groucho

Targets of Activity

Several different mechanisms have evolved for preventing gene transcription. One class of repressors function as transcriptional 'poisons', forming protein complexes with activators to retain them in the cytoplasm, rendering them unable to bind DNA. Other repressors act by competing for common or overlapping DNA target sites, thereby excluding activators from access to target promoters (Tailless, for example). A third class of negative transcriptional regulators also binds DNA but appears to act in a more instructive manner. Some such repressors (Snail, Kruppel, and Knirps for example) act at short range to mask adjacent, proximally bound activators. Others act over long-range, probably by directly interfering with the general transcriptional promoter sites. This latter class of DNA-bound repressors is likely to be assisted by general cofactors, by analogy to transcriptional activators that recruit co-activators to target gene promoters. Hairy and Enhancer of split proteins, and their corepressor Groucho, are included in this latter class of proteins (Paroush, 1997).

Tailless (Tll) acts as a repressor of Kruppel and knirps in the central domain of the recently fertilized embryo. Groucho acts throughout the embryo to repress the repressor of Kruppel and knirps, allowing the expression of these gap genes in the central domain of the embryo. Patterning of the non-segmental termini of the Drosophila embryo depends on signaling via the Torso receptor tyrosine kinase (RTK). Activation of Torso at the poles of the embryo triggers expression of the terminal zygotic gap genes tailless and huckebein (hkb). The Groucho (Gro) corepressor acts in this process to confine terminal gap gene expression to the embryonic termini. Embryos lacking maternal gro activity display ectopic tll and hkb transcription; in turn, tll then leads to lack of abdominal expression of the Kruppel and knirps gap genes. torso signaling permits terminal gap gene expression by antagonizing Gro-mediated repression. Groucho-mediated repression of tailless is relieved by the torso pathway suggesting that Groucho is the nuclear target for MAP kinase signaling. It is suggested that Groucho functions as a corepressor along with an unknown protein unrelated to Hairy, since Groucho mediated repression takes place in the absence of known Hairy-related bHLH proteins. Thus, the corepressor Gro is employed in diverse developmental contexts and, probably, by a variety of DNA-binding repressors (Paroush, 1997).

Maternal Groucho is essential for segmentation. groucho appears to be required for hairy and even-skipped dependent repression of fushi tarazu (Paroush, 1994). Sex-lethal is also a target of groucho. groucho-deadpan complexes act directly to repress early sex-lethal transcription (Paroush, 1994).

In normal growth of imaginal discs, posterior specific genes like hedgehog and engrailed are required to be restricted to the posterior compartment. Hedgehog has the potential to activate engrailed in the anterior compartment, but both hedgehog and engrailed are specifically repressed in anterior cells by the activity of neurogenic gene groucho. In groucho mutant discs, hh and en are expressed in the anterior compartment leading to ectopic activation of decapentaplegic and patched, and to a localized increase in cell growth associated with pattern duplications. This leads to a transformation of the anterior into the posterior (de Celis, 1995). Groucho also acts during embryonic development to restrict engrailed and hedgehog to the anterior parasegmental compartment (de Celis, 1995).

Transcriptional control of the Drosophila terminal gap gene huckebein (hkb) depends on Torso (Tor) receptor tyrosine kinase (RTK) signaling and the Rel/NFB homolog Dorsal (Dl). Dl acts as an intrinsic transcriptional activator in the ventral region of the embryo, but under certain conditions, such as when it is associated with the non-DNA-binding co-repressor Groucho (Gro), Dl is converted into a repressor. Gro is recruited to the enhancer element in the vicinity of Dl by sequence-specific transcription factors such as Dead Ringer (Dri). The interplay between Dl, Gro and Dri on the hkb enhancer was examined and it was shown that when acting over a distance, Gro abolishes rather than converts Dl activator function. However, reducing the distance between Dl- and Dri-binding sites switches Dl into a Gro-dependent repressor that overrides activation of transcription. Both of the distance-dependent regulatory options of Gro -- quenching and silencing of transcription -- are inhibited by RTK signaling. These data describe a newly identified mode of function for Gro when acting in concert with Dl. RTK signaling provides a way of modulating Dl function by interfering either with Gro activity or with Dri-dependent recruitment of Gro to the enhancer (Hader, 1999).

The cis-acting element has been identified that mediates expression of the Drosophila gene hkb, which is necessary for terminal pattern formation and to size the mesoderm anlage in the blastoderm embryo. Deletion analysis of this element reveals a 162 base pair (bp) sub-element that integrates the activities of the Tor-dependent RTK signaling cascade and the morphogen Dl. This element, termed hkb ventral element (VE), comprises a 112 bp ventral activator element (VAE) and a 50 bp ventral repressor element (VRE) (Hader, 1999).

The VAE contains a Dl-binding site, identified in vitro, and mediates gene activation along the ventral side of the embryo. VAE-mediated gene expression is absent in embryos lacking Dl activity and extends throughout Toll^10b mutants, in which Dl is present in all nuclei of the embryo. The expression pattern is not altered in embryos lacking snail and twist, the zygotic mediators of Dl. It is also not affected in embryos that lack Tor or express constitutively active Tor^Y9, which causes RTK signaling throughout the embryo. In contrast, the VE fails to activate in the absence of Tor and mediates broad ventral expression in tor^Y9 embryos not seen in the absence of Dl activity. This indicates that VAE mediates transcriptional activation by Dl, that the VRE, which by itself fails to activate transcription, is necessary to prevent Dl-dependent activation in the central region of the embryo, and that the activity of the unknown repressor, mediated by the VRE, is relieved by RTK signaling (Hader, 1999).

The evolutionarily conserved co-repressor Gro acts as a repressor of Dl activity, since both hkb expression and VE-driven gene expression expand along the ventral side of embryos lacking groucho (gro) activity. However, VAE-driven gene expression and the terminal expression domains of hkb are not significantly affected by lack of Gro. Thus, Gro functions as a repressor of VAE-directed, Dl-dependent transcriptional activation in the ventral region of the embryo and must act through the VRE (Hader, 1999).

