that the short-range signaling activity of EGL-20 was only mildly affected, while the long-range signaling of EGL-20 was fully impaired in C. elegans (Coudreuse, 2006). The distinct responses of target genes are attributed to differences in their sensitivity to extracellular Wnt protein levels. Indeed, it was noticed that while Sens is strikingly reduced, only relatively weak reduction of Dll expression is observed in the absence of retromer activity (Belenkaya, 2008).

Several models have been proposed for the function of retromer in Wnt-gradient formation. These models include roles of retromer in (1) Wnt maturation by directly interacting with other enzymes such as Porcupine and (2) facilitating the interaction of Wnt protein with lipoprotein particles (Coudreuse, 2007; Coudreuse, 2006). Although these mechanisms of retromer's role in Wnt activity cannot be ruled out completely, in light of the data, it is argued strongly that the main function of retromer in Wnt signaling is to maintain Wls protein levels. It was shown that Wls is substantially reduced in the absence of retromer activity and it interacts with retromer in cells. In further support of this view, it was shown that overexpression of Wls can restore Wg secretion defects in Dvps35 depleted wing discs. Since Wls is likely to be required for the secretion of all of Wnt members (Banziger, 2006; Hausmann, 2007), it is suggested that retromer may be essential for the secretion of all Wnt proteins (Belenkaya, 2008).

One main issue related to Wls function is its subcellular distribution. While Bartscherer (2006). identified Wls at the plasma membrane of Drosophila and HEK293T cells, Banziger (2006) showed localization of Wls in the Golgi and in vesicles between the Golgi and the surface of mammalian cells. To clarify this issue, Wls distribution was carefully examined in wing discs and in HeLa cells. The results showed that Wls protein localizes on plasma membrane, TGN, and vesicles. Data from the biotinylation experiment, the extracellular staining of Wls, and colocalization with plasma membrane markers provide compelling evidence that some of the Wls protein is present on the plasma membrane. Importantly, the data revealed colocalization of Wls protein with early endosome markers including Rab5, EEA1, and Hrs. Thus, it is likely that Wls is trafficking from the TGN onto cell membrane and is then subsequently endocytosed from the cell surface. A previous study showed that Wls is able to bind Wnt (Banziger, 2006). Therefore, it is likely that Wls acts as a Wnt cargo receptor for the delivery of Wnt ligands from the Golgi onto the cell surface for secretion. Consistent with this view, accumulation of Wg in the Golgi was observed in the absence of Vps35 activity. Perhaps, the absence of retromer activity causes depletion of Wls in the Golgi, thereby resulting in Wg accumulation in the Golgi (Belenkaya, 2008).

These experiments further demonstrate that Wls is actively internalized and degraded through Dynamin-mediated endocytosis. Since Wls is enhanced in endosomes in the absence of Hrs activity, the data also suggest that Wls is likely to undergo the lysosomal degradation. Endocytosis-mediated protein degradation has been shown to be a mechanism for regulation of a number of signaling receptors such as Patched, Thickveins, and EGF receptor. Thus, this work builds on the known principle that lysosomal targeting of receptors regulates signaling in the responding cell, by showing that lysosomal targeting of a putative cargo receptor can also attenuate the production and presentation of ligand in the first place (Belenkaya, 2008).

Extensive colocalization of Wls with Vps35 protein was found in endosomes in HeLa cells, arguing that endosomes are the main sites of retromer activity. Studies in both yeast and mammalian cells suggest an essential role of retromer in retrieving membrane proteins from endosomes back to TGN (Seaman, 2005). In light of the data, it is suggested that the retromer complex is involved in recycling Wls from endosomes to TGN for its further function in Wnt secretion. This is likely mediated by the interaction of Wls with the retromer complex. Consistent with this view, it was found that cell surface Wls can be internalized and returned to the Golgi. In the absence of retromer activity, internalized Wls is likely to be sorted into lysosomes for degradation. In support of this view, it was demonstrated that overexpression of Vps35 can significantly enhance levels of Wls in mammalian cells even in the absence of Wnt ligand (Belenkaya, 2008).

On the basis of these findings, the following model is proposed. At the ER, Wnt protein is dually lipid-modified by the Porc. The modified Wnt exits from the ER and enters the Golgi, where it binds Wls. Wls carries and sends Wnt proteins from TGN onto the cell surface for secretion. Wnt is delivered to the cell surface, the unloaded Wls protein on the plasma membrane will subsequently be internalized through a Dynamin-mediated endocytosis process. In the endosomes, the internalized Wls can have two fates. (1) The retromer complex interacts with Wls and retrieves Wls from endosomes back to TGN, thereby maintaining the normal levels of Wls protein. (2) In the absence of retromer activity, Wls protein is subsequently delivered into lysosomes for degradation. In this model, Wls acts as a Wnt cargo receptor. More experiments will be needed to define the mechanism by which Wls protein transports Wnt from TGN onto the cell surface for secretion. In addition, it also remains to be determined whether Wls is required for Wnt modification or its interaction with other proteins such as lipoprotein particles before its secretion (Belenkaya, 2008).

