cubitus interruptus

Protein interactions (part 1/2)

Post-transcriptional regulation of CI is observed by stage 11. CI transcript levels are uniform across the posterior compartment, but protein levels are higher next to the compartment boundaries. The distribution of the CI protein is altered in fused, hedgehog and wingless mutants suggesting cell-cell signaling may regulate CI protein levels (Motzny, 1995).

Members of the Hedgehog (Hh) and Wnt/Wingless (Wg) families of secreted proteins control many aspects of growth and patterning during animal development. Hh signal transduction leads to increased stability of the transcription factor Cubitus interruptus (Ci), whereas Wg signal transduction causes increased stability of Armadillo (Arm/beta-catenin), a possible co-factor for the transcriptional regulator Lef1/TCF. A new gene, slimb (for supernumerary limbs), is described which negatively regulates both of these signal transduction pathways. Loss of slimb function results in a cell-autonomous accumulation of high levels of both Ci and Arm, and the ectopic expression of both Hh- and Wg- responsive genes. Clones of slimb1 cells in the leg or wing disc ectopically express dpp or wg when they arise in the anterior (but not the posterior) compartments of these discs. Anterior clones reorganize normal limb pattern, creating supernumerary 'double-anterior' limbs. Slimb, like PKA, is a negative regulator that normally prevents activity of the Hh signal transduction pathway in the absence of ligand. slimb mutant cells that arise in the presumptive wing blade ectopically express Scute and differentiate ectopic sensory bristles instead of epidermal hairs on the surface of the wing blade. Both phenotypes are strictly autonomous to the mutant cells, as is the case when the Wg signal transduction pathway is constitutively activated, but not when Wg is ectopically expressed. The slimb gene encodes a conserved F-box/WD40-repeat protein related to Cdc4p, a protein in budding yeast that targets cell-cycle regulators for degradation by the ubiquitin/proteasome pathway. It is proposed that Slimb protein normally targets Ci and Arm for processing or degradation by the ubiquitin/proteasome pathway, and that Hh and Wg regulate gene expression, at least in part, by inducing changes in Ci and Arm, which protect both Ci and Arm from Slimb-mediated proteolysis (Jiang, 1998).

The slimb phenotype differs from those of all previously known genes, in that it is the first gene found to deregulate both wg and dpp expression in the D/V axis. Disrupting components of the Hh signaling pathway deregulate wg and dpp only along the A/P axis. Thus, the control of wg and dpp expression in the D/V axis is not disrupted by disruption of the Hh pathway. The mechanism restricting wg and dpp in the D/V axis is not known. The mutant phenotype of slimb- clones in discs provides the first evidence that wg and dpp expression in the D/V axis is actively regulated during imaginal disc development, and is not solely defined during embryonic development. Since the Hh pathway regulates wg and dpp expression in the A/P axis, these results suggest that a pathway different from Hh may operate in imaginal discs to restrict their expression in the D/V axis. This pathway cannot be either the Wg or Dpp signaling pathway since inactivation of Wg or Dpp signaling is known to affect either dpp or wg expression, but not both. The slimb phenotypes described here were not observed in the previous study which used weak slimb alleles and revealed only A/P defects (Jiang, 1998). Jiang proposed that Slimb protein normally targets Ci and Arm for processing or degradation by the ubiquitin/proteasome pathway, and that Hh and Wg regulate gene expression, at least in part, by inducing changes in Ci and Arm, which protect both Ci and Arm from Slimb-mediated proteolysis. The phenotypic differences probably reflect the fact that a null allele was used in the current study instead of hypermorphic alleles. In addition to D/V defects, slimb mutant clones also deregulate wg and dpp expression in the A/P axis. slimb is the first identified gene that regulates both wg and dpp expression in the A/P as well as D/V axes (Theodosiou, 1998).

The function of Costa (also known as Costal2 or Cos2), a kinesin-like protein, in Hedgehog signal transduction is unexpected, and the reason for its involvement is far from clear. Perhaps Cos2 serves to anchor Cubitus interruptus in the cytoplasm, preventing its transport into the nucleus where it functions as a transcription factor. Of particular interest is the observation that although Ci is found in a complex with Cos2, Cos2 activity reduces Ci staining in anterior compartment cells. Cos2 somatic clones in the anterior compartment of wing discs express high levels of cytoplasmic Ci staining and cause mirror-image duplications of the wing (Sisson, 1997). These results also conflict with the idea that Cos2 stabilises cytoplasmic Ci.

Whereas Cos2 activity is associated with decreased Ci levels, Cos2 levels are posttranscriptionally elevated in the anterior compartment. COS2 mRNA levels are uniform throughout discs. Therefore, the elevated level of Cos2 protein in anterior compartment cells must be due to differences between anterior and posterior cells in either the production or the stability of Cos2 protein. The uniform level of Cos2 throughout the anterior compartment of imaginal discs is inconsistent with Hh signal regulating its accumulation. Perhaps Ci stabilizes Cos2 in a macromolecular complex in anterior cells or Ci heightens translation of COS2 mRNA (Sisson, 1997).

Cos2 is cytoplasmic and binds microtubules (see betaTubulin56D) in a taxol-dependent, ATP-insensitve manner, while kinesin heavy chain binds microtubules in a toxol-dependent, ATP-insensitive manner. Ci associates with Cos2 in a large protein complex, suggesting that Cos2 directly controls the activity of Ci. This association does not involve microtubules. Elevated cytoplasmic Ci staining is seen in cos2 clones in the anterior compartment. The level of Ci staining is independent of the clone's distance from the A/P border. Nuclear Ci is not evident in the clones (Sisson, 1997).

The discovery of a multiprotein complex in the cytoplasm provides some of the explanation for regulation in the Hedgehog pathway, but the dynamic roles of Cos2 and Fused are not yet well understood and the fine details are still obscure. Stimulation of cells with Hh leads to an additional serine phosphorylation for both Fused and Cos2. The protein kinase(s) responsible for these phosphorylations have not been identified. The Hh-induced phosphorylation of Fused appears as long as 30 minutes after induction, suggesting that it represents a feedback device rather than an event in initial signal transduction. This leads in turn to the possibliity that Fused is not autophosphorylating, even though the phosphoryation can be abolished by mutations in the catalytic domain of Fused. Similarly, Fused is apparently not directly responsible for the phosphorylation of Cos2, which occurs even when inactivating mutations are present in the kinase domain of Fused (Robbins, 1997).

The Hedgehog (Hh) family of signaling proteins mediates inductive interactions either directly or by controlling the transcription of other secreted proteins through the action of Gli transcription factors, such as Cubitus interruptus (Ci). In Drosophila, the transcription of Hh targets requires the activation of the protein kinase Fused (Fu) and the inactivation of both Suppressor of fused [Su(fu)] and Costal-2 (Cos-2). Fu is required for Hh signaling in the embryo and in the wing imaginal disc and acts also as an antitumorigen in ovaries. All fu- phenotypes are suppressed by the loss of function of Su(fu). Fu, Cos-2 and Ci are co-associated in vivo in large complexes which are bound to microtubules in an Hh-dependent manner. The role of Su(fu) in the intracellular part of the Hh signaling pathway has been investigated. Using the yeast two-hybrid method and an in vitro binding assay, Su(fu), Ci and Fu are shown to interact directly to form a trimolecular complex, with Su(fu) binding to both its partners simultaneously. Su(fu) and Ci also co-immunoprecipitate from embryo extracts. It is proposed that, in the absence of Hh signaling, Su(fu) inhibits Ci by binding to it and that, upon reception of the Hh signal, Fu is activated and counteracts Su(fu), leading to the activation of Ci (Monnier, 1998).

Drosophila CBP (Nejire) is a co-activator of cubitus interruptus in hedgehog signaling. Drosophila CBP predicts a protein of relative molecular mass 332,000; the gene maps to position 8F/9A on the X chromosome. Mutants for dCBP gene, nejire, die at stages 9 or 10 during embryogenesis, although some embryos survive to hatching. The most severe phenotype of the nej hemizygotes is the twisting of the embryo at germband elongation. The expression of wingless is strikingly reduced at the posterior margin of each parasegment in mutants. In addition, engrailed expression, which is maintined by WG protein, is significantly lower than in wild type. These observations suggest the Drosophila CBP might contribute to the functioning of some transcription factors involved in the wingless-engrailed signaling pathway. Cubitus interruptus protein physically interacts with Drosophila CBP (dCBP). A series of deletion mutants of ci indicate that a region of CI between amino acids 1020 and 1160 is required for phosphorylation independent interaction with dCBP. This region is part of the CI transactivation domain, C-terminal to five putative PKA sites. dCBP expression augments transactivation by CI up to a maximum of 62 fold. The dominant gain-of-function ciD mutant phenotype in which the longitudinal vein 4 of the adult wing is shortened, some posterior row hairs are missing, and the posterior wing margin is flattened, can be explained by the inappropriate expression of ci in the posterior compartment of the wing imaginal disc, where it is usually repressed by Engrailed. A subset of the ciD wing defects is suppressed by haploinsufficiency of dCBP. Thus dCBP is required for the activation of Cubitus interruptus target genes such as patched, and CBP is required for the activator function of CI but not for the repressor function. dCBP binds to dCREB2, the Drosophila homolog of CREB, in a phosphorylation-dependent manner, whereas the dCBP-CI interaction is phosphorylation-independent. These findings raise the possiblilty that a limited amount of dCBP might be recruited to PKA-phosphorylated dCREB2, resulting in a decrease in CI activity, explaining the antagonistic actions of PKA and Hedgehog (Akimaru, 1997).

The secreted Drosophila Hedgehog (Hh) protein induces transcription of specific genes by an unknown mechanism that requires the serpentine transmembrane protein Smoothened (Smo) and the transcription factor Cubitus interruptus (Ci). Protein kinase A (PKA) has been implicated in the mechanism of Hh signal transduction because it acts to repress Hh target genes in imaginal disc cells that express Ci. Changes in Ci protein levels, detected by an antibody that recognizes an epitope in the carboxy-terminal half of Ci, have been suggested to mediate the positive effects of Hh and the negative effects of PKA on Hh target gene expression in imaginal discs. The effects of PKA on Hh target genes were examined by expressing a mutant regulatory subunit, R*, to reduce PKA activity in embryos. The alterations of wingless, patched and ventral cuticle patterns due to PKA inhibition resemble those induced by low-level ubiquitous expression of Hh but are less pronounced than those elicited by high levels of Hh or strong patched mutations. A constitutively active mouse PKA catalytic subunit transgene (mC*) expressed in Drosophila embryos causes ectopic expression of wingless and patched . Responses to elevated PKA activity require smoothened and cubitus interruptus, but not hedgehog. The absolute requirement for Smo to observe transcriptional induction by PKA hyperactivity is consistent with two mechanisms: (1) either PKA acts on Smo, directly or indirectly, perhaps to uncouple it from the inhibitory influence of Ptc or, (2) alternatively, Smo has Hh-independent activity that acts in parallel with PKA to stimulate wingless and patched expression. There is considerable evidence that phosphorylation can alter activity of G protein-coupled receptors. (Ohlmeyer, 1997).

PKA inhibition, like Hh, leads to increased "carboxy-terminal" Ci staining and Hh target gene expression in embryos. Hh and Smo can stimulate target gene expression at constant Ci levels; increased PKA activity can induce ectopic Hh target gene expression in a manner that requires Smo and Ci activity but does not involve changes in Ci protein concentration. Nevertheless, elevated PKA suppresses the elevation of Ci-C-terminal antibody staining normally elicited by Hh at the borders of each Ci expression stripe. This suggests a branching pathway of Hh signal transduction downstream of Smo and that PKA exerts opposite effects on the two branches. Two PKA targets (direct targeting of Smo and targeting of Ci) with opposing actions on Hh target gene expression can account for the initially surprising observation that both PKA inhibtion and PKA hyperactivity induce wingless and patched expression in embryos. The negative target, relevant to regulating Ci protein levels, is sensitive almost exclusively to reduction of PKA activity. Hh signaling in embryos does not depend on cAMP-dependent regulation of PKA activity (Ohlmeyer, 1997).

A wingless enhancer region has been described whose Cubitus interruptus (Ci) binding sites mediate Ci-dependent transcriptional activation in transiently transfected cells. Hedgehog (Hh) and Patched (Ptc) act through those Ci binding sites to modulate the level of Ci-dependent transcriptional activation in S2 cells. To test for effects of Ptc and Hh, titrations of Ci cDNA in cultured cells were performed on an expression vector regulated by this enhancer region. The titrations were performed either in the presence of Hh cDNA or in the presence of Ptc cDNA. Reporter activity is reduced 3-fold in the presence of co-transfected Ptc. The addition of Hh results in a 1.5-fold increase in reporter activity over that observed for Ci alone. This same wg enhancer region is Hh responsive in vivo and its Ci binding sites are necessary for its activity. This provides strong evidence that Hh affects wg transcription through post-translational activation of Ci (Von Ohlen, 1997b).

Protein kinase A directly regulates the activity and proteolysis of cubitus interruptus

A 75-kDa proteolytic product of the full-length Cubitus interruptus (Ci) protein translocates to the nucleus and represses the transcription of Ci target genes. In cells that receive the hh signal, the proteolysis of Ci is inhibited and the full-length protein can activate the hh target genes. Because protein kinase A (PKA) inhibits the expression of the hh target genes in developing embryos and discs and the loss of PKA activity results in elevated levels of full-length Ci protein, PKA might be involved directly in the regulation of Ci proteolysis. It has been demonstrated that the PKA pathway antagonizes the hh pathway by phosphorylating Ci (Chen, 1998).

Mutant Ci proteins that cannot be phosphorylated by PKA have increased transcriptional activity compared with wild-type Ci. A search for the RRXS/T consensus PKA phosphorylation motif in Ci shows that there are four consensus PKA phosphorylation sites clustered in the C-terminal transactivation domain. To determine whether PKA affects Ci transcriptional activity directly, the serines in the PKA sites were mutated to alanines either singly [m1 (Ser to Ala at aa 838), m2 (Ser to Ala at aa 856), m3 (Ser to Ala at aa 892), m4 (Thr to Ala at aa 1,006)] or in various combinations, including the mutation of all four sites (null). These PKA target site mutants were cotransfected with a Ci target reporter construct and the CAT activities were compared with those of wild-type Ci. M1, m2, m3, and the mutation that is null for all four PKA sites greatly stimulates Ci activity, whereas m4 has less of an effect. This result suggests that the inhibition of phosphorylation of any one of the first three Ci PKA sites is sufficient to stimulate Ci transcriptional activity. Although the major effect of PKA activity in this system is negative, when Ci can no longer be the target for PKA phosphorylation, a minor positive effect of PKA has been unmasked. PKA might phosphorylate other factors crucial for Ci-mediated transactivation, such as dCBP, to activate CI-mediated transcription. This dual role played by PKA is in agreement with the genetic result by Ohlmeyer (1997) that overexpression of the PKA catalytic subunit increases the hh target gene expression in a ci- and smo-dependent manner (Chen, 1998).