Previous results have shown that Gro switches the transcriptional activator Dl into a potent silencer of transcription. This requires the formation of a multiprotein repressor complex of which Dl and Gro are obligatory components. Complex formation requires that Gro is recruited next to Dl by sequence-specific transcription factors such as Cut or Dri. Lack of either Gro or Dri activity results in VE-driven gene expression along the ventral axis of the embryo, indicating that both factors are necessary for repression of Dl-dependent activation. A single binding site has been identified for Dri in the VRE. Replacement of 5 bp in this site (VE-^DRI) results in loss of repression in the central region of the embryo, indicating that Dri is necessary for recruitment of Gro to the VE (Hader, 1999).

The VE differs from the cis-acting elements of the genes zerknullt (zen) and decapentaplegic (dpp), both of which mediate long-range Dl-dependent transcriptional silencing by Gro. In these elements, binding sites for Dri and Dl are directly adjacent, whereas in the VE they are some 90 bp apart. This distance suggested the possibility that Gro cannot associate with Dl on the VE, implying that Gro must prevent Dl-dependent activation by a means other than formation of a long-range silencing complex, for example, by short-range quenching. This proposal was tested by monitoring gene expression patterns directed by a cis-acting activator element of the gene knirps (kni-element) to which the VRE, the VAE, the VE or molecularly defined variants of the VE were fused (Hader, 1999).

The kni-element drives gene expression throughout the embryo except in the posterior pole region. It mediates activation in response to the transcriptional activators Bicoid (Bcd) and Caudal (Cad) and acts in a Dl-independent fashion. Addition of the VRE to the kni-element does not cause ventral repression, nor does addition of the VE or the VAE. This indicates that within the VE, Gro abolishes the activator function of Dl instead of converting Dl into a long-range repressor that interferes with transcriptional activation by Bcd and Cad (Hader, 1999).

To investigate whether this action of Gro on Dl is determined by the arrangement of Dri- and Dl-binding sites in the VE, the transcription patterns driven by a modified VE-kni-element were examined in which the normal distance of 91 bp between the binding sites was reduced to 45 bp. This reduction results in Dl-dependent repression along the ventral side of wild-type embryos. Repression is not observed in the absence of Gro or Dl or in embryos expressing the constitutively active Tor^Y9 protein. In contrast, the repression domain expands anteriorly in tor mutant embryos, which lack RTK signaling, and is found to be Dl-dependent. This suggests that the spatial arrangement of the Dl- and Dri-binding sites dictates the mechanism by which Gro and Dl act within the enhancer element. In one case, Dl is suppressed by Gro, in the other, Dl is converted into a potent silencer of transcription that can override activation by Bcd and Cad. Both modes of repression are controlled by Tor-dependent RTK signaling (Hader, 1999).

In the zen and dpp cis-acting elements, Gro causes Dl-mediated long-range silencing. Gro functions either by inhibiting the assembly and function of the core RNA polymerase II complex, by positioning nucleosomes over the core promoter and/or by recruiting the histone deacetylase Rpd3 to the template, where the enzyme can modulate local chromatin structure. However, in the VE, Gro only inhibits Dl-dependent activation without converting Dl into a repressor. The different modes of Gro function, that is, long-range silencing and short-range quenching, as shown here, are dependent on the distance between the Dl- and Dri-binding sites and/or their orientation on the enhancer, since shortening of the spacer distance converts the VE into a dpp- or zen-like element. This suggests that the way in which Gro regulates Dl activity depends on whether or not the two proteins can directly interact in vivo. Furthermore, both regulatory options of Gro on Dl are abolished by RTK signaling, a phenomenon that corresponds to the observation that Dl-dependent repression of dpp and zen is relieved by local Tor activity in the pole regions of the embryo. RTK-dependent phosphorylation may therefore interfere with the binding of Dri to the DNA template, the recruitment of Gro, or with both. Phosphorylation of the vertebrate Gro homolog TLE1 has been demonstrated, and many potential phosphorylation sites have been noted in Dri. Thus, local RTK-dependent phosphorylation may render one or both factors inactive, preventing Gro-dependent repression of Dl in the termini of the wild-type embryo (Hader, 1999).

These results establish that the cooperation between two maternal signaling systems, which determines the spatial limits of the Drosophila mesoderm anlage through hkb expression, is based on the management of the ubiquitously distributed factors Gro and Dri by local RTK signaling and that Gro can act through different modes on Dl. Lack of dead ringer (dri) activity does not result in an overt expansion of hkb expression on the ventral side of the embryo. However, as has been observed for VE-dependent gene expression, it causes only weak defects in mesoderm formation as compared with Gro-deficient embryos or embryos that express hkb under the control of the VAE. Thus, the interactions shown here represent only the Dri-dependent aspect of Gro's effect on hkb expression. The full picture of hkb control is likely to involve additional and redundantly acting factor(s) that recruit Gro to sites flanking the VE within the hkb control region (Hader, 1999).

ro^Dom is a dominant allele of rough (ro) that results in reduced eye size due to premature arrest in morphogenetic furrow (MF) progression. The ro^Dom stop-furrow phenotype is sensitive to the dosage of genes known to affect retinal differentiation, in particular members of the hedgehog (hh) signaling cascade. ro^Dom interferes with Hh's ability to induce the retina-specific proneural gene atonal (ato) in the MF and normal eye size can be restored by providing excess Ato protein. ro^Dom was used as a sensitive genetic background in which to identify mutations that affect hh signal transduction or regulation of ato expression. In addition to mutations in several unknown loci, multiple alleles of groucho (gro) and Hairless (H) were recovered. Analysis of their phenotypes in somatic clones suggests that both normally act to restrict neuronal cell fate in the retina, although they control different aspects of ato's complex expression pattern (Chanut, 2000).

Loss-of-function ro mutations cause eye roughness, due to mis-specification of photoreceptors R2 and R5, and the formation of ommatidia with more than one R8 photoreceptor. Repression of R8 cell fate has been attributed to inhibition of ato expression by the Ro homeodomain protein. In support of this proposal, Rough and Atonal proteins appear in complementary sets of cells behind the MF, and ato expression is expanded behind the MF in ro mutants. Generalized expression of ro under a heat-shock promoter (hs-ro) leads to loss of ato expression in the MF and eventually results in furrow arrest. Furrow arrest in ro^Dom is also accompanied by loss of ato expression in the MF. By analogy to the hs-ro phenotype, it is proposed that ro^Dom leads to excess Ro production, although that excess is not detectable by antibody staining (Chanut, 2000).

ro^Dom is very sensitive to alterations of ato gene dosage, since it is enhanced by loss-of-function ato alleles and almost completely rescued when high levels of ato expression are restored ahead of the MF. The ro^Dom phenotype therefore appears to result primarily from inhibition of ato expression due to excess Ro protein. On the basis of this understanding, the role of two of the strongest suppressors isolated in this screen, new alleles of gro and H, were analyzed on ato regulation and furrow progression (Chanut, 2000).

gro encodes a transcription inhibitor that combines with b-HLH genes of the E(spl) complex to inhibit expression of proneural genes such as achaete and scute. In gro mutant clones, expression of ato persists behind the MF longer than in wild-type tissue. This is consistent with a role for Gro in the N signaling events that lead to the refinement of ato expression behind the MF (Chanut, 2000).