Active Wnt proteins are secreted on exosome

Wnt signalling has important roles during development and in many diseases. As morphogens, hydrophobic Wnt proteins exert their function over a distance to induce patterning and cell differentiation decisions. Recent studies have identified several factors that are required for the secretion of Wnt proteins; however, how Wnts travel in the extracellular space remains a largely unresolved question. This study shows that Wnts are secreted on exosomes both during Drosophila development and in human cells. Exosomes carry Wnts on their surface to induce Wnt signalling activity in target cells. Together with the cargo receptor Evi/WIs, Wnts are transported through endosomal compartments onto exosomes, a process that requires the R-SNARE Ykt6. This study demonstrates an evolutionarily conserved functional role of extracellular vesicular transport of Wnt proteins (Gross, 2012).

The present study demonstrated by biochemical and genetic approaches that a portion of functional Wnts is secreted on exosomes. The discrepancy between the hydrophobic properties of Wnts and their trafficking over longer distances has been a puzzling observation, leading to different hypotheses on how Wnts disperse. The results demonstrate that Wnts are secreted on exosomes in vivo and ex vivo. These secreted microvesicles originate through inward budding in multivesicular bodies (MVBs) and influence intercellular trafficking. Several lines of evidence support these conclusions. First, by biochemical fractionation it was found Wnt co-segregating with exosomes derived from mammalian and Drosophila cells. Ultrastructural analysis demonstrated that Wg is found on the outer membrane of exosomes. Second, it was demonstrate that Wnts are shuttled to MVBs by their cargo receptor Evi; this step is impaired by inhibition of MVB maturation or depletion of the ESCRT-0 complex, which mediates cargo recognition of exosomal proteins. Third, it was shown that a fraction of extracellular Wg co-localizes with several exosomal markers in vivo. Fourth, Ykt6 was identified as a protein required for secretion of Wnts and exosomal proteins. The results argue for an alternative route of Wnt secretion in vivo that is independent of lipoprotein particles. The biochemical fractionation method differs from the parameters used to biochemically isolate lipoprotein particles; in addition, lipoprotein particles were not found by ultrastructural analysis in the preparations. Taken together, it is believed that this study provides solid evidence for a Wnt secretion route through exosomes (Gross, 2012).

Ykt6 belongs to the longin type of R-SNARE involved in various trafficking events. Studies in yeast have suggested that Ykt6 localizes to the Golgi and to endosomal and vacuolar membranes. This study shows that in Drosophila Ykt6 does not affect secretion of transmembrane proteins such as Ptc and Fmi, and that on loss of Ykt6 in HeLa cells Evi accumulates in early endosomes but not in the Golgi. This finding is consistent with the role of Ykt6 in early/recycling endosomes described in mammalian cells; however, further functional roles for Ykt6 might exist and might also be modulated in a cell-type-specific manner (Gross, 2012).

Interestingly, the results imply that the retromer complex acts upstream of MVB sorting of Evi-Wnt complexes, as demonstrated by the effect of ESCRT and Ykt6 RNAi on secretion of Wnt/Evi and the exosomal marker CD81. The retromer-SNX3 sorting route might decide whether unloaded Evi is recycled to the Golgi or whether Evi-Wnt complexes are packaged onto exosomes. This is supported by the finding that depletion of retromer leads to lysosomal sorting, thereby reducing the pool of Evi as well as inhibiting the secretion of Wnts on exosomes. It has been shown that SNX3 together with HGS functions in sorting and membrane invagination at the MVB level. In yeast, SNX3, similar to Ykt6, retrieves cargo receptors to Golgi before homotypic fusion of vacuoli. Thus, it is tempting to speculate that a retromer-SNX3-dependent step retrieves empty Evi to Golgi, whereas Wnts (and Evi) are packaged onto exosomes for functional release (Gross, 2012).

Different extracellular forms of Wnts might not be mutually exclusive. The results show that removal of vesicle-bound Wnt only partially reduced Wnt activity, implying that other secretion routes exist, possibly through the direct release from the plasma membrane, which might have different extracellular trafficking properties (Gross, 2012).

It is tempting to speculate that morphogens provide spatial information in other forms than concentrations of signalling molecules, involving a 'digital' rather than 'analogue' mode of encoding signals. Different packaging modes of Wnt molecules and their control in Wnt-producing cells might have a profound effect on spreading and signalling properties of morphogens (Gross, 2012).

return to wingless Protein Interactions part 1/3 | part 2/3

wingless continued: Biological Overview | Evolutionary Homologs | Transcriptional regulation |Targets of Activity | Developmental Biology | Effects of Mutation | References

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.