The PKA-mediated phosphorylation of Ci promotes its proteolysis from the full-length active form to the 75-kDa repressor form. Kc cells were transfected with HA-CI (WT) or HA-CI (null) (Ci cDNA fused to an N-terminal HA tag) and the exogenous HA-CI protein was precipitated from whole-cell extracts by a rat monoclonal anti-HA antibody. The HA antibody detects both the 155-kDa full-length form and the 75-kDa form when HA-CI wild type is transfected into Kc cells. When HA-CI (null), which has all four PKA sites mutated, was transfected into cells, the 75-kDa CI protein could not be detected, indicating that PKA phosphorylation of CI is required for the proteolysis of CI. Kc cells were also cotransfected with HA-CI (WT) or HA-CI (null) and either PKA or PKI, an inhibitor of PKA. Cotransfection of PKI with HA-CI (WT) inhibits the formation of the 75-kDa CI protein, whereas the cotransfection of HA-CI (WT) and PKA stimulates the formation of the 75-kDa CI product This result suggests that the phosphorylation of CI by PKA stimulates the proteolytic processing of CI in Kc cells (Chen, 1998).

Drosophila CBP has several consensus PKA sites, one of which is located between the third and fourth zinc fingers and is well conserved among members of the p300/CBP gene family. Results from other research has shown that a GAL-CBP1678-2441 fusion protein, which includes this conserved PKA site, is a transcription activator and that the activity of this chimera increases when PKA activity is stimulated in PC12 cells; however, there is no evidence as yet that PKA increases CBP activity by phosphorylation of this site. In addition, the finding that the activity of GAL-CBP1-460 is greatly stimulated by PKA treatment in PC12 cells (17-fold) further supports the idea that PKA modulates the activity of CBP. However, the current results do not rule out the possibility that PKA phosphorylates and activates other transcriptional factors involved in CI-mediated gene activation as well. Whether the hh-smo pathway directly affects PKA function is unclear (Chen, 1998).

Proteolysis of Cubitus interruptus in Drosophila requires phosphorylation by Protein Kinase A

The Hedgehog signal transduction pathway is involved in diverse patterning events in many organisms. In Drosophila, Hedgehog signaling regulates transcription of target genes by modifying the activity of the DNA-binding protein Cubitus interruptus (Ci). Hedgehog signaling inhibits proteolytic cleavage of full-length Ci (Ci-155) to Ci-75, a form that represses some target genes, and also converts the full-length form to a potent transcriptional activator. Reduction of protein kinase A (PKA) activity also leads to accumulation of full-length Ci and to ectopic expression of Hedgehog target genes, prompting the hypothesis that PKA might normally promote cleavage to Ci-75 by directly phosphorylating Ci-155. A mutant form of Ci lacking five potential PKA phosphorylation sites (Ci5m) is not detectably cleaved to Ci-75 in Drosophila embryos. Moreover, changes in PKA activity dramatically alters levels of full-length wild-type Ci in embryos and imaginal discs, but does not significantly alter full-length Ci5m levels. These results are corroborated by showing that Ci5m is more active than wild-type Ci at inducing ectopic transcription of the Hh target gene wingless in embryos and that inhibition of PKA enhances induction of wingless by wild-type Ci but not by Ci5m. It is therefore proposed that PKA phosphorylation of Ci is required for the proteolysis of Ci-155 to Ci-75 in vivo. It is also showm that the activity of Ci5m remains Hedgehog responsive if expressed at low levels, providing further evidence that the full-length form of Ci undergoes a Hedgehog-dependent activation step (Price, 1999).

How does PKA phosphorylation of Ci-155 lead to its proteolysis? Loss of cos2 activity in wing disc clones induces high levels of Ci-155, suggesting that the integrity of the multiprotein cytoplasmic complex that contains Ci or the association of this complex with microtubules may be necessary in order for proteolysis to occur. It is possible that Ci phosphorylation also affects proteolysis by altering these interactions. A more direct role for PKA phosphorylation of Ci has been proposed based on the sequence and properties of the Slimb protein, which affects the conversion of Ci-155 to Ci-75. Slimb belongs to a family of F-box/ WD40-repeat proteins implicated in binding to and targeting phosphorylated molecules for ubiquitin-mediated degradation. It was recently shown that the vertebrate Slimb homolog, beta-TRCP, targets IkappaB and beta-catenin for ubiquitin-mediated degradation by binding specifically to a phosphorylated motif (DSGXXS, where both serines must be phosphorylated) present in both proteins. Whether Slimb participates in such a direct manner in Ci proteolysis is not clear. Slimb has not been shown to bind to Ci, and Ci proteolysis has not been shown to involve ubiquitination; Ci proteolysis is also unusual in being incomplete, leaving a stable 75 kDa product. Sequences around three Ci sites show some extended similarity to each other but are quite different from the IkappaB and beta-catenin consensus. It will be interesting to determine if Slimb, or another F-box protein, can bind directly to these regions of Ci when phosphorylated by PKA. Since Slimb recognition requires phosphorylation at multiple residues and the PKA site consensus in Ci contains additional serines, it is worth considering that the activity of another protein kinase in addition to PKA may also contribute to the regulation of Ci proteolysis (Price, 1999 and references).

Protein kinase A antagonizes Hedgehog signaling by regulating both the activator and repressor forms of Cubitus interruptus

In Drosophila, Hh transduces its signal via Cubitus interruptus (Ci), a transcription factor present in two forms: a full-length activator and a carboxy-terminally truncated repressor that is derived from the full-length form by proteolytic processing. The proteolytic processing of Ci is promoted by the activities of protein kinase A (PKA) and Slimb, whereas it is inhibited by Hh. PKA inhibits the activity of the full-length Ci in addition to its role in regulating Ci proteolysis. Whereas Ci processing is blocked in both PKA and slimb mutant cells, the accumulated full-length Ci becomes activated only in PKA but not in slimb mutant cells. Moreover, PKA inhibits an uncleavable activator form of Ci. These observations suggest that PKA regulates the activity of the full-length Ci independent of its proteolytic processing. Evidence exists that PKA regulates both the proteolytic processing and transcriptional activity of Ci by directly phosphorylating Ci. It is proposed that phosphorylation of Ci by PKA has two separable roles: (1) it blocks the transcription activity of the full-length activator form of Ci, and (2) it targets Ci for Slimb-mediated proteolytic processing to generate the truncated form that functions as a repressor (G. Wang, 1999).

A recent study by Methot (1999) demonstrates the existence of distinct activator and repressor forms of Ci. These two forms play separable roles in Drosophila limb development by regulating different sets of Hh target genes. In the developing wing, the expression of ptc and en appears to be exclusively controlled by the activator form of Ci whereas dpp expression is governed by both the activator and repressor forms. Interestingly, preventing Ci proteolysis with an uncleavable form of Ci is not sufficient to convert Ci into a constitutive activator, suggesting that the full-length activator form of Ci encounters additional regulatory block(s) that need to be alleviated by Hh signaling (Methot, 1999). Evidence is provided that PKA activity exerts such a block. Initial evidence that PKA regulates the activator form of Ci comes from a close examination of PKA and slimb mutant phenotypes. In slimb mutant cells, Ci processing is nearly abolished, and, as a consequence, full-length Ci accumulates. However, the expression of ptc-lacZ and en is not induced, suggesting that the full-length form of Ci that accumulates in slimb mutant cells is transcriptionally silent. In contrast, PKA mutant cells express ptc-lacZ and en even though they accumulate full-length Ci at levels comparable to slimb mutant cells. This suggests that the full-length Ci that accumulates in PKA mutant cells is transcriptionally active. Furthermore, slimb mutant cells with reduced PKA activity ectopically express ptc-lacZ, arguing that the lack of Ci activity in slimb mutant cells is due to an inhibitory role for PKA rather than a positive requirement for Slimb in the Hh signaling pathway. Further evidence that PKA regulates the activator form of Ci independent of its processing come from the gain-of-function studies. The ectopic expression of ptc-lacZ induced by the uncleavable activator form of Ci (Ci^U) is suppressed by overexpression of a constitutively active form of PKA (mC*) (G. Wang, 1999).

In support of the view that Ci is a direct target for PKA in regulating Hh signaling, it was found that a modified form of Ci with three PKA phosphorylation consensus sites mutated is not processed but exhibits constitutive activity when expressed in the developing wings. Although these observations suggest that PKA antagonizes Hh signaling by directly phosphorylating Ci and targeting it for proteolysis, they do not to address whether phosphorylation of Ci by PKA also regulates the activity of the full-length activator form of Ci. The low levels of constitutive activity exhibited by the PKA phosphorylation-deficient form of Ci could be secondary to the lack of Ci processing, which results in a dramatic increase in the levels of the full-length activator form of Ci, since it has been shown that overexpression of a full-length wild type form of Ci can activate ptc expression in wing discs. To define the role of PKA phosphorylation in regulating the activity of the full-length Ci, advantage was taken of the uncleavable activator form of Ci (Ci^U). Mutating multiple PKA phosphorylation sites in Ci^U dramatically alters its transcriptional activity and renders it constitutively active. This observation suggests that the activity of Ci^U is normally blocked by PKA phosphorylation, even though its processing is impaired. This result provides compelling evidence that PKA phosphorylation of Ci inhibits the activator form of Ci, independent of its role in promoting Ci processing. Taken together, these results suggest the following working model for the inhibitory function of PKA in the Hh pathway. It is proposed that PKA phosphorylation of Ci in its carboxy-terminal region has two separable roles: (1) it blocks the activity of the full-length activator form of Ci, and (2) it targets the full-length Ci for Slimb-mediated proteolysis to generate the truncated repressor form of Ci. Such a dual regulation ensures that only the repressor form of Ci is active in the absence of any Hh signaling. This model accounts for the difference between PKA and slimb mutant phenotypes. In slimb mutant cells, Ci is not processed to the repressor form but accumulates in an inactive phosphorylated form, and, as a consequence, dpp is derepressed at low levels but ptc and en are not activated. In PKA mutant cells, however, Ci accumulates in an unphosphorylated or hypophosphorylated active form, and, as a consequence, ptc and en are activated (G. Wang, 1999).

How phosphorylation of Ci regulates its activity and proteolytic processing remains to be explored. It has been proposed that Su(fu) attenuates Hh signaling activity by blocking a maturation step that converts Ci into a short-lived nuclear transcriptional activator. Analyses of slimb Su(fu) double mutant and slimb Su(fu) PKA triple mutant phenotypes suggest that inhibition of Ci activity by PKA is independent of Su(fu). When Ci processing is blocked, removing Su(fu) only partially stimulates the activity of the full-length Ci whereas simultaneously removing Su(fu) function and reducing PKA activity leads to virtually full activation of Ci. These observations suggest that PKA and Su(fu) act in parallel through independent mechanisms to regulate the activity of the full-length Ci. In slimb Su(fu) double mutant cells, the majority of unprocessed full-length Ci appears to be transformed into a labile nuclear form, and yet the activity of this nuclear form of Ci seems to be inhibited by PKA. This implies that PKA might inhibit Ci at a step after it enters the nucleus. For example, phosphorylation of Ci by PKA might prevent the formation of an active Ci transcription complex or might attenuate its ability to bind DNA. Another possible mechanism by which PKA exerts its influence on Ci is to regulate its nuclear trafficking. It has been shown recently that Hh signaling increases the nuclear import of full-length Ci. As PKA and Hh act antagonistically, it is possible that PKA phosphorylation of Ci might tether the full-length Ci in the cytoplasm in the absence of Hh signaling (G. Wang, 1999).

Su(fu), Cos2, and the Ser/Thr kinase Fu form a multiprotein complex with Ci and the complex associates with microtubules in a manner regulated by Hh. It has been proposed that the assembly of the microtubule-associated Ci complex is critical for inhibiting Ci activity, possibly by tethering Ci in the cytoplasm. The relationship between PKA phosphorylation and the formation of Ci complex is not known. It is not clear whether they are two independent processes or whether one step might regulate the other. The nearly identical phenotypes caused by loss of PKA or Cos2 function in limb development suggest that these two regulatory events might be intimately related. For example, Cos2 might target Ci for efficient phosphorylation by PKA, allowing PKA to exert its negative regulation on Ci. Alternatively, phosphorylation of Ci by PKA might regulate the complex formation, allowing Cos2 to exert its influence on Ci. Further genetic and biochemical studies are required to resolve this important issue (G. Wang, 1999).

It has been proposed that phosphorylation of Ci by PKA allows Slimb to bind Ci and target it for ubiquitin/proteasome-mediated proteolysis. The epistatic relationship between PKA and Slimb defined by this study is consistent with this hypothesis. Moreover, it has been shown recently that proteasome is involved in Ci proteolytic processing. However, no evidence has been obtained to indicate that Ci is ubiquitinated. It is possible that the polyubiquitin chains added to Ci might be unstable and thus might escape detection. Alternatively, the proteolytic processing of Ci might not be directly mediated by ubiquitination, and Slimb might regulate Ci processing indirectly. For example, the so-called SCF (Skp1, Cdc53, and F-box) ubiquitin ligase complex (SCF^Slimb) might promote the ubiquitination and degradation of an inhibitor of a protease that cleaves Ci (G. Wang, 1999).

Another important question that remains largely unanswered is how Hh antagonizes PKA. The structural similarity between Smo and G protein-coupled seven-transmembrane receptors suggests that Hh signaling might antagonize PKA by down-regulating its cAMP dependent kinase activity. However, the observations that a constitutively active cAMP independent form of PKA (mC*) can rescue PKA mutant phenotypes without perturbing normal Hh signaling both in embryos and in imaginal discs strongly argue against this possibility. The finding that high but not low levels of mC* are able to override Hh signaling is more consistent with a model in which Hh and PKA act competitively and antagonistically on Ci. For example, Hh may activate a phosphatase that removes the phosphates added to Ci by PKA. In support of this view, pharmacological evidence suggests that Hh stimulates target gene expression via a PP2A-like phosphatase in tissue culture cells. However, there is no genetic evidence for the involvement of a phosphatase in the Hh pathway (G. Wang, 1999).