However, Gro is also known to form inhibitory complexes with other transcription factors of the b-HLH class, such as Hairy, or of other classes, such as the c-Rel homolog Dorsal or the homeodomain, segment polarity regulator Engrailed. Association of Gro with Hairy deserves to be envisaged here, since Hairy has been implicated in inhibition of ato as well. It is found, however, that gro mutant clones expand ato expression posterior to the MF, whereas h inhibits ato expression anterior to the MF. Another intriguing possibility is that Gro associates with Ro to mediate inhibition of ato expression behind the MF. Although this hypothesis cannot be completely eliminated, it is unlikely, because complete loss-of-function phenotypes of ro and gro are different. While both of them lead to increased ato expression and imperfect R8 resolution, this effect is much more extensive and long lasting in gro mutant tissue than in ro mutant tissue. In addition, neuronal hyperplasia is not observed in ro mutant tissue, which suggests that at least this gro function must involve factors other than Ro. However, removal of E(spl) function results in neuronal hyperplasia and excess R8 development very similar to removal of gro function. Therefore, the hypothesis is favored that Gro restricts ato expression by combining with proteins of the E(spl) complex whose expression is induced by N signaling (Chanut, 2000).

Even in the complete absence of gro [or E(spl)] function, some refinement of ato expression still occurs, which indicates that factors independent of N and Gro play important roles in patterning Ato behind the MF. Candidates include Ro, the Egfr inhibitor Argos, and Hh. Moreover, outer photoreceptors differentiate in large excess between the R8 precursors and are the main cause of neuronal hyperplasia. Neuronal hyperplasia could occur as a direct consequence of the excess of R8 precursors in gro [and E(spl)] mutant tissue, which would, through the normal serial induction process, recruit an excess of neighboring cells into ommatidial clusters. However, differentiation of extra outer photoreceptors was observed with a hypomorphic gro allele in the absence of excess R8 differentiation. The excess of all photoreceptor types observed with a stronger gro allele may therefore reflect an involvement of gro in the restriction of cell fates at each step of ommatidial formation (Chanut, 2000).

gro mutant clones can also induce extensive overgrowth of head capsule and retinal tissues. In the wing, gro clones have been found to cause overgrowth via the induction of ectopic hh expression. Hh is also a powerful inducer of overgrowth in eye discs, when provided in excess or ectopically. However, overgrowth due to ectopic hh expression is accompanied by ectopic and premature photoreceptor differentiation, a phenotype not observed in overgrown gro mutant tissue. It is therefore unlikely that gro mutant clones cause ectopic hh expression in the eye. Besides, if gro mutations allowed increased Hh production, one would expect enhancement, rather than suppression, of the ro^Dom phenotype. While scenarios cannot be eliminated for ro^Dom in which a slight increase in cell proliferation allows the MF to progress further, it is more likely that gro mutations suppress ro^Dom by allowing Ato protein to persist longer in the MF (Chanut, 2000).

Finding that similar levels of suppression can be achieved by loss-of-function mutations in H and gro (which act in opposite directions in the N pathway) is not unique. A similar situation was encountered in another study where mutations in gro and H were both found to enhance the wing and bristle phenotypes associated with loss-of-function mutations in Egfr. The observation that mutations in both genes elevate ato expression in the vicinity of the MF, but at different stages of the differentiation process, helps resolve this paradox. The results also indicate that the exact timing (or location) of ato expression might not be crucial to MF progression, provided adequate levels are reached. This conclusion is supported by the finding that Ato supplied anterior to its normal expression domain, in the h expression domain, restores normal eye size in a ro^Dom background. Whether proper R8 spacing and ommatidial patterning can be achieved under these conditions remains to be shown (Chanut, 2000).

A function of Gro in imaginal development has been investigated, namely the repression of hedgehog in anterior wing pouch cells. hh is repressed in anterior compartments at least partly via Ci[rep], a form of the multifunctional transcription factor Cubitus interruptus (Ci). Cells in the wing primordium close to the AP boundary need gro activity to maintain repression of hh transcription, whereas in more anterior cells gro is dispensable. This repressive function of Gro does not appear to be mediated by Ci[rep]. Analysis of mutant gro transgenes has revealed that the Q and WD40 domains are both necessary for hh repression. Yet, deletion of the WD40 repeats does not always abolish Gro activity. These findings provide new insights both into the mechanisms of AP patterning of the wing and into the function of Gro (Apidianakis, 2001).

Although Ci[rep]-mediated repression can account for the lack of hh expression away from the AP boundary, it has not been firmly established that Ci[rep] is operational close to the AP boundary. These cells receive high Hh signal and as a result not only do they not process Ci to Ci[rep], but also they activate full-length Ci into a strong activator, Ci[act], by post-translational modification. There is indirect evidence that Hh-receiving cells do not contain sufficient Ci[rep] levels to repress hh: in posterior cells, ci is repressed by En; other than this, the cellular mechanism for Hh signal transduction is present. When full-length ci is provided by ectopic expression in the posterior compartment, hh-lacZ is not repressed. This suggests that these cells cannot produce appreciable amounts of Ci[rep], consistent with their responding to Hh signaling. That this is indeed the case was shown by the fact that ectopic expression of ci does repress posterior hh-lacZ in smo loss-of-function clones, where the Hh signal transduction has been disrupted. If anterior cells that are exposed to Hh behave similarly, then the lack of hh expression there cannot be attributed to Ci[rep]. It is proposed that a Gro-dependent repression complex supplies this function, since gro^- clones exhibit strong derepression of hh-lacZ near the AP boundary. The Gro complex is not required in anterior cells far from the boundary, because those receive no Hh signal and thus contain sufficient Ci[rep] to repress hh. Accordingly, by supplying increased levels of Ci[rep] near the AP boundary via the ci^Ce2 allele, the need for Gro-mediated hh repression is able to be largely abolished, with the exception of the DV boundary. Since Gro is a ubiquitous co-repressor, one has to postulate the existence of a DNA-tethering factor, which will be referred to as 'X' for the purpose of this discussion, and some process of spatial regulation of the X-Gro complex activity. The possibility that X is a form of Ci itself was tested and the answer was negative: using three different assays -- GST pulldowns, yeast two-hybrid and transfection colocalization -- no interaction between Gro and either form of Ci could be shown. Most importantly, the fact that Ci[rep] does not require Gro to repress hh in anterior cells away from the boundary supports a model where Ci and Gro repress hh independently of each other (Apidianakis, 2001).