In vertebrates, Hh signaling is mediated by three members of the Gli family of transcription factors: Gli1, Gli2, and Gli3. Like Ci, all three Gli proteins contain multiple PKA phosphorylation consensus sites at conserved positions, so they are likely to be direct targets for PKA regulation in the vertebrate Hh signaling pathway. Among the three Gli proteins, Gli3 is both structurally and functionally related to Ci. Gli3 has been implicated to have both activator and repressor function depending on the developmental contexts. Moreover, PKA appears to promote Gli3 processing to generate a putative repressor form. Thus, the mechanism by which PKA targets Ci for Slimb-mediated processing may well be conserved from invertebrates to vertebrates. Gli1 and Gli2 appear to function mainly as positive regulators in the vertebrate Hh signaling pathway. Unlike Gli3 and Ci, Gli1 and Gli2 do not undergo PKA-dependent processing, however, their activities are likely to be regulated by PKA. For example, it has been shown that overexpression of a constitutive active form of PKA represses the transcriptional activity of Gli1 in mammalian culture cells. Thus, whereas the mechanism by which PKA regulates Ci processing may only apply to Gli3, the processing-independent inhibitory mechanism defined by this study may well apply to all three Gli proteins and is likely to be a more general mechanism by which PKA negatively regulates Hh signaling (G. Wang, 1999 and references therein).

Identification of a tetrameric hedgehog signaling complex

Hedgehog (Hh) signal transduction requires a large cytoplasmic multi-protein complex that binds microtubules in an Hh-dependent manner. Three members of this complex, Costal2 (Cos2), Fused (Fu), and Cubitus interruptus (Ci), bind each other directly to form a trimeric complex. This trimeric signaling complex exists in Drosophila lacking Suppressor of Fused [Su(fu)], an extragenic suppressor of fu, indicating that Su(fu) is not required for the formation, or apparently function, of the Hh signaling complex. However, Su(fu), although not a requisite component of this complex, does form a tetrameric complex with Fu, Cos2, and Ci. This additional Su(fu)-containing Hh signaling complex does not appear to be enriched on microtubules. Additionally, it has been demonstrated that, in response to Hh, Ci accumulates in the nucleus without its various cytoplasmic binding partners, including Su(fu). A model is discussed in which Su(fu) and Cos2 each bind to Fu and Ci to exert some redundant effect on Ci such as cytoplasmic retention. This model is consistent with genetic data demonstrating that Su(fu) is not required for Hh signal transduction proper and with the elaborate genetic interactions observed among Su(fu), fu, cos2, and ci (Stegman, 2000).

Suppressor of Fused opposes Hedgehog signal transduction by impeding nuclear accumulation of the activator form of Cubitus interruptus

It is proposed that, in the absence of Hh signaling, Su(fu) inhibits Ci by binding to it and that, upon reception of the Hh signal, Fu is activated and counteracts Su(fu), leading to the activation of Ci (Monnier, 1998). Since the Hedgehog signal transduction pathway is complex, involving multiple inputs to create the repressor and activator forms of ci, Ci[rep] and Ci[act] respectively, the components of the pathway must be manipulated independently in order to discover the role of Su(fu). In particular, neutralization of the effects of the kinase PKA is necessary in order to unravel the effects of the Fused kinase, and its interacting protein Su(fu). The absence of Cos2 or PKA activity prevents formation of the cleaved form of Ci, namely Ci[rep]. However mere prevention of Ci cleavage does not suffice for Ci[act] formation (Methot, 1999). To test whether, in these mutant contexts, prevention of Ci[rep] formation is linked to the generation of Ci[act], PKA or Cos2 were removed Ci[act] activity was assessed. While the repression of hh transcription is indicative for the presence of Ci[rep], the upregulation of ptc transcription serves as an indicator for the presence of Ci[act] (Methot, 1999). ptc expression was examined in A cells mutant for either cos2 or PKA; in both cases, ptc is upregulated. These results indicate that A cells mutant for cos2 or PKA generate Ci[act], and suggest that neutralization of the activities or effects of either Cos2 or PKA is an important step for the formation of Ci[act]. Interestingly, although the C terminus of Fu is required for Ci[rep] formation, its absence does not lead to ectopic ptc-lacZ expression and Ci[act] formation (Methot, 2000).

Therefore, Ci[act] is generated in cells that lack PKA activity. An equivalent situation can be created by mutating the PKA phosphorylation sites of Ci. One (Ci^PKA1) or four (CiPKA4) PKA phosphorylation sites were mutated and the ability of these mutants to activate ptc-lacZ in wing imaginal discs was compared. Ubiquitous weak expression of wild-type Ci leads to ptc-lacZ activation only in Hh-exposed cells. This indicates that under physiological conditions, transcriptional activity of Ci is under the control of Hh. In contrast, Ci^PKA1 activates ptc-lacZ in all cells, regardless of their exposure to Hh. Thus, Ci^PKA1 is constitutively active in vivo. Identical results have recently been described by several groups, and are taken as indication of a role for PKA in Ci activation. Interestingly, closer examination of discs ubiquitously expressing Ci^PKA1 reveals that P cells express ptc-lacZ at higher levels than A cells. This suggests that the transcriptional activity of Ci^PKA1 can be further enhanced by Hh signaling. Similar results were obtained with Ci^PKA4. To test whether it is the reception of the Hh signal that stimulates the basal activity of Ci^PKA, the function of Smo was removed from a subset of P cells expressing Ci^PKA4. Such cells transcribe ptc-lacZ at lower levels than their smo⁺ neighbors, levels that are similar to those found in anterior smo⁺ Ci^PKA cells. Thus Ci protein that is constitutively active due to mutated PKA phosphorylation sites is further stimulated in its transcriptional activity by the Hh signal (Methot, 2000).

It is possible that the reduction of ptc-lacZ expression in posterior smo^-;Ci^PKA4cells is due to formation of some Ci[rep], which could compete with the Ci[act] activity of Ci^PKA. The repressor assay described above was used to test whether Ci^PKA4 can be converted to a transcriptional repressor; hh transcription in smo^- posterior cells is essentially unaffected. This indicates that Ci^PKA4 cannot be converted to Ci[rep] and that the reduction of ptc expression in smo^- Ci^PKA4 cells is not due to the presence of Ci[rep] (Methot, 2000).

Components of the Hh pathway that could stimulate the basal activity of Ci^PKA were then sought. Fu is one of the few proteins that have a positive input on Hh signaling. ptc-lacZ levels were examined in wing imaginal discs that ubiquitously express Ci^PKA1, in either a wild-type or fu¹ background. ptc-lacZ expression is clearly reduced in P cells of fu¹ discs. Similar results have also been obtained with Ci^PKA4. Slightly elevated ptc-lacZ can still be seen near the AP compartment boundary in fu¹ discs, and may be the result of cumulative activities of endogenous Ci[act] and Ci^PKA. It is concluded that Fu kinase enhances the basal activity of Ci^PKA (Methot, 2000).

Beyond this basal activity, Fu stimulates Ci^PKA by inhibiting Su(fu) activity. fu is tightly linked to Su(fu), both genetically and biochemically. To test whether the modulation of Ci^PKA activity involves Su(fu), Ci^PKA4 was ubiquitously expressed together with myc-tagged Su(fu) or GFP as a negative control. Ci^PKA4 (in the presence of GFP) induces ptc-lacZ expression everywhere in the wing imaginal disc but at higher levels in the P compartment. Co-expression with mycSu(fu) abolishes ptc-lacZ expression in the A compartment and reduces ptc-lacZ levels in P cells. Thus Su(fu) inhibits the activity of Ci^PKA4. This result is strengthened by the converse experiment, where the absence of Su(fu) [in Su(fu)LP homozygous animals] reduces the difference in ptc-lacZ levels between A and P Ci^PKA1-expressing cells. To determine whether Su(fu) negatively acts on Ci^PKA4 by direct protein-protein interaction, a mutant form of Ci with impaired Su(fu) binding was created. Su(fu) interacts with Ci within a region that encompasses amino acids 244-346 (Monnier, 1998). Indeed, an N-terminal fragment of Ci (amino acids 5-440) interacts with GST-Su(fu). A deletion removing amino acids 212-268 of Ci almost abrogates Su(fu) binding to an N-terminal in vitro translated product of Ci. Removal of amino acids 268 to 346 also reduces Su(fu) binding, but to a lesser extent. The Delta212-268 deletion was introduced into Ci^PKA4, to create CiDeltaN^PKA4. This mutant is constitutively active, with P cells expressing higher ptc-lacZ levels than A cells. The activity of CiDeltaN^PKA4 is slightly reduced when introduced into a strong fu background, but the reduction is much less pronounced compared to that observed for Ci^PKA4. It is concluded that inhibition of Su(fu) activity by Fu kinase is an important step toward stimulating the basal activity of Ci^PKA4 (and by analogy Ci[act]) (Methot, 2000).

A possible mechanism by which Fu stimulates and Su(fu) counteracts Ci[act] could be the promotion or impediment of nuclear Ci[act] accumulation, respectively. The subcellular distribution of Ci^PKA in cells expressing or lacking Su(fu) was examined. Strikingly while wild-type cells show a cytoplasmic distribution of Ci^PKA4, this protein is mostly nuclear in Su(fu) mutant salivary gland cells. Identical results were obtained with wild-type Ci fused N terminally to GFP. This suggests that Su(fu) influences the nuclear localization of Ci. The effect of Su(fu) on Ci localization was further tested by overexpressing Su(fu) with a GFP-tagged form of Ci75, which has been shown to be mainly nuclear. Expression of mycSu(fu) reduces the amount of GFPCi75 that accumulates in the nucleus. These experiments were also performed with full-length Ci (CiGFP). Co-expression of Su(fu) in discs reduces the amount of nuclear CiGFP, both in A and P cells. It is concluded that Su(fu) downregulates the Hh pathway by preventing nuclear accumulation of Ci[act] (Methot, 2000).

It is suggested that Fu, rather than being involved in Ci[act] formation per se, stimulates the Hh pathway by permitting nuclear accumulation of Ci[act]. Su(fu) is not involved in the formation of Ci[rep] or Ci[act]. Rather, Su(fu) appears to restrict the activities of Ci[act]. This is evident from the observation that Su(fu) overexpression substantially curbs the transcriptional activity of constitutively active Ci^PKA, and is suggestive of Su(fu) acting after Ci[act] formation. There are several ways by which Su(fu) could fulfill such a role. One possibility is that it impedes entry of Ci[act] into the nucleus. Alternatively, Su(fu) might promote nuclear export of Ci[act]. It is difficult to distinguish between these two possibilities. The observation that little Su(fu) accumulates in the nuclei suggests that Su(fu) functions primarily in the cytoplasm and hence might exert a negative effect on Ci[act] by preventing its nuclear entry. It cannot be excluded, however, that a minor fraction of Su(fu) negatively affects the activity, stability or localization of Ci[act] in the nucleus (Methot, 2000).

Fu, as the main regulator of Su(fu) activity, is also controlled by Hh. In fu¹ discs, Ci^PKA expression leads to similar levels of ptc transcription in A and P cells but, in fu⁺ discs, Ci^PKA expression causes higher ptc levels in P cells. In other words, Fu enhances CiPKA activity only in Hh-exposed cells. From this, it can be concluded that Fu activity is subject to Hh control (Methot, 2000).

One puzzling aspect regarding Su(fu) is that it is dispensable for viability. Animals that lack Su(fu) protein do not exhibit Hh-independent Ci[act] activity. This paradox can be partly explained by viewing Su(fu) only as a partial inhibitor of Ci[act] activity, which exerts its function subsequent to Ci[act] formation. Other elements ensure tight control over the generation of Ci[act]. The problem of how full-length Ci protein is converted into Ci[act] is more challenging. Fu has been implicated in this process, but as in the case of Su(fu), Fu kinase activity is partially dispensable in wild-type discs and entirely dispensable in animals lacking Su(fu). This suggests that the Fu kinase functions only to prevent Su(fu) from negatively acting on the Hh pathway. If we accept that Su(fu) acts subsequent to Ci[act] formation, it must be concluded that the same is true for the Fu kinase. In short, it is proposed that the activity of the Fu kinase is only required to maximize the output of an already activated form of Ci, for example by opposing cytoplasmic tethering of Ci[act] by Su(fu). The precise mechanism of how these components act is not understood. No substrate for the Fu kinase has been identified and the significance of nuclear Fu protein is unclear (Methot, 2000 and references therein).

PKA and Cos2 prevent Ci[act] formation and the same components are required for Ci[rep] formation (Methot, 2000). This observation closely links the two events. Cos2, Fu and Ci are found in a large cytoplasmic complex that is associated with microtubules. Fu derived from type II alleles, lacking the C-terminal portion, fails to locate to this complex. Indeed, Ci[rep] is not generated in cells expressing only type II-mutant Fu protein. In addition, exposure to Hh releases Cos2 from microtubules. This links Ci[rep] formation to complex formation. Together, these findings lead to the idea that complex formation fulfills two roles: one is to tether Ci to microtubules, thereby preventing nuclear entry. The other is to localize Ci to the site of proteolytic processing for the formation of Ci[rep]. Hh signaling would promote the formation of Ci[act] by releasing this complex (or Ci) from microtubules, and as a consequence would prevent the cleavage of Ci. Upon release, Ci[act] would be subjected to Su(fu) action, possibly by cytoplasmic tethering. Stimulation of Fu kinase activity by Hh inhibits Su(fu) and enables nuclear accumulation of Ci[act]. A challenging question to be answered is whether the Hh-dependent events are all catalyzed by a single biochemical step (Methot, 2000 and references therein).

Interactions with Costal2 and Suppressor of fused regulate nuclear translocation and activity of Cubitus interruptus

Costal2 (Cos2) and Suppressor of Fused [Su(fu)] inhibit Ci by tethering it in the cytoplasm, whereas Hh induces nuclear translocaltion of Ci through Fused (Fu). A 125 amino acid domain in the C-terminal part of Ci has been identiifed that mediates response to Cos2 inhibition. Cos2 binds Ci, prevents its nuclear import, and inhibits its activity via this domain. Su(fu) regulates Ci through two distinct mechanisms: (1) Su(fu) blocks Ci nuclear import through the N-terminal region of Ci, and (2) it inhibits the activity of Ci through a mechanism independent of Ci nuclear translocation. Cos2 is required for transducing high levels of Hh signaling activity, and it does so by alleviating the blockage of Ci activity imposed by Su(fu) (G. Wang, 2000).

Wild-type wing discs accumulate Ci in the nucleus in Hh receiving cells after treatment with Leptomycin B (LMB), a drug that blocks CRM1 dependent nuclear export. Ectopic hh expression in anterior (A) compartment cells away from the compartment boundary induces LMB-dependent nuclear translocation of Ci in these cells. The stimulation of LMB-dependent nuclear import by Hh appears to be much more efficient in the developing imaginal discs than in cultured cl-8 cells. One possible explanation is that cl-8 cells might not fully recapitulate all the Hh signaling properties. The ability of LMB to block nuclear export of Ci in cultured imaginal discs provides an opportunity to address the roles of Cos2 and other Hh signaling components in regulating Ci nuclear import. cos2 mutation results in constitutive nuclear translocation of Ci independent of Hh signaling. In contrast, fu mutation attenuates Ci nuclear translocation induced by Hh. Taken together, these experiments show that Cos2 and Hh have opposing influences on Ci nuclear import: Cos2 exerts a block on Ci nuclear translocation, whereas Hh stimulates Ci nuclear translocation through Fu (G. Wang, 2000).

Using deletion analysis coupled with in vivo coexpression assays, a 125 amino acid domain has been identified in the C-terminal part of Ci (aa 961-1065) that mediates transcriptional repression and cytoplasmic retention by Cos2. This domain has been named CORD for Cos2 responsive domain. Ci deletion mutants that lack CORD are insensitive to Cos2 repression and are no longer sequestered in the cytoplasm by Cos2 in these assays. Moreover, CORD is sufficient to mediate Cos2-dependent cytoplasmic retention when fused to a heterologous protein. In yeast two hybrid assay, CORD is found to be the only region of Ci that binds Cos2. Taken together, these data provide strong evidence that Cos2 inhibits Ci activity by tethering it in the cytoplasm via directly binding to CORD (G. Wang, 2000).

A Ci region from aa 703 to aa 850 can act to sequester heterologous proteins in the cytoplasm. However, this region does not mediate cytoplasmic retention by Cos2 because Ci deletion mutants that retain it fail to be sequestered by Cos2 and are resistant to Cos2 inhibition in the in vivo assay. Moreover, Ci fragments containing this region fail to bind Cos2 in yeast. Rather, this Ci domain appears to mediate Ci nuclear export as its effect on nuclear localization is abolished by LMB treatment (G. Wang, 2000).

The mechanism by which Su(fu) inhibits Hh signaling has remained controversial. Overexpression studies using mammalian cultured cells have shown that Su(fu) can sequester Gli1 in the cytoplasm. However, a different result was obtained from overexpression study using Drosophila cultured cells. For example, overexpressing Su(fu) in Drosophila cl-8 cells fails to block LMB-induced nuclear accumulation of Ci. In addition, several studies have revealed that Su(fu) can interact with Gli1 on DNA, raising the possibility that Su(fu) might affect Gli activity in the nucleus. In this study, genetic evidence is provided that Su(fu) regulates Ci/Gli by both blocking its nuclear import and affecting its activity after nuclear translocation. To overcome the problem of Ci instability in Su(fu) mutant cells, the effect of Su(fu) mutation was examined on nuclear translocation of overexpressed Ci that appears to saturate the mechanism responsible for degrading Ci in the absence of Su(fu). Overexpressed Ci is significantly retained in the cytoplasm in A compartment cells of wild-type wing discs but is largely accumulated in the nucleus in Su(fu) mutant wing discs. Removal of Su(fu) binding domain has a similar effect on Ci nuclear translocation to Su(fu) mutation, suggesting that Su(fu) sequesters Ci in the cytoplasm by directly binding to the N-terminal region of Ci. The ability of Su(fu) to sequester Ci in the cytoplasm appears to depend on Cos2, as Su(fu) does not prevent LMB-dependent Ci nuclear import in cos2 mutant cells (G. Wang, 2000).

Evidence arguing that Su(fu) affects Ci transcriptional activity in the nucleus comes from analysis of cos2 mutant phenotypes. In wild-type wing discs, A compartment cells abutting the A/P compartment boundary transduce high levels of Hh signaling activity; these high levels convert Ci into a labile transcription activator by antagonizing the inhibitory role of Su(fu). As a consequence, these cells activate en and show low levels of Ci staining. cos2 mutant cells abutting the compartment boundary accumulate high levels of Ci and show low levels of Hh signaling activity as they fail to activate en. Thus, it appears that the majority of Ci in cos2 mutant cells remains in a latent stable form, likely in a complex with Su(fu). In support of this view, it has been shown that removal of Su(fu) from cos2 mutant cells restores high levels of Hh signaling activity and simultaneously decreases the concentration of Ci in these cells. Because Hh induction of Ci nuclear import is not affected by Su(fu) in cos2 mutant cells near the A/P compartment boundary, it is concluded that Su(fu) inhibits Ci activity at a step after it translocates into the nucleus. A possible mechanism by which Su(fu) inhibits Ci activity in the nucleus is to prevent it from forming an active transcriptional complex, since it has been shown that Su(fu) can interact with Gli on DNA (G. Wang, 2000).

Cos2 was identified as a negative component in the hh pathway by previous genetic studies. A novel, positive role for Cos2 in the hh pathway has now been uncovered. In addition to blocking Hh signal transduction in A compartment cells away from the compartment boundary, Cos2 is required for transducing high levels of Hh signaling activity by antagonizing Su(fu) in A compartment cells near the A/P compartment boundary. In addition, this requirement is a general property of Cos2 that applies to all A compartment cells. Thus, these results underscore an unusual relationship between Cos2 and Su(fu): in the absence of Hh signaling, Cos2 acts cooperatively with Su(fu) to block Ci nuclear import by forming a complex with Ci; in cells receiving high dose of Hh signal, Cos2 is required to alleviate the block on Ci transcriptional activity imposed by Su(fu) (G. Wang, 2000).

Based on the evidence presented here and elsewhere, a working model for how Cos2, Fu and Su(fu) regulate the nuclear translocation and activity of Ci is proposed. Su(fu) and Cos2 bind Ci via the N- and C-terminal domains, respectively, and the complex binds microtubules through Cos2 and retains Ci in the cytoplasm. In addition, Cos2 promotes the proteolysis of Ci to generate a truncated repressor form (Ci75), a process that also requires the activities of PKA, Slimb, and proteasome. Hh stimulates Ci nuclear translocation through Fu kinase and inhibits Ci processing possibly through dephosphorylating Ci. The transcriptional activity of full-length Ci is attenuated in the nucleus by Su(fu), which also stabilizes the latent form of Ci. High levels of Hh signaling activity convert Ci into a labile and active form, possibly by dissociating it from Su(fu), and this process requires the activities of Fu and Cos2 (G. Wang, 2000).

Several important issues regarding this model need to be addressed. For example, how Hh antagonizes Cos2 and Su(fu) to promote Ci nuclear translocation remains an important unsolved problem. It is likely that Hh stimulates Ci nuclear import by dissociating Ci complex from microtubules and, further, by releasing Ci from the complex. In support of this view, it has been shown that Hh can induce dissociation of Cos2 from microtubules. Moreover, it has also been implicated that dissociation of Ci tetrameric complex might proceed the nuclear translocation of Ci. However, no biochemical evidence has been obtained indicating that Hh induces dissociation of Ci from Cos2 and Su(fu) (G. Wang, 2000).

Fu kinase appears to be required for Hh to stimulate Ci nuclear translocation, because Ci is retained significantly in the cytoplasm in fu mutant cells that receive Hh signal. The substrate for Fu kinase still remains a mystery. One attractive candidate is Su(fu), which binds Fu and whose function is antagonized by Fu. Since Su(fu) is a PEST domain protein, phosphorylation of Su(fu) might cause its degradation and subsequent disassembly of Ci complex. Another good candidate for a Fu substrate is Cos2, which also interacts with Fu. It has been shown that Hh induces phosphorylation of Cos2; however, the kinase responsible for Hh-dependent phosphorylation of Cos2 has not been identified. It remains to be determined if Fu contributes to Cos2 phosphorylation. Although the biological significance of Cos2 phosphorylation has not been shown yet, it is conceivable that such phosphorylation could cause dissociation of Cos2 from microtubules or from Ci, leading to Ci nuclear translocation. In support of this view, it has been found that the effect of fu mutation on Ci nuclear translocation can be suppressed by removal of Cos2, arguing that Fu promotes Ci nuclear import by antagonizing Cos2 (G. Wang, 2000).

Finally, how Cos2 positively regulates Hh signaling activity remains to be determined. The finding that Cos2 is required for Hh to antagonize Su(fu) could be explained by the observation that Cos2 forms a complex with Fu and Su(fu). One scenario is that Cos2 might simply play a structural role in which it antagonizes Su(fu) by recruiting Fu. Alternatively, Cos2 might play a more active role in which it recruits other positive components in close proximity to Fu while also regulating Fused activity in response to Hh signaling. Structure and functional analysis of Cos2 and identifying other Cos2 interacting proteins may help to resolve this important issue (G. Wang, 2000).

Homeotic proboscipedia function acts throught Ci to modulate hedgehog-mediated organizer activity to pattern adult Drosophila mouthparts

Drosophila proboscipedia (HoxA2/B2 homolog) mutants develop distal legs in place of their adult labial mouthparts. How pb homeotic function distinguishes the developmental programs of labium and leg has been examined. The labial-to-leg transformation in pb mutants occurs progressively over a 2-day period in mid-development, as viewed with identity markers such as dachshund (dac). This transformation requires hedgehog activity, and involves a morphogenetic reorganization of the labial imaginal disc. These results implicate pb function in modulating global axial organization. Pb protein acts in at least two ways. (1) Pb cell autonomously regulates the expression of target genes such as dac; (2) Pb acts in opposition to the organizing action of hedgehog. This latter action is cell-autonomous, but has a nonautonomous effect on labial structure, via the negative regulation of wingless and decapentaplegic. This opposition of Pb to hedgehog target expression appears to occur at the level of the conserved transcription factor cubitus interruptus/Gli that mediates hedgehog signaling activity. These results extend selector function to primary steps of tissue patterning, and leads to the notion of a homeotic organizer (Joulia, 2005).

The labial palps, the drinking and taste apparatus of the adult fly head, are highly refined ventral appendages homologous to legs and antennae. As for most adult structures, these mouthparts are derived from larval imaginal discs, the labial discs. Wild-type pb selector function acts together with a second Hox locus, Scr, to direct the development of the labial discs giving rise to the adult proboscis. In the absence of pb activity, the adult labium is transformed to distal prothoracic (T1) legs, reflecting the ongoing expression and function of Scr in the same disc. Though the pb locus shows prominent segmental embryonic expression, as for the other Drosophila homeotic genes of the Bithorax and Antennapedia complexes, it is unique in that it has no detected embryonic function and null pb mutants eclose as adults that are unable to feed. Thus, normal pb selector function is required relatively late, in the labial imaginal discs that proliferate and differentiate during larval/pupal development to yield the adult labial palps. Though the genetic pathway guiding development of the ventral labial imaginal discs to adult mouthparts remains relatively unexplored both in flies and elsewhere, study of P-D patterning has identified several genes subject to pb regulation in the labial discs (notably Dll, dac, and hth) and a distinct organization of normal labial discs has been indicated compared to other imaginal discs (Joulia, 2005).

This study pursued an investigation of how pb homeotic function distinguishes between labial and leg developmental programs. The results implicate pb function at the level of global axial organization. Employing identity markers such as dachshund (dac), a 2-day period late in larval development has been identified when normal pb function is required for labial development. The labial-to-leg transformation occurs during the third larval instar stage, involves a progressive morphogenetic reorganization of the labial imaginal disc, and is hedgehog-dependent. This analysis of the transformation indicates that normal pb action is required at least at two distinct levels. One is in the cell-autonomous regulation of target genes such as dac likely to be implicated in cell identity. A second level involves an autonomous action with a nonautonomous effect on labial structure, through the negative regulation of wingless and decapentaplegic downstream of hh signaling. This opposition to hh targets is likely to occur at the level of the transcription factor cubitus interruptus/Gli, a crucial and conserved mediator of hh signaling activity. These results led to a proposal that homeotic function may exist in intimate functional contact with the hedgehog organizer signaling system: the 'homeotic organizer' (Joulia, 2005).

Segmental organization in the imaginal discs involves the reiterated deployment of segment polarity genes that organize the fundamental segmental form. This involves a cascade proceeding from posteriorly expressed Engrailed protein through a short-range Hh morphogen gradient in anterior cells favoring the activator form of Ci transcription factor, which in turn activates wg and dpp to establish two concurrent, instructive concentration gradients that structure gene expression along the proximo-distal axis. In contrast with this elaborate choreography of the segment polarity genes, the homeodomain proteins encoded by Hox genes are expressed in a segmental register, which obscures how they can direct the differentiation of distinct cell types within the segment. The present investigation of homeotic proboscipedia function during labial palp formation indicates a multipronged action for pb in the labial disc. Pb acts cell-autonomously in the negative regulation of target genes including dac, which is normally extinguished in Pb-expressing cells of labial or leg imaginal discs but is activated in labial discs in the absence of pb activity. This activation of dac in mutant labial cells is hh-dependent and is likely a response to wg and dpp morphogen signals as for leg discs. The data further indicate that pb acts cell autonomously to regulate the level of both wg and dpp expression in response to hh. Thus, pb appears to negatively regulate dac expression directly, but also by withholding positive instructions from Wg and Dpp morphogens. The interweaving of homeotic selector proteins with strategic target genes including morphogens (wg, dpp) and targets of signaling activity (dac, Dll) may influence segment patterning from global size and shape to specific local pattern and cell identity. This positioning offers a powerful yet economical mode of selector function that helps to better understand how a single selector gene can integrate global patterning with cellular identity (Joulia, 2005).

This view invoking multiple and overlapping modes of regulation by a homeotic selector protein supports and extends the vision from analyses seeking to explain how Ultrabithorax (Ubx) selector function differentiates between the serially homologous wing and haltere appendages. This analysis supports a role for Ubx in fruit flies transforming a dorsal default state (wing) to haltere, by repressing the accumulation of Wg in the posterior part of the haltere, and by regulating a subset of Dpp or Wg activated targets such as vestigial and spalt related. Additionally, clear evidence has been presented for a nonautonomous action of Ubx via its activity in cells of the D-V organizer where wg is expressed. Ubx thus acts to down-regulate wg in the haltere, but also intervenes to modulate the expression of targets of both dpp and wingless signaling pathways. An analysis of mutants for maxillopedia (mxp), the Tribolium pb homolog, revealed augmented transcription of flour beetle wg within the transformed labial segment. This observation, in full accord with the above results for Ubx, and the current results for Drosophila pb, supports a conserved role for homeotic regulation of nonautonomous signaling input in appendage development. At the same time, mxp mutants show a precocious maxilla-to-leg transformation in larvae, demonstrating a prior, embryonic requirement for mxp. This result is of particular interest since it highlights a temporal aspect of pb action in the fly labial disc: the absence of pb function early has no apparent effect on the labial discs in early L3 larvae, which appear normal. It is only subsequently that these diverge toward leg structure. Thus, the globally conserved activity of mxp/pb in equivalent beetle or fly organs is nonetheless employed in temporally different ways among species. Though it is not clear whether this reflects the existence of species-specific co-factors or rather of the effects of expression dosage and timing, such modifications might offer important possibilities for changing form. Variations on all these themes can probably contribute to the diversification of organism form, within and among species (Joulia, 2005)

The roles of diffusible Wg and Dpp morphogens induced by Hh at the A-P boundary, and the transcriptional programs they induce according to their concentrations within a gradient, are considered central to organizing the group of cells constituting a segment. The present work indicates that pb normally acts downstream of Hh within the organizer, where it maintains Wg and Dpp at low levels in labial imaginal tissue. Overexpressing Wg or Dpp in the labial discs results in malformed, overgrown or transformed 'labial' tissue. These observations support the viewpoint that limiting morphogen accumulation is essential to ensuring that the labial program is correctly applied. This study underlines the potential importance of the absolute levels of wg and dpp-encoded signaling molecules deployed for tissue organization. While a gradient may in principle be formed from any source, part of the spectrum of threshold levels necessary for stimulating specific gene responses is likely removed from the repertoire in the labial environment. The absolute level of activation or inhibition of diverse signaling pathways thus may be in itself a tissue-specific property, allowing gradients of related form but with different instructive capacities that can be a distinctive element in guiding tissue formation and specifying ultimate identity. This integration of diverse sorts of information -- the hh organizer linked to the Hox selector -- may confer order to tissue organization and identity (Joulia, 2005).