The quantitative aspect of hh derepression in gro^- clones is intriguing: clones abutting the AP boundary (type I) express the highest hh-lacZ levels, which drop gradually as clones arise further from the P compartment. This might reflect the fact that Ci[rep]-dependent repression gradually increases away from the boundary, and this is independent of gro. This interpretation assumes that basal (unrepressed) hh transcription in the A compartment would be high and subject to the dual repressors (Ci and X-Gro). Alternatively, basal hh transcription could be low, but, in addition to the repression control, hh could display a positive response to Hh signaling at the AP boundary. The latter model is consistent with the fact that in ci^- cells, basal hh expression appears to be low. It also agrees with the behavior of large type I gro^- clones in the present study. In these clones, high levels of hh-lacZ could be observed throughout the clone, even at a distance from the AP boundary. This could be accounted for by Hh signaling, which, having risen over some threshold owing to hh derepression, further stimulates hh transcription to a high level. This effect would spread to the edge of the clone, beyond which activation of the X-Gro repressor would silence hh transcription. The putative inducer of hh by Hh signaling may be Ci[act], as with all other direct Hh target genes; alternatively, it may be another factor induced by Ci[act]. The hypothesis that Ci[act] itself can activate hh transcription is not unreasonable, since hh should contain a regulatory region(s) that bind(s) Ci[rep]. Ci[act] and Ci[rep] contain the same DNA-binding domain and recent work has shown that the two forms of Ci bind the same target sites, although some enhancers may be configured in such a way as to respond preferentially to either the activator or repressor form (Apidianakis, 2001).

For the sake of simplicity, the existence of a low level ubiquitous activator of hh (basal levels) with a stronger activator located in P cells is postulated to account for the high levels of hh expression in P cells. In A cells that do not receive the Hh signal, the basal activity of hh is repressed by Ci[rep] and gro is not required. In A cells close to the Hh source, the basal transcription of hh would be enhanced by positive autoregulation; however, the presence of the repressive X-Gro complex does not allow this activation to take place. Implicit in this model is that X is itself activated by Hh (e.g. transcriptionally induced via Ci[act]), so that it only functions in Hh-receiving cells. In addition X production/activity should be spatially limited to the A compartment (e.g. repressed by En), since ectopic expression of full-length ci in the posterior cannot induce X-Gro activity to repress endogenous hh. According to this model, ci^- clones close to the AP boundary express basal hh levels, since they lack both the X-Gro repressor (no activation of X in the absence of Ci[act]) and the activator of hh transcription (Ci[act] itself or a downstream target). By contrast, gro^- clones in the same region only lack the repressive X-Gro complex and thus actively transcribe hh in response to Ci[act]; the high levels of hh produced are sufficient to initiate Hh signaling, which can propagate this effect of hh derepression throughout the clone (Apidianakis, 2001).

gro^- clones near the DV boundary behave somewhat aberrantly. hh-lacZ derepression there is more efficient, observable in further anteriorly arising clones, compared with equivalent clones away from the DV boundary -- it even occurs in the presence of increased Ci[rep]. Although the mechanism remains to be discovered, one way to account for this special behavior, without invoking additional regulators, is that Ci[rep] is less active near the DV boundary and/or Ci[act] is more active, and this modulation of Ci activity in favor of the activator form allows high level hh expression at a greater distance from the Hh source and even in the ci^Ce2/+ background. Interestingly, ci^- clones show little or no hh-lacZ derepression at the DV boundary, consistent with Gro, rather than Ci[rep], being the major hh repressor there (Apidianakis, 2001).

The model put forward here is perhaps the simplest, but by no means the only one that fits the existing data. For example, Gro might interact with Ci[act] itself, switching it from an activator into a repressor, given the right enhancer context, much like the effect Gro has on other activators, such as Dorsal. This interaction may be weak and/or require additional factors, accounting for the inability to detect it. To resolve the mechanism of hh repression at the AP boundary will necessitate detailed molecular dissection of the hh regulatory regions and characterization of relevant trans acting factors. Whatever the mechanism, it appears that a Gro-containing complex is deployed in the wing to block the spread of hh expression anteriorly from the AP compartment boundary. This should ensure a spatially fixed organizer (dpp expression stripe), in contrast to a moving one, as found in the fly retina (Apidianakis, 2001).

Gro is the founder of a family of transcriptional co-repressors encountered in invertebrates and vertebrates. Gro proteins are multipurpose co-repressors, since they can interact with a good number of DNA-binding repressors. A number of Gro mutants were tested both for subcellular localization. Grocdc2^- and GroDeltaQ show the same nuclear accumulation as wild-type Gro. GroDeltaWD40 is also nuclear, but it shows a striking departure from the rather uniform wild-type pattern, since it localizes predominantly to a small number of subnuclear particles. GroNLS^- is both nuclear and cytoplasmic, whereas GroDeltaGCS is exclusively cytoplasmic. This suggests that the GP, CcN and SP domains contain at least two different regions needed for efficient nuclear accumulation, one of which is the canonical NLS. It can be speculated that other such regions might be those necessary for association with histones or with DNA-bound repressors, which might promote nuclear accumulation of Gro even in the absence of the NLS (Apidianakis, 2001).