The fine-tuning of morphogen signals by Hox selectors coupled with the concomitant regulation of downstream targets thus appears to offer a strategic control point for achieving reliable developmental control coupled with evolutionary flexibility. The modulation of different cell signaling pathways by pb activity implies it can regulate both the tissue “context” generated by the signaling pathways activated in a tissue, and the cellular response to this context. This capacity to meld large-scale patterning with cellular identities merits emphasis (Joulia, 2005).

While the logic described above appears to be conserved, its application leads to widely different results according to the species and the tissue. Quite recently, an analysis of vertebrate Hox function has led to the identification of an intimate developmental link between Hox selector function and hedgehog signaling. This analysis reveals a direct physical interaction between the mouse Ci homolog Gli and Hox homeodomain transcription factors. It thus provides a compelling complement to the present work, since the molecular framework of a direct link between Gli and Hox proteins goes far to rationalise the dose-sensitive interplay between Ci and Pb that was observed in Drosophila. If Hox proteins indeed compete for available nuclear Gli/Ci, this molecular mechanism may also help to understand other phenomena including phenotypic suppression in flies or posterior prevalence in mice. Correspondingly, the current data place Pb in antagonism to Ci within the hedgehog organizer, where it modulates output from the wg and dpp genes and the instructive morphogens they encode. These complementary observations from insect and vertebrate models suggest the existence of an evolutionarily conserved machinery with enormous potential for generating morphological diversity. It will be exciting to know more about how the homeotic selector function is integrated in known cascades that make use of conserved molecules both to ensure the fidelity of normal form, as well as to generate new form (Joulia, 2005).

Hedgehog-regulated Costal2-kinase complexes control phosphorylation and proteolytic processing of Cubitus interruptus

Hedgehog (Hh) proteins control animal development by regulating the Gli/Ci family of transcription factors. In Drosophila, Hh counteracts phosphorylation by PKA, GSK3, and CKI to prevent Cubitus interruptus (Ci) processing through unknown mechanisms. These kinases physically interact with the kinesin-like protein Costal2 (Cos2) to control Ci processing and Hh inhibits such interaction. Cos2 is required for Ci phosphorylation in vivo, and Cos2-immunocomplexes (Cos2IPs) phosphorylate Ci and contain PKA, GSK3, and CKI. By using a Kinesin-Cos2 chimeric protein that carries Cos2-interacting proteins to the microtubule plus end, it was demonstrated that these kinases bind Cos2 in intact cells. PKA, GSK3, and CKI directly bind the N- and C-terminal regions of Cos2, both of which are essential for Ci processing. Finally, it was shown that Hh signaling inhibits Cos2-kinase complex formation. It is proposed that Cos2 recruits multiple kinases to efficiently phosphorylate Ci and that Hh inhibits Ci phosphorylation by specifically interfering with kinase recruitment (Zhang, 2005).

To facilitate detection of protein-protein interaction between Cos2 and its binding partners in vivo, a Kinesin/Cos2 chimeric protein (Kinco) was generated in which the microtubule binding domain of Cos2 is replaced by the motor domain of Drosophila KHC. Kinco moves to the microtubule plus end and accumulates near the basal surface of imaginal disc epithelial cells. Strikingly, Kinco carries all the known Cos2 binding proteins to the same subcellular compartment, leading to colocalization. PKAc, GSK3, and two CKI isoforms, CKIα and CKIϵ, all colocalize with Kinco at the microtubule plus end, demonstrating that these kinases associate with Cos2 in intact cells. Hence, Kinco provides a powerful tool to determine if a protein interacts with Cos2 in vivo. In addition, Kinco colocalizes with Cos2-interacting proteins in cultured Drosophila cells such as S2 and cl8 cells. It is conceivable that one can use such a cell-based colocalization assay to identify additional proteins that form a complex with Cos2. Furthermore, it is also possible to extend this approach to other protein complexes by generating appropriate Kinesin chimeric 'bait' proteins (Zhang, 2005).

By using immunoprecipitation and GST pull-down assays, the kinase interaction domains were mapped to the microtubule binding (MB) and C-terminal (CT) of Cos2. GST fusion proteins containing either of these domains bind purified recombinant PKAc, GSK3, and CKI, suggesting that these kinases directly bind Cos2. However, the possibility cannot be rule out that these kinases may have additional contacts with other components in the Cos2 complex. Indeed, it was found that CKI can bind Ci in yeast (Zhang, 2005).

Several lines of evidence suggest that Cos2/kinase interaction plays an important role in regulating Ci phosphorylation and processing: (1) Ci phosphorylation is compromised in cos2 mutants; (2) the kinase-interacting domains in Cos2 are essential for Ci processing; (3) overexpressing multiple kinases can bypass the requirement of Cos2 for Ci processing (Zhang, 2005).

PKAc, GSK3, and CKI appear to bind competitively to Cos2; however, since Cos2 can dimerize and each Cos2 protein contains two kinase binding domains, a Cos2 dimer could in principle bind all three kinases simultaneously. It is possible that these kinases might not form a tight complex with Cos2 in a stoichiometric manner, which could explain why purification of endogenous Cos2 complexes failed to identify any of these kinases. However, by using in vitro kinase assay and Western blot analysis, the association of PKAc, GSK3, and CKI with endogenous Cos2 was detected. It is likely that interactions between Cos2 and kinases are transient; however, such interactions could increase local concentrations of these kinases; this greatly facilitates Ci hyperphosphorylation (Zhang, 2005).

It has been demonstrated that Hh induces Ci dephosphorylation in cl8 cells; however, it is not clear whether Hh blocks Ci phosphorylation by all three kinases or a subset of them. By using an antibody that specifically recognizes a phosphorylated PKA site in Ci, it was found that Hh partially inhibits PKA phosphorylation of Ci in wing discs. Consistent with this, Hh only partially blocks Cos2/PKA interaction. In contrast, Hh appears to have a more profound influence on the interaction between Cos2 and CKI or GSK3. Furthermore, CKI and GSK3 kinase activities associated with endogenous Cos2 diminishes upon Hh stimulation and Cos2IPs phosphorylates Ci to a lesser extent after Hh treatment. These observations suggest that Ci phosphorylation by CKI and GSK3 is likely to be inhibited by Hh in vivo (Zhang, 2005).

Several mechanisms may contribute to the regulation of Cos2-Ci-kinase complex formation by Hh. (1) The finding that PKAc, GSK3, and CKI bind Cos2 domains that also interact with Smo raises a possibility that Smo/Cos2 interaction may exclude kinases from binding to Cos2. Indeed, a membrane-tethered form of SmoCT (Myr-SmoCT) interferes with Cos2-Ci-kinase complex formation. (2) Smo/Cos2 interaction at the cell surface may induce conformational change in Cos2, which could mask its kinase interacting domains. (3) Cos2 is phosphorylated in response to Hh. Phosphorylation of Cos2 could regulate its interaction with one or more kinases. (4) There is evidence that Hh induces dissociation of Ci from Cos2, which may further decrease the accessibility of Ci to the kinases. This may explain why Hh induces more significant dissociation of PKAc from Ci than from Cos2. (5) Hh induces degradation of Cos2 in P compartment cells as well as in cells immediately adjacent to the A/P boundary; this may lead to a chronic destruction of Cos2-Ci-kinase complexes. However, it appears that only high levels of Hh induce Cos2 degradation in vivo. Low levels of Hh may prevent Ci phosphorylation with different mechanisms such as those described above (Zhang, 2005).

The following model is proposed for the regulation of Ci phosphorylation by Cos2 and Hh. In the absence of Hh, Cos2 scaffolds multiple kinases and Ci into proximity, thus increasing the accessibility of Ci to these kinases and facilitating extensive phosphorylation of Ci. Upon Hh stimulation, Cos2 complexes are recruited to the cell surface via Smo, leading to disassembly of Cos2-Ci-kinase complexes. As a consequence, Ci phosphorylation is compromised and Ci processing is blocked. This model has several interesting parallels to that proposed for the Wnt pathway. In quiescent cells, both pathways employ large protein complexes to bring kinases and their substrates in close proximity, resulting in phosphorylation and proteolysis of the transcription factor (Ci) or effector (β-catenin). Upon ligand stimulation, both pathways recruit the cytoplasmic signaling complex to the cell surface and cause dissociation of the complex, leading to dephosphorylation and stabilization of the transcription factor/effector. Interestingly, both pathways use common kinases, including GSK3 and CKI. However, these kinases together with their substrates form distinct signaling complexes assembled by pathway-specific scaffolding proteins (Cos2 and Axin in the Hh and Wnt pathways, respectively). Pathway activation is achieved by a specific interaction between the receptor system and the scaffolding protein (Smo/Cos2 interaction in the Hh pathway and LPR5/6/Axin interaction in the Wnt pathway). Thus, each pathway only controls the pool of kinases in the same complex with the pathway effector, leading to pathway-specific regulation of substrate phosphorylation. The combinatorial mechanism by which pathway-specific scaffolds bring common kinases into proximity with their substrates thus appears to be a general one and may apply to other signaling pathways that utilize a common set of kinases (Zhang, 2005).

Sex lethal is part of the Hedgehog signaling complex: Rate of nuclear entry of full-length Cubitus interruptus are enhanced by Sxl

Sex-lethal (Sxl), the Drosophila sex-determination master switch, is on in females and controls sexual development as a splicing and translational regulator. Hedgehog (Hh) is a secreted protein that specifies cell fate during development. Sxl protein has been shown to be part of the Hh cytoplasmic signaling complex and Hh promotes Sxl nuclear entry (Vied, 2001; Horabin, 2003). In the wing disc anterior compartment, Patched (Ptc), the Hh receptor, acts positively in this process. This study shows that the levels and rate of nuclear entry of full-length Cubitus interruptus (Ci), the Hh signaling target, are enhanced by Sxl. This effect requires the cholesterol but not palmitoyl modification on Hh, and expands the zone of full-length Ci expression. Expansion of Ci activation and its downstream targets, particularly decapentaplegic the Drosophila TGFß homolog, suggests a mechanism for generating different body sizes in the sexes; in Drosophila, females are larger and this difference is controlled by Sxl. Consistent with this proposal, discs expressing ectopic Sxl show an increase in growth. In keeping with the idea of the involvement of a signaling system, this growth effect by Sxl is not cell autonomous. These results have implications for all organisms that are sexually dimorphic and use Hh for patterning (Horabin, 2005).

Drosophila Hh is synthesized as a 45 kDa precursor that is shortened to a mature form with two lipid modifications; palmitic acid at the N terminus and cholesterol at the C terminus. Maturation involves autoproteolytic processing under the control of the C-terminal domain of Hh. To test whether either of the lipid modifications plays a role in Hh promoted Sxl nuclear entry, female wing discs expressing Hh with only a single modification were examined. HhN encodes the N-terminal region of Hh that is palmitoylated but, because it does not undergo autoproteolytic processing, does not contain the cholesterol moiety. This form of Hh is functional for Ci activation and full-length Ci is detected distantly anterior of the AP boundary. Where HhN levels are maximal, there is a reduction of full-length Ci, most likely from the activation of en, which inhibits Ci transcription. HhN does not increase Sxl nuclear levels, however. The normal high nuclear levels in the posterior compartment and graded nuclear localization in the anterior compartment (Horabin, 2003) are detected, with no change in the cells expressing HhN (Horabin, 2005).

The alternative single modification [cholesterol without palmitoyl (C84S-Hh)], by contrast, is active with respect to Sxl. The dppGAL4 driver was used to drive expression of C84S-Hh. Relative to endogenous Hh, about threefold less of the nuclear export inhibitor Leptomycin B (LMB) was required to detect nuclear Sxl in these discs, suggesting that the nuclear Sxl is effected primarily by the ectopic Hh (Horabin, 2005).

C84S-Hh has been shown to dominantly destabilize Ci, decreasing the expression of Hh target genes. Patterning of the wing is compromised and the size of the region between veins L3 and L4 is reduced. C84S-Hh is also unable to rescue the embryonic segmentation phenotype caused by loss of Hh. C84S-Hh destabilizes Ci, but only in males. Females show the opposite effect, increasing the levels of full-length Ci (Horabin, 2005).

This sex specificity, coupled with the observation that Sxl is present in the Hh cytoplasmic complex, suggests that Sxl may be acting to stabilize Ci on Hh signaling. If this is the case, expressing Sxl in males should increase the levels of full-length Ci. Indeed, male discs expressing Sxl (MS3 isoform), as well as C84S-Hh under the control of dppGAL4, now show higher levels of full-length Ci and the protein is more nuclear, as seen in females. Taken together, these results suggest that when the cholesterol moiety is present on Hh, Sxl enhances the production of full-length Ci (Horabin, 2005).

Curiously, the presence of Sxl does not temper the wing patterning defect caused by the ectopic expression of C84S-Hh; the reported narrowing between wing veins L3 and L4 is the same in the two sexes. The form of Ci that Sxl stabilizes through C84S-Hh must not be the form responsible for Hh patterning (Horabin, 2005).

The data presented here show that when Sxl is present, the Hh signal is augmented. This is seen as an increase in full-length Ci in whole-mount tissue, and in Western blots which give a more quantitative sense of protein levels. In addition to elevating the levels of full-length Ci, several of the Hh downstream targets, including ptc, dpp and some of the downstream targets of Dpp, show an increase in expression. Conversely, removal of Sxl in female cells shows a reduction in the strength of the Hh signal (Horabin, 2005).

Sxl also enhances the nuclear entry rate of Ci, with either endogenous Hh or Hh that has only the cholesterol modification. In females, when Sxl is co-expressed with Hh with only the cholesterol modification, the amount of LMB required to detect nuclear Ci is reduced (by almost sixfold), further supporting the idea that Sxl affects Ci nuclear entry rate on Hh signaling (Horabin, 2005).

Hh enhancement of Sxl nuclear entry also depends on the cholesterol and not the palmitoyl modification. Given that Ci and Sxl are in a complex in the cytoplasm and both respond to the Hh cholesterol modification, it is tempting to speculate, although the data presented does not address this issue, that the two proteins may also enter the nucleus as a complex. This may be the method by which Sxl stabilizes Ci, diverting it from rapid proteolysis, particularly the highly activated form that is functionally detectable but has not been identified biochemically (Horabin, 2005).

Stabilization of full-length Ci by Hh with only the cholesterol modification in females is in contrast to what occurs in males. this form of Hh can destabilize Ci as well as compromise the Hh signal, but only in males. The effect of the cholesterol moiety contrasts with the palmitoyl that potentiates Hh in activating Ci for patterning. This is generally also true in vertebrates, where the cholesterol modification appears to have less of a role in patterning and a more significant role in the release and extracellular transport of the Hh ligand (Horabin, 2005).

In both sexes, ectopic expression of Sxl shows an increase in intensity of ptc expression, indicating it is possible to further elevate the Hh response. Other than en, which was difficult to score in these experiments, ptc requires the highest levels of Ci activation for its transcription (Horabin, 2005).

In females, the ectopic Sxl elevates ptc expression in the cells near the AP boundary, but the depth of the cells showing this highest level of Ci activation is reduced. A reduction in the number of cells transcribing ptc, when compared with the wider but less intense width of ptc transcription in the control half of the disc, suggests a restriction in Hh diffusion. Elevated ptc transcription is expected to produce more Ptc at the membrane, which should sequester more Hh close to the AP boundary. This result shows that Sxl can both enhance the Hh response and effectively alter the Hh gradient (Horabin, 2005).

In males, the increase in ptc transcription induced by Sxl both intensifies and widens the ptc expression zone. This suggests that the activation of Ci is at a lower peak in males than in females, and its enhancement by ectopic Sxl does not reach the same maximum that additional Sxl in females produces (Horabin, 2005).

Ectopic expression of Sxl in the dpp expression zone has been shown to adversely affect female wing development, narrowing the region between veins L3 and L4. This defect was taken to suggest that the relative concentrations of both Ci and Sxl are important for their normal function (Horabin, 2003). The data presented in this study support this conclusion while providing an explanation for the apparent decrease in effectiveness of the Hh signal. When additional Sxl is expressed, the slope of the Hh gradient becomes steeper. Since Hh directly patterns the L3 to L4 wing vein region, a steeper gradient of Hh will reduce the area patterned because the normal Hh patterning minimum is reached more rapidly. The L3 to L4 intervein region should correspondingly become narrower. No adult males expressing Sxl were recovered (presumably because of upsets in dosage compensation) so their wings could not be scored (Horabin, 2005).

Depending on the expression driver used, ectopic Sxl is not only lethal to males but also females. This is perhaps not altogether surprising given that Sxl can modulate the signal strength of a molecule crucial to the development of numerous tissues. The in vivo concentration of Sxl is, most likely, tightly controlled. It has been shown that Sxl negatively regulates translation of its own mRNA. Combined with its positive autoregulatory splicing feedback loop, which ensures that essentially all of the Sxl mRNA is spliced in the productive female mode in females, this dual negative and positive autoregulation implies a homeostasis that keeps the concentration of Sxl in a predetermined fixed range. The potent effect of Sxl on the Hh signal makes the requirement for this dual regulation more readily understood (Horabin, 2005).

Mutations in Sxl that produce sex transformed females generally result in animals that are small and male-like in size. Females transformed by mutations in tra appear as males but maintain the female size, indicating that sexual dimorphic body size is controlled by Sxl (Horabin, 2005).

The enhanced levels of full-length Ci suggest that Sxl promotes disc growth. Indeed, when ectopic Sxl is being expressed in the dorsal half, many of the discs, both male and female, show an overgrowth phenotype with the dorsal half of the wing pouch frequently expanded and distorted. This growth effect is non autonomous, indicating that it is affected by a system that signals beyond the cells expressing Sxl. This is consistent with the idea that Hh signaling is augmented to result in the overgrowth. The experiments described here do not rule out the possibility that Sxl may additionally regulate growth autonomously (Horabin, 2005).

Hh with only the cholesterol modification has the greater impact on Sxl and its stabilization of full-length Ci. However, the Ci that is stabilized does not appear to accomplish Hh patterning. This raises the mechanistic question of how Sxl achieves growth of the entire disc (Horabin, 2005).

Simply reducing the levels of the repressor form of Ci (which is accomplished by increasing the levels of full-length Ci) should increase the expression of the growth factor dpp. This is because dpp is affected by Ci at two levels: absence of the Ci repressor ameliorates repression to give low levels of dpp expression, while activated full-length Ci further elevates dpp transcription. Indeed, while the wing patterning defect caused by the ectopic expression of C84S-Hh narrows the region between wing veins L3 and L4 equally in the two sexes (due to its dominant-negative effect on endogenous Hh), the overall sexual dimorphic size difference is maintained. Consistent with this idea, co-expressing Sxl and Hh with only the cholesterol modification produces an overgrowth phenotype in discs, indicating Sxl can promote disc growth through this form of Hh (Horabin, 2005).

The growth induced by Dpp has been described as 'balanced', involving both mass accumulation as well as cell cycle progression. The net effect is that cell size does not change, nor does the ploidy. This is in contrast to growth induced by hyperactivation of Ras, Myc or Phosphoinositide 3 kinase, which increase growth but do not induce a progression through the G2/M phase of the cell cycle and, as a result, increase cell size (Horabin, 2005).

It is proposed that in the wild-type gradient of Hh with both its lipid modifications, Sxl augments the overall Hh signal to increase both full-length as well as activated full-length Ci. The two Hh targets (Ci and Sxl) respond differentially to the various components of the pathway (Horabin, 2003). Since Sxl is able to alter signal strength, the final outcome of the Hh signal must reflect the balance in activities of the components, modulated by the lipid moieties recognized, the membrane proteins used (Ptc versus Smo) and the proteins present in the Hh cytoplasmic complex. The studies reported here provide a strong rationale for why Sxl resides within the Hh cytoplasmic complex (Horabin, 2005).

Sxl not only elevates expression of dpp and its downstream targets to induce growth, but is able to elevate ptc expression. Enhancing ptc suggests that the Hh signal is 'corrected' for the enlarged patterning field, since short-range patterning has to be controlled by Hh. By enhancing dpp, Sxl indirectly also enhances the long-range patterning system of the disc. Augmenting the Hh signal would thus appear an elegant solution for increasing overall size without changing the basic body plan or pattern. Since Sxl is expressed in all female tissues from very early in development and this expression is maintained for the rest of the life cycle, Sxl is constantly available to upregulate the Hh signal. This augmentation must be kept within check, however, because, as argued above, too high an increase can change the overall slope of the Hh gradient, effectively changing the final patterning of the tissue (Horabin, 2005).

The Hh pathway can also control body size in mammals. ptc1 mutations in mice provide an overgrowth phenotype with large body size, while increasing ptc1 expression decreases body size. Humans with basal cell nevus syndrome, an autosomal-dominant condition caused by the inheritance of a mutant ptc allele, have been reported to have multiple developmental abnormalities and, relevant to this study, larger body size. Whether the mechanism described in this study is global to sexually dimorphic organisms that use Hh for patterning remains to be seen (Horabin, 2005).

A conserved processing mechanism regulates the activity of transcription factors Cubitus interruptus and NF-kappaB

The proteasome degrades some proteins, such as transcription factors Cubitus interruptus (Ci) and NF-kappaB, to generate biologically active protein fragments. This study identifies and characterizes the signals in the substrate proteins that cause this processing. The minimum signal consists of a simple sequence preceding a tightly folded domain in the direction of proteasome movement. The strength of the processing signal depends primarily on the complexity of the simple sequence rather than on amino acid identity, the resistance of the folded domain to unraveling by the proteasome and the spacing between the simple sequence and folded domain. Two unrelated transcription factors, Ci and NF-kappaB, use this mechanism to undergo partial degradation by the proteasome in vivo. These findings suggest that the mechanism is conserved evolutionarily and that processing signals may be widespread in regulatory proteins (Tian, 2005).

Proteasomal proteolysis controls the cellular concentrations of hundreds of regulatory proteins. Normally, the proteasome degrades its substrates completely into small peptides by sequentially running along their polypeptide chain and hydrolyzing the peptide bonds approximately every 8 residues. The proteasome can also function as a processing enzyme that produces functional protein fragments from larger precursors by partial degradation. Processing occurs when the proteasome encounters a stop signal during its sequential hydrolysis of a substrate protein. The stop signal consists of two components: a sequence of low compositional complexity followed by a tightly folded domain in the direction of proteasome movement. Glycine-alanine repeat regions in the Epstein-Barr virus nuclear antigen-1 are known to protect the protein from proteasomal degradation. This study finds that many different simple sequences can cause partial degradation, but only in combination with a tightly folded domain and at the appropriate spacing (Tian, 2005).

Partial degradation can modulate the function of regulatory proteins, as shown for p105 and Ci, and provides a simple mechanism for directly switching a signaling pathway from an active state to a repressed state and vice versa. Processing can also produce more subtle changes in activity. The amount of fragment formed during the degradation of a protein depends on the strength of the processing signal. More fragment is formed the less complex the simple sequenceand the more stable the folded domain. For example, the low-complexity region of Ci is not as simple as the glycine-rich region in p105. Therefore, less Ci repressor is formed in the absence of Hh signaling, and a reporter target gene is less tightly repressed than possible if the low-complexity regions of Ci were replaced with the p105 glycine-rich region. The ratio of Ci activator to Ci repressor in turn determines the activity of the Ci target genes, and, therefore, processing efficiency affects the shape of the Ci activity gradient at the distal edge of Hh signaling. The second component of the processing signal is the susceptibility of folded domains to unraveling by the proteasome, which depends on the stability of the local structure first encountered by the proteasome. A very stable domain, such as methotrexate-stabilized dihydrofolate reductase, can lead to fragment formation without a neighboring simple sequence. It may be possible to modulate the strength of a processing signal in the cell by modifying the simple sequence, for example by phosphorylation, or by adjusting the stability of the folded domain, for example by ligand binding (Tian, 2005).

Processing signal function depends on the direction of degradation, because the proteasome has to encounter the simple sequence before the folded domain and because the susceptibilities of folded domains to unraveling from the N or C terminus can differ. This polarity provides a simple mechanism for the degradation of the protein fragments when they are no longer needed. In the eye, the protein-ubiquitin ligase Cul3 targets Ci for complete degradation using unknown ubiquitination sites. It is predicted that Cul3 ubiquitinates in the N-terminal region of Ci, which would lead to its complete degradation by the proteasome unimpaired by the processing signal (Tian, 2005).

Other examples of protein processing by the proteasome probably exist. The NF-kappaB subunit p52 is synthesized as the larger precursor p100, which is homologous to p105 and presumably processed by the same mechanism. Vertebrates have three Ci homologs, Gli1, Gli2 and Gli3, but only the latter two are processed to a smaller form, probably in a proteasome-dependent manner. Consistent with this observation, only Gli2 and Gli3 seem to have a processing signal and Gli1 may not be ubiquitinated. Processing could also exist in proteins unrelated to Ci or NF-kappaB and does not have to be limited to transcription factors. Regions of low compositional complexity are common and found in half of all predicted eukaryotic proteins, but to form a processing signal, a simple sequence must be positioned adjacent to a tightly folded domain at the appropriate spacing. Standard sequence alignments cannot detect the processing signals, because folded domains can be formed by unrelated sequences and the function of the simple sequence does not depend on the identity of the repeated amino acids (Tian, 2005).

The ubiquitin-proteasome system is also involved in the activation of the two membrane-bound yeast transcription factors Spt23 and Mga2 by the release of N-terminal fragments of these proteins from the membrane. Processing requires folded domains that are homologous to the Rel-homology domain of p105, but the proteins do not contain simple sequences at the expected places. Notably, p105 processing in yeast also does not depend on the presence of a simple sequence. However, because Spt23 and Mga2 are membrane anchored and degradation seems to proceed from an internal loop, it is also possible that the processing is more complicated than the simple mechanism described in this study (Tian, 2005).

The biochemical mechanism by which the processing signal causes partial degradation is not known. Prokaryotic ATP-dependent proteases can release their substrate when they reach protein domains that are hard to unfold. In all the substrates described in this study, the ubiquitination sites had been degraded by the time the proteasome reached the folded domain in their substrate. Thus, the protease was associated with its substrate only through the part of the substrate that was about to be degraded, which contained the simple sequence. The spacing requirement between simple sequence and folded domain for processing differs with the direction of proteasome movement, and this disparity may indicate that the proteasome interacts with its substrates differently depending on the direction of degradation. The simple sequences could then lead to processing if they reduced the affinity of the substrate for the proteasome. In apparent agreement with this proposal, simple sequences cannot serve as efficient degradation initiation sites. Once the substrate is released from the proteasome, it has escaped proteasomal degradation because the ubiquitination site has already been removed. Thus, according to this model, protein processing occurs when a tightly folded domain located at the entrance to the degradation channel stalls the progression of the proteasome along the polypeptide chain and a simple sequence in the channel accelerates the release of the folded domain and the remaining protein from the proteasome (Tian, 2005).

In summary, a combination of a simple sequence followed by a folded domain in the direction of proteasome movement inhibits proteasome progression; this results in the accumulation of a partially degraded fragment. Proteasomal processing by partial degradation has an essential function in at least two unrelated cellular signaling pathways. The processing mechanism is conserved between flies and humans, and more examples of this process probably exist (Tian, 2005).