In vivo activity was tested by assaying the ability of mutant Gro proteins to repress anterior hh-lacZ expression. GroDeltaQ and GroDeltaWD40 proteins were inactive in this assay. In contrast, Grocdc2^- was as active as wild-type Gro. The inability of GroDeltaQ to function as a co-repressor is expected, since the Q domain is the strongest repression domain and is needed both for tetramerization as well as for histone interaction. The inactivity of the GroDeltaWD40 mutant might be accounted by its inability to interact with the X-factor tether. Or one could suggest an alternative explanation based on the localization data: that GroDeltaWD40 is retained in subnuclear particles and as a result cannot gain access to target genetic loci. Whether the aberrant subnuclear localization of GroDeltaWD40 is a cause or a consequence of its inactivity is a matter for further study. Despite its aberrant localization, GroDeltaWD40 is as active as wild-type Gro and Grocdc2^- when overexpressed by omb-Gal4: all three transgenes results in abnormal leg development. Gro-DeltaQ, -NLS^- and -DeltaGCS did not have such an effect. This shows that GroDeltaWD40 retains some activity, although in the absence of data regarding the cause of defects in leg patterning, the function of the mutant protein is as yet unknown. 'Short' Gro family proteins that lack WD40 repeats exist in vertebrates. These, human AES and mouse Grg5, contain only Q and GP domains, thus they are not entirely equivalent to the DeltaWD40 mutant. It has been shown that these proteins are cytoplasmic, although they are readily transported to the nucleus upon interaction with a Tcf partner. Their role in transcription seems to be context dependent, since they can act as co-repressors in some cases, whereas in others they might counter repression by 'long' Gro proteins. One study suggests that this anti-repression effect is not necessarily due to the absence of the CcN/SP/WD40 domains, but rather due to the inability of the GP domain of the 'short' proteins to interact with HDAC1. In this study, GroDeltaWD40 was active in one assay and inactive in another. It will be interesting to determine its activity in additional biological contexts where Gro is required (Apidianakis, 2001).

Low-level ectopic expression of the Runt transcription factor blocks activation of the Drosophila melanogaster segmentation gene engrailed (en) in odd-numbered parasegments and is associated with a lethal phenotype. By using a genetic screen for maternal factors that contribute in a dose-dependent fashion to Runt-mediated repression, it is shown that there are two distinct steps in the repression of en by Runt. The initial establishment of repression is sensitive to the dosage of the zinc-finger transcription factor Tramtrack. By contrast, the co-repressor proteins Groucho and dCtBP, and the histone deacetylase Rpd3, do not affect establishment but instead maintain repression after the blastoderm stage. The distinction between establishment and maintenance is confirmed by experiments with Runt derivatives that are impaired specifically for either co-repressor interaction or DNA binding. Other transcription factors can also establish repression in Rpd3-deficient embryos: this indicates that the distinction between establishment and maintenance may be a general feature of eukaryotic transcriptional repression (Wheeler, 2002).

Long- and short-range transcriptional repressors induce distinct chromatin states on repressed genes

Transcriptional repression is essential for establishing precise patterns of gene expression during development. Repressors governing early Drosophila segmentation can be classified as short- or long-range factors based on their ranges of action, acting either locally to quench adjacent activators or broadly to silence an entire locus. Paradoxically, these repressors recruit common corepressors, Groucho and CtBP, despite their different ranges of repression. To reveal the mechanisms underlying these two distinct modes of repression, chromatin analysis was performed using the prototypical long-range repressor Hairy and the short-range repressor Knirps. Chromatin immunoprecipitation and micrococcal nuclease mapping studies reveal that Knirps causes local changes of histone density and acetylation, and the inhibition of activator recruitment, without affecting the recruitment of basal transcriptional machinery. In contrast, Hairy induces widespread histone deacetylation and inhibits the recruitment of basal machinery without inducing chromatin compaction. This study provides detailed mechanistic insight into short- and long-range repression on selected endogenous target genes and suggests that the transcriptional corepressors can be differentially deployed to mediate chromatin changes in a context-dependent manner (Li, 2011).

To directly compare functional aspects of Hairy- and Knirps- mediated repression in the Drosophila embryo, these proteins’ interactions were studied with two segmentally expressed pair-rule genes. Hairy directly represses fushi tarazu (ftz), a secondary pair-rule gene expressed in the blastoderm embryo in a seven-stripe pattern. ftz is regulated by both regionally acting gap genes and the segmentally expressed hairy pair-rule gene. Chromatin immunoprecipitation (ChIP) experiments have revealed dense clusters of peaks around the ftz gene for key transcription factors active in the blastoderm embryo, including Caudal, Hunchback, Knirps, Giant, Huckebein, Krüppel, and Tailless. These transcription factors bind to the promoter-proximal Zebra element, the stripe 1+5 enhancer located 3' of ftz, and a presumptive 5' regulatory region located between 23 kbp and 28 kbp. Hairy has been found to bind in vivo to all of these regions. This repressor is expressed in a striped pattern in the blastoderm embryo; therefore, the ftz gene is active in some nuclei and repressed in others. In order to obtain a homogeneous population of nuclei for chromatin studies, Hairy protein was overexpressed in embryos using a heat-shock driver, which results in complete repression of ftz. This repression requires the recruitment of the Groucho corepressor, because a mutant version of Hairy that does not bind to Groucho fails to repress ftz (Li, 2011).

Interestingly, a titration of heat-shock induction resulted in a nonuniform, progressive loss of specific ftz stripes, with stripe 4 being the most sensitive and stripe 1+5 the least. This result points to the intriguing possibility that Hairy can act locally on specific enhancers, at least very transiently, although the end result of Hairy repression is complete silencing of all enhancer elements. The asynchronous repression of the ftz locus also suggests that Hairy-mediated long-range repression does not act solely by direct targeting the basal promoter, as suggested by a previous model for this class of repressor, because this mechanism should cause uniform inhibition of stripe elements. Similar to ftz, the pair-rule gene even skipped (eve) is also expressed in a seven-stripe pattern and is regulated by multiple modular enhancers. eve is a well-characterized target of the short-range repressor Knirps, which sets posterior boundaries of eve stripe 3 and 4 and anterior borders of eve stripe 6 and 7. After substantial overexpression of Knirps (20 min heat-shock induction), the repressor is able to repress all of the eve stripe enhancers except for the stripe 5 enhancer. When the induction is titrated, Knirps represses individual enhancers in a stepwise manner, with the most sensitive enhancers downregulated earliest, at a low dose of Knirps. Together, these experiments indicate that Hairy can initially act locally but ultimately acts in a globally dominant fashion, whereas Knirps acts in a restricted manner (Li, 2011).