Processing of the Drosophila hedgehog signaling effector Ci-155 to the repressor Ci-75 is mediated by direct binding to the SCF component Slimb

Signaling by extracellular Hedgehog (Hh) molecules is crucial for the correct allocation of cell fates and patterns of cell proliferation in humans and other organisms. Responses to Hh are universally mediated by regulating the activity and the proteolysis of the Gli family of transcriptional activators such that they induce target genes only in the presence of Hh. In the absence of Hh, the sole Drosophila Gli homolog, Cubitus interruptus (Ci), undergoes partial proteolysis to Ci-75, which represses key Hh target genes. This processing requires phosphorylation of full-length Ci (Ci-155) by protein kinase A (PKA), casein kinase 1 (CK1), and glycogen synthase kinase 3 (GSK3), as well as the activity of Slimb. Slimb is homologous to vertebrate ß-TRCP1, which binds as part of an SCF (Skp1/Cullin1/F-box) complex to a defined phosphopeptide motif to target proteins for ubiquitination and subsequent proteolysis. Phosphorylation of Ci at the specific PKA, GSK-3, and CK1 sites required in vivo for partial proteolysis stimulates binding to Slimb in vitro. Furthermore, a consensus Slimb/ß-TRCP1 binding site from another protein can substitute for phosphorylated residues of Ci-155 to direct conversion to Ci-75 in vivo. From this, it is concluded that Slimb binds directly to phosphorylated Ci-155 to initiate processing to Ci-75. The phosphorylated motifs in Ci that are recognized by Slimb have been explored and some evidence is provided that silencing of Ci-155 by phosphorylation may involve more than binding to Slimb (Smelkinson, 2006).

The mechanism and consequences of Hh signaling have been studied extensively in the developing Drosophila wing imaginal disc, where Hh, secreted from posterior compartment cells, induces a strip of nearby, responsive anterior cells (AP border cells) to express a small set of target genes, including decapentaplegic (dpp), that subsequently pattern the developing wing. In anterior cells far from Hh, Ci-155 is processed slowly to Ci-75, which crucially represses potential Hh target genes, including hh itself and dpp, to ensure that they are not ectopically expressed. Even low-level Hh signaling at the AP border blocks Ci-75 production, thereby also increasing the concentration of Ci-155. Hh further activates Ci-155 in a dose-dependent manner by facilitating its nuclear accumulation and potentially also by modifying its binding partners in the nucleus. Because formation of Ci-75 requires both Ci-155 phosphorylation and the activity of Slimb, it was proposed that Slimb might promote partial proteolysis of Ci-155 by directly binding to phosphorylated Ci-155 and catalyzing its ubiquitination. However, despite some support for this hypothesis, Ci-155 contains no obvious consensus binding site for Slimb/β-TRCP1: there are only two well-studied examples where proteasomal degradation of a ubiquitinated protein is incomplete (NF-KappaB precursors, p100 and p105), and ubiquitinated Ci-155 has not been detected when Ci-155 is stabilized by inhibiting the proteasome (Smelkinson, 2006).

To determine whether Slimb can bind to Ci in a manner dependent on phosphorylation, a purified GST fusion protein was used that includes the key phosphorylation sites of Ci. This GST-Ci protein undergoes a significant mobility shift in SDS polyacrylamide gels when phosphorylated by PKA and CK1 together and an even greater shift if GSK3 is also included. GST-Ci binds more avidly than GST alone to ³⁵S-labeled full-length Slimb produced by in vitro translation in a reticulocyte lysate and to HA-tagged full-length Slimb from crude extracts of transiently transfected Drosophila Kc cells. This binding is reproducibly increased by using PKA together with CK1, and to a much greater extent by using all three protein kinases to phosphorylate GST-Ci prior to the binding assay. The synergistic contribution of GSK3 was clearest in the HA-Slimb binding assay, and so this assay was used to investigate further the characteristics of Ci binding to Slimb (Smelkinson, 2006).

Three PKA sites ('P1-3'), the neighboring three PKA-primed CK1 sites ('C1-3') and the two adjacent PKA-primed GSK3 sites ('G2,3') are required for Ci-155 to be converted to Ci-75 in Drosophila embryos and wing discs. When the serine residues at these PKA, CK1, or GSK3 sites were replaced with alanines, significant stimulation of binding of GST-Ci to HA-Slimb was no longer seen by any combination of PKA, CK1, and GSK3. Thus, strong binding of HA-Slimb to a Ci fragment in vitro requires the same protein kinases and the same phosphorylation sites that are required in vivo to convert Ci-155 to Ci-75 (Smelkinson, 2006).

Whether a defined, minimal Slimb binding site from another protein could direct processing of Ci-155 to Ci-75 was tested. β-catenin is a prototypical substrate for the β-TRCP1 SCF complex, in which a dually phosphorylated motif (DpSGIHpS, where pS stands for phosphoserine) is the critical recognition element for binding. This motif is conserved in Drosophila β-catenin (Armadillo), and Armadillo proteolysis depends on both this sequence and Slimb activity. The motif is also expected to serve as a direct binding site for Slimb, the Drosophila homolog of β-TRCP1. Tests revealed that this consensus Slimb/β-TRCP1 binding site, engineered into Ci, is functional and directs Slimb binding in vitro with an apparent avidity similar to that seen for fully phosphorylated wild-type Ci (Smelkinson, 2006).

Binding assays were used to search for the direct Slimb recognition elements in Ci because no established Slimb/β-TRCP consensus binding sites are apparent in the sequence of Ci or Gli proteins. Each of the three PKA sites in Ci is required for detectable processing to Ci-75 in Kc tissue-culture cells; therefore, whether each site contributes to Slimb binding in vitro was tested. Replacement of all three PKA-primed CK1 sites (and the two predicted CK1-primed CK1 sites) with acidic residues abolished any stimulation of HA-Slimb binding by phosphorylation of GST-Ci (Smelkinson, 2006).

It cannot be readily determined whether the PKA sites and PKA-primed CK1 sites are directly recognized by Slimb. However, the evident contribution of each PKA site to Slimb binding implies that each must nucleate at least one direct Slimb binding site. Surrounding the three PKA sites there are two types of motifs that are related to previously recognized or postulated Slimb/β-TRCP binding motifs (DSGXXS, DSGXXXS, TSGXXS, EEGXXS, DDGXXD, and DSGXXL. (1) The six-amino-acid motif (D/pS)(pS/pT)(Q/Y)XX(pS/pT) might be created in three places if phosphorylation occurs, as is suspected, at some nonconsensus sites. These motifs most closely resemble the β-TRCP1 binding site postulated for p100 (DpSAYGpS), but the presence of glutamine or tyrosine at the third position in place of glycine in the ideal consensus would be expected to reduce binding by more than an alanine substitution. (2) Many five-amino-acid motifs are created in which two acidic residues (DpS, pSpS, or pSpT) are separated from a phosphorylated residue (pS or pT) by only two amino acid residues. Four of these eight motifs include a glutamine at the third position. However, this residue does not appear to be instrumental in Slimb binding because substitution with alanine has little effect. On the basis of the crystal structure of β-TRCP1, it is hard to predict the affinity of the designated five-amino-acid motifs for Slimb, but it is likely to be lower than for any of the motifs cited above. Regardless of which precise motifs contribute most significantly to Slimb binding, it is clear from the mutational analysis that Ci must be highly phosphorylated over a region spanning almost sixty amino acid residues to generate several suboptimal binding sites that must collaborate to provide physiologically significant affinity for Slimb. This follows a precedent established for degradation of the yeast cell-cycle regulator Sic1 by binding of the SCF_CDC4 complex to multiple low-affinity phosphodegrons. The requirement for extensive Ci phosphorylation could account for the essential role of the scaffolding protein Cos2 in facilitating phosphorylation and for the relatively slow conversion of Ci-155 to Ci-75 seen in vivo (Smelkinson, 2006).

In summary, processing of Ci-155 to Ci-75 is initiated by direct binding of Slimb to Ci-155 molecules that have been extensively phosphorylated by PKA, GSK3, and CK1. This presumably leads to ubiquitination of Ci-155 and its partial proteolysis by the proteasome, generating a transcriptional repressor that plays a key developmental role in cells that are not exposed to Hh. Whether phosphorylation also prevents Ci-155 from activating transcription through an additional mechanism remains to be explored, as does the mechanism by which proteolysis of Ci-155 is limited to preserve its N-terminal domains as Ci-75 (Smelkinson, 2006).

Regulation of the Drosophila transcription factor, Cubitus interruptus, by two conserved domains

Hedgehog signaling is required for the development of many organisms, including Drosophila. In flies, Hh patterns the embryonic epidermis and larval imaginal discs by regulating the transcription factor, Cubitus interruptus (Ci). To date, three levels of regulation have been identified: proteolytic processing into a repressor, nuclear import, and activation. In this report, the function of two Ci domains that are conserved in the vertebrate homologues, GLI1, GLI2, and GLI3 has been tested. One domain includes the first two of five C2-H2 zinc-fingers. While conserved in all members of the GLI/Ci family, the first two fingers do not appear to make significant contacts with the DNA target sequence. Ci protein lacking this region is still able to interact with the cytoplasmic complex and activate transcription in embryos and wing imaginal discs, but it is no longer processed into the repressor form. The second domain, termed NR for 'N-terminal Regulatory', binds Suppressor of Fused. Deletion of this region has little effect on embryonic patterning, but compromises cytoplasmic retention of Ci. Analysis of the amino acid sequence of this domain identifies 11 perfectly conserved serines and one tyrosine. It is proposed that this region may be modified, possibly by phosphorylation, to regulate Ci nuclear import (Croker, 2006).

Despite 500 million years of divergence between higher vertebrates and Drosophila, many proteins have homologues with extensive sequence conservation. This is well exemplified by the GLI family of transcription factors and their fly counterpart, Ci. Over their entirety, Ci and the human GLIs share over 22% amino acid identity. Ci and human GLI3 are even more similar, at greater than 27% identity. Though the proteins' sequences have diverged, the two domains studied here remain strikingly similar. This information was exploited to further the understanding of the complex regulation of Ci in Drosophila (Croker, 2006).

Amino acid comparisons of the mouse and human GLI1, GLI2, and GLI3 proteins with Drosophila Ci show an expected high degree of conservation in the DNA-binding domain, which consists of five tandem C₂–H₂ zinc-fingers. This domain in all of the GLI homologs appears to recognize the same DNA target sequence. Furthermore, the Gli proteins have been shown to recapitulate the multifaceted functions of Ci in flies. This study focused attention on the first two of five zinc-fingers, which have not been implicated in significant DNA contacts. Sequence analysis shows 58% identity over the 60 amino acids that constitute the region. While this is substantially less than the 91% identity of the three DNA binding zinc-fingers, it is nonetheless impressive (Croker, 2006).

A second domain N-terminal to the DNA-binding domain also shares considerable sequence identity with the vertebrate homologues. In this case, the region, which has been termed “NR”, appears to be present in Ci, GLI1, GLI2, and GLI3. There exists no obvious NR motif in the nematode homologue Tra-1. This region retains greater than 57% identity over 49 amino acids examined in the mouse, human, and fly proteins (72% identity between Ci and its closest homolog GLI3). Over one quarter of this domain consists of perfectly conserved serines, threonines, and tyrosines. It is speculated that this module contains phosphorylation sites that contribute to the proper regulation of these transcription factors, perhaps by modulating interaction with Su(fu) (Croker, 2006).

Analysis of en and 4bsLacZ (a wing-specific decapentaplegic enhancer) gene expression in wing disc clones expressing UAS-CiΔZ2myc induced by the Actin5Cp-“flip-out”-GAL4 driver shows that a Ci molecule lacking the first two zinc-fingers retains the ability to bind DNA and activate transcription. This is consistent with the crystal structure of the DNA–Gli1 complex. Gene expression in this context is Hh-independent, but it is a possible consequence of overexpression, since a wild type UAS-Ci-cDNA can activate these target genes if it is expressed at very high levels (Croker, 2006).

To better gauge the regulation of CiΔZ2myc, it was also analyzed in ci⁹⁴ null mutant embryos using the ci-GAL4 driver. Embryonic cuticles expressing UAS-CiΔZ2myc in the anterior compartment of every segment show a gain-of-function phenotype reminiscent of cos2 and ptc mutants. This suggests that the first two zinc-fingers provide a negative regulatory or repressive function. Supporting this hypothesis is the observation that CiΔZ2myc fails to repress hh-lacZ expression in posterior smo clones. The lost repressive activity could be accounted for in a number of different ways. It is possible that the first two zinc fingers recruit a co-repressor protein or that they are required for proper modification of Ci. Alternately, it is possible that this construct is not processed into the repressor form in the absence of Hh signaling. The latter hypothesis is favored since Western blot analysis of the CiΔZ2myc protein fails to show a truncated repressor form. Further, there is a broadening of the domain of ectopic CiΔZ2myc in embryos compared to overexpression of both wild type Ci and CiΔNRmyc, which is consistent with a processing defect of this molecule. These results extend previous work showing that the zinc finger domain as a whole is required for processing into the Ci-75 repressor. In addition to the apparent defect in proteolytic processing, the nuclear import of CiΔZ2myc appears to be Hh-independent. These phenotypes could be caused by a failure of the CiΔZ2myc protein to efficiently complex with Fu, Su(fu), and Cos2, but the known binding sites for these proteins are all retained in CiΔZ2myc. Rather, this is likely an effect of overexpression as co-expression with Su(fu) leads to cytoplasmic retention of CiΔZ2myc (Croker, 2006).

While the CiΔZ2myc construct shows some Hh-independent function, the protein does not appear to be fully active in the absence of Hh signaling. In ptc mutant embryos, where the pathway is fully activated, the stripe of en expression is duplicated anterior to each expanded Wg stripe. This does not occur with CiΔZ2myc (Croker, 2006).

Studies on the function of a Ci construct in which sequences amino-terminal to the zinc-finger DNA binding domain were deleted have been reported previously. Expression of the CiZnC protein in the posterior compartment of embryos leads to ectopic ptc activation. These findings extend the former observations by using ci-GAL4 to drive continuous expression of transgenes in the anterior compartment. In the case of embryonic cuticle patterning, it is shown that the CiZnC protein induced by ci-GAL4 actually has a gain-of-function phenotype when expressed in embryos lacking endogenous Ci. The gain-of-function phenotype suggests the existence of domains in the amino-terminus of the Ci protein that regulate the function of the full-length molecule or contribute to the repressor function of Ci-75. Two obvious candidates are amino acid regions 1-346 and 346-430 which bind Su(fu) and Cos2, respectively. The NR region (AA 208-260) is likely to be important for Ci binding to Su(fu); amino acids 116-125 of the human Gli1 protein are critical for Su(fu) binding, especially the SYGHLS amino acid motif at the end of the domain. This interaction domain sequence corresponds to amino acids 245-254 of the Ci NR domain, which retains perfect conservation of the SYGHXS motif (Croker, 2006).