To compare the effects of repression by Hairy and Knirps, chromatin changes associated with repression of ftz and eve were studied via ChIP. No significant change of histone H3 occupancy were detected at regions sampled throughout the ftz locus after Hairy overexpression (although some regions showed modest differences. In contrast, Knirps repression of eve resulted in significantly increased histone H3 density, particularly in two of the three regions corresponding to the Knirps-sensitive enhancers, namely stripe 4+6 and stripe 2. Little change was noted in the promoter region, transcribed region, or the stripe 1 and 5 enhancers, which are not readily repressed by Knirps. An apparent increase in histone H3 density on the repressed stripe 3+7 enhancer, although of low statistical significance, correlates with other alterations common to repressed enhancers, noted below (Li, 2011).

To provide a more detailed picture of chromatin structure, a micrococcal nuclease (MNase) mapping protocol used in yeast and cultured cells was adapted for Drosophila embryos. MNase mapping showed that Hairy repression had little effect on chromatin accessibility throughout the ftz locus, whereas Knirps induced a significant increase in MNase insensitivity specifically at the eve stripe 3+7, 2, and 4+6 enhancers and a minor increase in stripe 1 protection. The promoter and the eve stripe 5 enhancer were little changed, mirroring the patterns noted for overall histone H3 occupancy. The changes noted for the eve locus appear to be specific, because Knirps did not induce any change of a nontargeted intergenic site on the third chromosome. Hairy also had no effect at this locus. The similar results from overall histone H3 density and MNase mapping suggest that Hairy-mediated long-range repression does not involve a general compaction of chromatin on the ftz locus. In contrast, repression by Knirps is associated with an increase in the histone density of targeted enhancer regions, which may result either from Knirps recruitment of factors that mediate chromatin condensation or the blocking of proteins responsible for loosening of chromatin. Recruitment of Groucho by other repressor proteins is also associated with distinct effects: Runt-dependent repression of slp1 does not involve changes in H3 density, but Brinker repression of the vgQ enhancer does. The distance dependence of these repressors has not been established, but in light of the current results, it is apparent that the Groucho corepressor can be involved in distinct effects depending on the context of recruitment (Li, 2011).

Histone acetylation is dynamically regulated on transcribed genes in eukaryotes, with histone acetylation generally correlated with active loci. The histone deacetylase Rpd3 is a component of both Hairy and Knirps corepressor complexes; therefore, histone acetylation levels were assayed across the eve and ftz genes before and after repression. Hairy repression resulted in widespread histone H4 deacetylation throughout the ftz locus. The ectopically expressed Hairy protein itself was not observed to spread but remained restricted to regions of the gene previously observed to bind endogenous Hairy. Using anti-H3-acetylation antibodies, similar widespread H3 deacetylation was also noted. This distributed effect on the ftz locus correlates with prior observations that Hairy-mediated long-range repression might involve a Groucho-mediated 'spreading' mechanism. By this means, Rpd3 may be delivered to extensive areas of a gene. To test whether a spreading of histone deacetylation might correlate with the successive inhibition of ftz enhancers, histone acetylation levels were investigated across ftz after a brief 5 min heat shock followed by immediate fixing, before the entire complement of enhancers can be repressed. In this setting, deacetylation was mostly concentrated around the stripe 1+5 enhancer and the immediate 5' regulatory region, areas that show Hairy occupancy in vivo. More distal 5' regulatory regions and the transcription unit itself showed little initial change, consistent with a spreading action of this repressor during the more extensive repression period (Li, 2011).

A different picture emerged from studies of Knirps acting on eve. Here, repression led to selective decreases in H3 and H4 acetylation levels, concentrated over the eve stripe 4+6 and stripe 2 enhancers, with lesser decreases noted at stripe 3+7 and stripe 1 enhancers. A local change in acetylation was also noted near the transcriptional initiation site, but not immediately 5' and 3' of this area. The reductions in histone acetylation levels seen on both eve and ftz are consistent with Hairy and Knirps recruiting deacetylases to their target genes. However, it is striking that the broad deacetylation mediated by Hairy on ftz is not associated with dramatic changes in histone density or resistance to nuclease accessibility, whereas increased histone density and resistance to nuclease digestion are associated with Knirps repression on eve. It is possible that in addition to inducing deacetylation, Knirps triggers additional histone modifications or interacts with nucleosome-remodeling complexes to further alter chromatin at the enhancers. H3 lysine 27 methylation is one chromatin signature associated with silenced genes; however, no significant change in this modification was noted at ftz or eve upon repression (Li, 2011).

Previous studies indicated that Hairy can effectively repress a reporter gene without displacing the activators. Attempts were made to test whether this was the case on an endogenous gene, ftz, by examining occupancy by Caudal, a transcription factor that also activates eve. Caudal activates the posterior stripes of both ftz and eve, and it was found that Caudal binds the ftz 5' regulatory region and the promoter-proximal Zebra element. Repression of the locus by Hairy did not affect the Caudal binding pattern, similar to the results obtained with a Hairy-regulated reporter gene. In contrast, Knirps repression decreased Caudal occupancy specifically at the eve 3+7 and 4+6 enhancers, bringing overall protein occupancy down to near baseline levels. This decrease is not an effect of global decrease of Caudal occupancy, because the Caudal binding peak at the eve promoter was not affected. A similar decrease in Caudal occupancy was also observed on a hunchback enhancer after repression by Knirps. Interestingly, Bicoid occupancy of the eve stripe 2 and stripe 1 enhancers was not altered by Knirps, although these enhancers were repressed. Clearly, loss of transcription factor occupancy is not required for short-range repression of a cis-regulatory element. It is possible that different transcriptional activators exhibit differential sensitivity to chromatin changes induced during repression (Li, 2011).

New insights have suggested that many developmental genes, including those regulated by short-range repressors such as Snail, feature RNA polymerase paused in the promoter region even in their inactive state, suggesting postrecruitment levels of regulation. Components of the core machinery were analyzed before and after repression by Hairy and Knirps. Upon Hairy repression, a marked decrease of RNA polymerase II (Pol II) occupancy was observed at the ftz locus. The same trend was observed for the preinitiation, initiation, and elongation forms of Pol II. These results suggest that Hairy directly or indirectly blocks recruitment of Pol II. Similar decreases were noted with levels of TATA box-binding protein (TBP) at the promoter (Li, 2011).