When the amino-terminal deletion is restricted specifically to the NR region (208-260), embryonic development is essentially normal. The CiΔNRmyc protein retains the ability to repress hh-lacZ in posterior smo clones and thus is presumably processed into the repressor form. In contrast, when CiΔNRmyc is expressed in wing imaginal disc clones using 'flip-out' Gal4, it is not properly sequestered in the cytoplasm. This phenotype may be in part a consequence of protein overexpression, but wild type Ci expressed at similar levels is properly sequestered in the cytoplasm in the absence of Hh signaling. These results might be expected if the role of the NR domain is to regulate the interaction of Ci with Su(fu). Su(fu) mutants are viable and only have subtle patterning defects, and thus, a mutation in Ci that disrupted Su(fu) interaction should be relatively normal. However, in conditions of Ci overexpression, it has been shown that Su(fu) is required for Ci cytoplasmic retention. Indeed, overexpression of Su(fu) can restore proper cytoplasmic localization of a highly expressed wild type Ci molecule, but cannot rescue the CiΔNRmyc nuclear localization phenotype suggesting that the NR domain does modulate Ci/Su(fu) interaction. Yeast two-hybrid results show that this conserved interaction between Ci and Su(fu) is direct. It is postulated that modification of the NR domain at conserved putative phosphorylation sites modulates the molecule's interaction with Su(fu). Su(fu) appears able to act as a Ci sponge and to sequester excess Ci in the cytoplasm. This may explain in part why Su(fu) mutants have such a subtle phenotype. Under normal circumstances, Cos2 and the other components of the cytoplasmic complex are able to retain Ci in the cytoplasm. Su(fu) provides a redundant mechanism for sequestering any Ci that escapes (Croker, 2006).

Given that deletion of the NR region fails to generate a strong gain-of-function phenotype, an additional negative regulatory element must exist in the amino-terminus outside of the NR region. Two candidates are the Cos2 binding region (346–430) and a pair alanine-rich clusters in the Ci N-terminus that may function as a repressor domain (Croker, 2006).

This analysis of the NR region and the first two zinc-fingers demonstrates that both are required for negative regulation of the Ci protein. The NR domain functions to regulate interaction with the Su(fu) protein which in turn modulates the subcellular localization of the transcription factor. In the case of the first two zinc-fingers, the interacting proteins are unknown, but are likely to play important roles in Ci regulation (Croker, 2006).

Regulation of Ci and Su(fu) nuclear import in Drosophila

The Hedgehog (Hh) signal transduction pathway plays a central role in the development of invertebrates and vertebrates. While much is known about the pathway, the role of Suppressor of fused [Su(fu)], a component of the pathway's signaling complex has remained enigmatic. Previous studies have linked Su(fu) to the cytoplasmic sequestration of the zinc finger transcription factor, Cubitus interruptus (Ci), while other studies suggest a role in modulating target gene expression. In examining the cell biology of the pathway, it was found that like its vertebrate homologue, Drosophila Su(fu) enters the nucleus. Furthermore, the nuclear import of Su(fu) occurs in concert with that of Ci in response to Hh signaling. This study examines the mechanism by which Su(fu) regulates Ci import by investigating the importance of the Ci nuclear localization signal (NLS) and the effect of adding an additional NLS. Finally, it is demonstrated that Ci can bring Su(fu) with it to a multimerized Ci DNA binding site. These results provide a basis for understanding the dual roles played by Su(fu) in the regulation of Ci (Sisson, 2006).

This study establishes the Hh dependent translocation of the Su(fu)-Ci complex into the nucleus, illustrating further conservation between the mammalian and Drosophila Hedgehog signaling pathways. This result is somewhat of a paradox since Su(fu) has been shown to assist in the sequestration of Ci in the cytoplasm. If Su(fu) contributes to the cytoplasmic sequestration of Ci, what is the mechanism that allows its release in response to Hh signaling? The likely event is phosphorylation by the Fu kinase. In the absence of Fu kinase function, Ci is not released in response to Hh signaling. However, in double mutants lacking both Su(fu) and Fu kinase activity, Ci is now able to enter the nucleus, suggesting that Fu kinase activity is required to regulate the cytoplasmic retention of the Su(fu)-Ci complex. It is not known whether Su(fu) or Ci are direct targets of the Fu kinase, and this need not be the case, since modification of other components of the pathway such as Cos2 could allow Su(fu)-Ci complex release (Sisson, 2006).

While previous studies demonstrated the functionality of the Ci NLS at AA R596-K600 and K611-K614, the data presented here indicate that Ci nuclear import in salivary glands and in wing discs does not absolutely require the presence of this NLS. Consistent with some decrease in NLS function, Ci-mutNLS gives substantial, but not complete, rescue of a ci null mutation. This suggests either the existence of an additional NLS within Ci or the presence of an additional protein that brings Ci into the nucleus (Sisson, 2006).

The results from salivary glands may favor the hypothesis that an additional NLS is present in Ci. Ci-mutNLS nuclear import is impeded but not prevented, implying that if an additional protein were necessary to bring Ci into the nucleus, it too would have to be present in salivary glands and would not be Ci specific. Another consideration is that perhaps the Ci-mutNLS mutation does not entirely destroy NLS function. This mutation only disrupts the second basic cluster within a bipartite NLS, because altering the first cluster would disrupt the last zinc finger and DNA binding (Sisson, 2006).

Addition of an exogenous SV40NLS to Ci leads to more rapid nuclear import in salivary glands and a variable gain of function phenotype in embryos. The increased rate of nuclear import appears to compromise the ability of Su(fu) to sequester Ci-SV40NLS in the cytoplasm of anterior wing imaginal disc clones that are away from Hh signaling. One could interpret this result to suggest that Su(fu) masks the endogenous Ci NLS but not the added SV40NLS. This seems unlikely since wild-type Su(fu)-Ci complexes readily enter the nuclei of salivary glands. An alternative explanation is that away from Hh signaling the Su(fu)-Ci complex has some affinity for a cytoplasmic tether and some low probability of being imported into the nucleus; increasing the rate of nuclear import shifts this equilibrium resulting in nuclear accumulation of Su(fu)-Ci (Sisson, 2006).

Potentially consistent with the role of Su(fu) in sequestering Ci in the cytoplasm is the observation that N-terminally myc-tagged Su(fu) appears to be tightly tethered in the cytoplasm resulting in the cytoplasmic retention of both it and Ci. It is possible that the addition of the myc tag causes a spurious interaction between Su(fu) and an unidentified cytoplasmic component, or it may be the case that the interaction is normal, but the addition of the myc tag prevents the release of Su(fu) from this component. Since these experiments were carried out in salivary glands, where there is little if any Fu, Smo or Cos2, the component tethering myc-Su(fu) is distinct from the known proteins in the Hh signal transduction cascade (Sisson, 2006).

Ci-SV40NLS is not sequestered in the cytoplasm of salivary glands by myc-Su(fu). Instead much of it escapes into the nucleus, but it does not bring myc-Su(fu) with it. Again, this result is likely a consequence of the increased rate of Ci-SV40NLS nuclear import. Myc-Su(fu) remains tightly tethered in the cytoplasm, and the distribution of Ci-SV40NLS will be determined by the relationship between the rate of nuclear import and the binding affinity to myc-Su(fu) (Sisson, 2006).

Mammalian studies have found that the addition of mouse Su(fu) can increase the binding affinity of the Glis to target DNA sequences. The salivary gland model system demonstrates that Su(fu), along with Ci, clearly bind an introduced target within the polytene chromosomes. The presence of Su(fu) at Hh target gene enhancers provides the opportunity for another level of Ci regulation. This regulatory role is likely to be restricted to the full-length form of Ci. The Ci repressor form is missing the C-terminal Su(fu) binding site and Ci-N[HA]Zn, which closely resembles the Ci repressor, does not bring Su(fu) with it to the DNA (Sisson, 2006).

Given the high degree of conservation between the mammalian and Drosophila Hh signaling pathways, one might expect Su(fu) to play homologous roles in the negative regulation of the two pathways. Su(fu) has been shown to act in the cytoplasm as a negative regulator of the pathway by contributing to the sequestration of Ci. In the absence of Hh, Fu and Ci are tethered to microtubules via their interaction with Cos2. In the presence of Hh, Smo is phosphorylated, the complex is released from the microtubules, and a tetrameric cytoplasmic complex is formed with the addition of Su(fu). Su(fu) may contribute to Ci sequestration in two ways, as part of the tetrameric cytoplasmic complex and as a heterodimer with Ci where it could act as a sink to sequester any excess Ci that is not bound to the Cos2-Fu complex (Sisson, 2006).

The presence of Su(fu) in the nucleus suggests a dual role for Su(fu) in the regulation of Ci. A direct nuclear role for Su(fu) negative regulation has been inferred by vertebrate studies documenting an interaction of Su(fu) with SAP18. While mammalian Su(fu) has been show to interact with SAP18 through GST pull-downs and yeast two-hybrid analysis, initial yeast two-hybrid studies did not reveal an interaction between Drosophila SAP18 and Drosophila Su(fu). Therefore, further studies are needed to delineate the function of Drosophila Su(fu) in the nucleus and to determine if SAP18 is indeed involved in the negative regulation of Ci (Sisson, 2006).

The question still remains of how differential responses to Hh signaling are generated. In the presence of Hh, Su(fu) is phosphorylated and it has been suggested that this modification at the A/P boundary may reduce Su(fu) repressive activity, thus allowing Ci to activate target genes requiring the highest levels of Hh. The modification of Su(fu) or Ci could change the nature of cofactors recruited to target enhancers and account for differential gene regulation (Sisson, 2006).

This model is consistent with observations on the phenotypes of cos2 and cos2; Su(fu) double mutant clones. In cos2 mutant clones, targets that require modest levels of Hh are activated while those that require high-level Hh are not. This suggests that the Cos2 protein is required for some modification of the Ci-Su(fu) heterodimer that is essential for 'activation'. When the Su(fu) gene is also eliminated, now target genes requiring high level Hh are activated. Thus, Su(fu) must contribute to the attenuation of Ci activity in response to modest levels of Hh. However, Su(fu) cannot be the entire story as animals mutant for Su(fu) are essentially normal. There must be a second factor that is partially redundant with Su(fu) in attenuating Ci activity. Since cos2; Su(fu) mutant clones have 'activated' Ci, it would seem that Cos2 is required for the function of this second factor (Sisson, 2006).

Roadkill attenuates Hedgehog responses through degradation of Cubitus interruptus

The final step in Hedgehog (Hh) signal transduction is post-translational regulation of the transcription factor, Cubitus interruptus (Ci). Ci resides in the cytoplasm in a latent form, where Hh regulates its processing into a transcriptional repressor or its nuclear access as a transcriptional activator. Levels of latent Ci are controlled by degradation, with different pathways activated in response to different levels of Hh. The roadkill (rdx) gene is expressed in response to Hh. The Rdx protein belongs to a conserved family of proteins that serve as substrate adaptors for Cullin3-mediated ubiquitylation. Overexpression of rdx reduces Ci levels and decreases both transcriptional activation and repression mediated by Ci. Loss of rdx allows excessive accumulation of Ci. rdx manipulation in the eye revealed a novel role for Hh in the organization and survival of pigment and cone cells. These studies identify rdx as a limiting factor in a feedback loop that attenuates Hh responses through reducing levels of Ci. The existence of human orthologs for Rdx raises the possibility that this novel feedback loop also modulates Hh responses in humans (Kent, 2006; full text of article).

The rdx locus was identified by an enhancer trap with embryonic expression in a pattern suggesting Hh-regulation. When genomic DNA flanking the insertion was used to screen a Drosophila embryonic cDNA library, cDNAs were obtained that initiated near the enhancer trap insertion and spliced into a cluster of seven downstream exons. These cDNAs represented the predicted gene CG10235 spliced into the predicted gene CG9924. ESTs recovered by the BDGP identified four additional isoforms (CG9924 A-D), which differ in their 5' ends but which share the cluster of seven downstream exons with rdxE, and are designated rdxA-rdxD (Kent, 2006).

The A, C/D and E forms are predicted to encode proteins with unique and novel amino termini fused to a common C terminus. The B form lacks unique coding sequence and is predicted to initiate translation within exon 7. The 398 C-terminal residues encoded by exons 7-13 contains two conserved domains: a MATH (Meprin and TRAF homology) domain and a BTB (Broad/Tramtrack/Bric-a-brac) domain. These two protein interaction domains are found together in an evolutionarily conserved protein family where the BTB domain binds to Cul3, while the MATH domain recruits specific substrates to the Cul3-based E3 ubiquitin ligase complex for ubiquitylation and subsequent degradation (Kent, 2006).

rdxA, rdxE and the initial enhancer trap produced expression patterns that were indistinguishable from those of a probe common to all rdx forms. Maternally deposited rdx transcripts were detected in early embryos, but disappeared during mid-cleavage stages. The first zygotic transcripts appeared in pole cells. During cellularization of the blastoderm, rdx transcripts appeared in two broad stripes in the head, in seven narrower stripes along the segment primordium, and in a ring surrounding the pole cells. Seven additional stripes appeared during germ band extension, so that by stage 8, rdx was expressed in 14 evenly spaced ectodermal stripes characteristic of segment polarity genes. At this time, strong expression was seen in the anterior and posterior midgut primordia. During stage 9/10, expression appeared in a subset of neuroblasts. During stage 10 each segmental stripe split so that by stage 11, ectodermal expression consisted of two thin stripes corresponding to the anterior and posterior margins of the former stripe. At this time, strong expression was seen in the mesoderm. As germ band retraction began, expression faded from most of the ectoderm, but was retained in the salivary glands and in abdominal segment 9. After stage 14, rdx expression was detected only in the clypeolabrum, anal plate and salivary glands (Kent, 2006).

Thus, rdx encodes a protein belonging to a phylogenetically conserved protein family of substrate-specific adaptors for Cullin3-based ubiquitin E3 ligases. rdx loss-of-function and gain-of-function studies suggest that rdx has at least two substrates: a regulator of early embryonic mitoses and the Hh regulated transcription factor Ci155. The data support a model where Rdx regulates the Hh-dependent degradation of Ci by acting as the adaptor that presents Ci to the Cul3-based E3 ubiquitin ligase. Because rdx is expressed in response to Hh, rdx is involved in a novel regulatory loop that attenuates Hh responses through reducing levels of Ci. In the wing, this feedback regulation of Ci by rdx plays a minor role, but in the eye it is essential for proper packing of ommatidia into a hexagonal array (Kent, 2006).

Hh is key regulator in human health. The haploinsufficiency of Ptc in humans and its activity as a morphogen in the spinal column argue that the level of Hh response is often crucial. Although there are differences in the Hh pathway between flies and vertebrates, many regulatory mechanisms are conserved. In particular, Gli2 and Gli3 are regulated much like Ci, becoming repressors or activators, depending on levels of Hh. The Rdx ortholog SPOP lies in 17q21.33, a chromosomal region that has been linked with ovarian cancer and cervical immature teratoma. Future studies will determine whether the Rdx orthologs SPOP or LOC339745 modulate Gli levels and Hh-mediated responses, and even contribute to cancer (Kent, 2006).

Continued: Cubitus interuptus Protein Interactions part 2/2