In contrast, induction of Knirps did not change Pol II occupancy at the eve transcription unit, even under condition where most enhancers were repressed. (Under conditions tested in this study, over three-quarters of the embryos had shut down expression of all but stripe 1 and/or 5.) Similarly, TBP occupancy remained at a comparable level before and after Knirps repression. The constant level of RNA polymerase on the eve transcription unit was a surprise in light of the sharp reduction in mRNA production as measured by in situ hybridization. However, there is precedence for this effect: Runt repression of slp1 appears to act through elongation control, which causes no change of the concentration of Pol II on slp1. Knirps may produce a similar effect by inducing a slower transit rate of Pol II on the repressed eve locus. Similar observations have been made at the hsp70 gene upon depletion of elongation factors such as Spt6 or Paf1 (Li, 2011).

The differential distance dependence of short- and long- range repressors such as Hairy and Knirps has been observed in many contexts. However, the mechanisms by which these proteins function have been poorly understood. With the recent demonstration that transcriptional factors considered to be short- and long-range repressors utilize shared cofactors, namely CtBP and Groucho, there has been a question of whether long-range repression is actually functionally distinct from short-range repression (Payankaulam, 2009). The current study provides evidence that the chromatin states associated with long- and short-range repressors are distinct in several ways. It is not yet knowm whether the effects seen on ftz are observed for all Hairy targets, although the similarity of changes observed on the lacZ reporter subject to Hairy repression suggests that they are conserved (Martinez, 2008). Similarly, the reproducibility of Knirps-induced changes at different eve enhancers indicates that this protein can effect related chromatin changes on cis-regulatory modules bound by different activators. Snail, another short-range repressor, also appears to mediate localized deacetylation and activator displacement; thus, this mechanism may be a common feature of this entire class of repressors (Qi, 2009; Y. Nibu, personal communication to Li, 2011). It will be interesting to determine how general are the observations made in this study for long- and short-range repression, a question that can be approached using genome-wide methods. In any event, the highly divergent activities of Knirps and Hairy demonstrated in this study not only underscore the fact that these proteins can mediate biochemically divergent events but also raise interesting questions about how similar cofactors can participate in such distinct effects in a context-dependent manner. It is possible that the corepressors adopt distinct conformations when recruited by different repressors, or the corepressor may form distinct complexes with unique activities. In addition to determining how cis- and trans-acting factors affect repression pathways, these mechanistic insights will provide important contextual information for interpretation of genome-wide transcription factor binding and chromatin modifications and will inform quantitative modeling of cis-regulatory elements for the aim of understanding the activity and evolution of enhancers (Li, 2011).

RTK signaling modulates the Dorsal gradient

The dorsoventral (DV) axis of the Drosophila embryo is patterned by a nuclear gradient of the Rel family transcription factor, Dorsal (Dl), that activates or represses numerous target genes in a region-specific manner. This study demonstrates that signaling by receptor tyrosine kinases (RTK) reduces nuclear levels and transcriptional activity of Dl, both at the poles and in the mid-body of the embryo. These effects depend on wntD, which encodes a Dl antagonist belonging to the Wingless/Wnt family of secreted factors. Specifically, it was shown that, via relief of Groucho- and Capicua-mediated repression, the Torso and EGFR RTK pathways induce expression of WntD, which in turn limits Dl nuclear localization at the poles and along the DV axis. Furthermore, this RTK-dependent control of Dl is important for restricting expression of its targets in both contexts. Thus, the results reveal a new mechanism of crosstalk, whereby RTK signals modulate the spatial distribution and activity of a developmental morphogen in vivo (Helman, 2012).

Specification of body axes in all metazoans is initiated by a small number of inductive signals that must be integrated in time and space to control complex and unique patterns of gene expression. It is therefore of utmost importance to unravel the mechanisms underlying crosstalk between different signaling cues that concur during early development. This study has elucidated a novel signal integration mechanism that coordinates RTK signaling pathways with the Dl nuclear gradient, and thus with terminal and DV patterning of the Drosophila embryo (Helman, 2012).

Previous work had identified an input by Torso signaling into specific transcriptional effects of Dl. The current results establish a general mechanism, which involves RTK-dependent control of the nuclear Dl gradient itself, and thus affects a large group of Dl targets. This regulatory input is based on RTK-dependent derepression of wntD, a Dl target that encodes a feedback inhibitor of the Dl gradient. Thus, Dl activates wntD effectively only when accompanied by RTK signaling, enabling region-specific negative-feedback control of the nuclear Dl gradient. In the absence of RTK signaling, wntD is not expressed and the levels of nuclear Dl are elevated. Consequently, Dl target genes are ectopically expressed, both at the poles and along the DV axis (Helman, 2012).

Torso RTK signaling depends on maternal cues and is independent of the Dl gradient. Thus, it can be viewed as a gating signal that operates only at the embryonic poles, where it controls Dl-dependent gene regulation. However, the activity of the EGFR RTK pathway later on in development crucially depends on Dl, which induces the neuroectodermal expression of rhomboid, a gene encoding a serine protease required for processing of the EGFR ligand Spitz. In this case, EGFR-dependent induction of WntD represents a negative feedback loop that reduces nuclear levels of Dl laterally and, consequently, limits the expression of multiple Dl targets along the DV axis (Helman, 2012).

It should be noted that the regulatory interactions that have been characterized do not preclude the existence of other mechanisms modulating nuclear Dl concentration or activity. For example, the progressive dilution or degradation of maternal components involved in Toll receptor activation upstream of Dl should cause reduced Dl nuclear accumulation and retraction of its targets as development proceeds. It is also possible that Torso- or EGFR-induced repressors block transcription of Dl target genes directly. Accordingly, the ectopic sna expression observed in embryos mutant for components of the Torso pathway such as DSor and trunk probably reflects both loss of WntD activity on Dl and loss of Hkb-mediated repression of sna. In this context, it is interesting to note that sna expression expands and colocalizes with Hkb at the poles of wntD mutants; perhaps repression of sna by Hkb is not sufficient to override increased Dl activation in this genetic background. Thus, the Torso pathway probably employs more than one mechanism to exclude Dl target expression from the termini. Furthermore, the existence of such additional regulatory mechanisms could explain why wntD mutants do not have a clear developmental phenotype, despite the broad effects on Dl-dependent gene expression patterns caused by the genetic removal of wntD. It is proposec that corrective mechanisms are present, which make the terminal and DV systems robust with respect to removal of the WntD-based feedback, such as RTK-induced repressors. Understanding the basis of this robustness will require additional studies (Helman, 2012).

This work shows that RTK-dependent relief of Gro- and Cic-mediated repression is essential for transcriptional activation of wntD by Dl. Correspondingly, in the absence of cic or gro, the early expression of wntD expands ventrally throughout the domain of nuclear Dl. The early onset of this derepression, and the presence of at least one conserved Cic-binding site in the proximal upstream region of wntD, indicate that repression of wntD may be direct. Interestingly, it is thought that Gro and Cic are also involved in assisting Dl-mediated repression of other targets such as dpp and zen, as gro and cic mutant embryos show derepression of those targets in ventral regions. However, as ectopic wntD expression in these mutants leads to reduced nuclear localization of Dl along the ventral region, it is conceivable that decreased Dl activity also contributes to the derepression of dpp and zen (Helman, 2012).

In conclusion, the data presented in this study demonstrate RTK-dependent control of nuclear Dl via wntD, based on multiple regulatory inputs, including negative gating, feed-forward loops and negative feedback control. Together, these mechanisms provide additional combinatorial tiers of spatiotemporal regulation to Dl target gene expression. Future studies will show whether other signal transduction cascades and/or additional developmental cues also impinge on the Dl morphogen gradient (Helman, 2012).

The Groucho co-repressor is primarily recruited to local target sites in active chromatin to attenuate transcription

Gene expression is regulated by the complex interaction between transcriptional activators and repressors, which function in part by recruiting histone-modifying enzymes to control accessibility of DNA to RNA polymerase. The evolutionarily conserved family of Groucho/Transducin-Like Enhancer of split (Gro/TLE) proteins act as co-repressors for numerous transcription factors. Gro/TLE proteins act in several key pathways during development (including Notch and Wnt signaling), and are implicated in the pathogenesis of several human cancers. Gro/TLE proteins form oligomers and it has been proposed that their ability to exert long-range repression on target genes involves oligomerization over broad regions of chromatin. However, analysis of an endogenous gro mutation in Drosophila revealed that oligomerization of Gro is not always obligatory for repression in vivo. This study used chromatin immunoprecipitation followed by DNA sequencing (ChIP-seq) to profile Gro recruitment in two Drosophila cell lines. Gro was found to predominantly bind at discrete peaks (<1 kilobase). It was also demonstrated that blocking Gro oligomerization does not reduce peak width as would be expected if Gro oligomerization induced spreading along the chromatin from the site of recruitment. Gro recruitment is enriched in 'active' chromatin containing developmentally regulated genes. However, Gro binding is associated with local regions containing hypoacetylated histones H3 and H4, which is indicative of chromatin that is not fully open for efficient transcription. It was also found that peaks of Gro binding frequently overlap the transcription start sites of expressed genes that exhibit strong RNA polymerase pausing and that depletion of Gro leads to release of polymerase pausing and increased transcription at a bona fide target gene. These results demonstrate that Gro is recruited to local sites by transcription factors to attenuate rather than silence gene expression by promoting histone deacetylation and polymerase pausing (Kaul, 2014; 25165826).

An ancient Pygo-dependent Wnt enhanceosome integrated by Chip/LDB-SSDP

TCF/LEF factors (see Drosophila Pangolin) are ancient context-dependent enhancer-binding proteins that are activated by β-catenin (see Drosophila Armadillo) following Wnt signaling. They control embryonic development and adult stem cell compartments, and their dysregulation often causes cancer. β-catenin-dependent transcription relies on the NPF motif of Pygo proteins. This study used a proteomics approach to discover the Chip/LDB-SSDP (ChiLS) complex as the ligand specifically binding to NPF. ChiLS also recognizes NPF motifs in other nuclear factors including Runt/RUNX2 and Drosophila ARID1, and binds to Groucho/TLE. Studies of Wnt-responsive dTCF enhancers in the Drosophila embryonic midgut indicate how these factors interact to form the Wnt enhanceosome, primed for Wnt responses by Pygo. Together with previous evidence, this study indicates that ChiLS confers context-dependence on TCF/LEF by integrating multiple inputs from lineage and signal-responsive factors, including enhanceosome switch-off by Notch. Its pivotal function in embryos and stem cells explain why its integrity is crucial in the avoidance of cancer (Fiedler, 2015).

TCF/LEF factors (TCFs) were discovered as context-dependent architectural factors without intrinsic transactivation potential that bind to the T cell receptor α (TCRα) enhancer via their high mobility group (HMG) domain. They facilitate complex assemblies with other nearby enhancer-binding proteins, including the signal-responsive CRE-binding factor (CREB) and the lineage-specific RUNX1 (also called Acute Myeloid Leukemia 1, AML1). Their activity further depends on β-catenin, a transcriptional co-factor activated by Wnt signaling, an ancient signaling pathway that controls animal development and stem cell compartments, and whose dysregulation often causes cancer. The context-dependence of TCFs is also apparent in other systems, for example in the embryonic midgut of Drosophila where dTCF integrates multiple signaling inputs with lineage-specific cues during endoderm induction. The molecular basis for this context-dependence remains unexplained (Fiedler, 2015).

In the absence of signaling, T cell factors (TCFs) are bound by the Groucho/Transducin-like Enhancer-of-split (Groucho/TLE) proteins, a family of co-repressors that silence TCF enhancers by recruiting histone deacetylases (HDACs) and by 'blanketing' them with inactive chromatin. TLEs are displaced from TCFs by β-catenin following Wnt signaling, however this is not achieved by competitive binding but depends on other factors. One of these is Pygopus (Pygo), a conserved nuclear Wnt signaling factor that recruits Armadillo (Drosophila β-catenin) via the Legless/BCL9 adaptor to promote TCF-dependent transcription. Intriguingly, Pygo is largely dispensable in the absence of Groucho, which implicates this protein in alleviating Groucho-dependent repression of Wg targets (Fiedler, 2015).

Pygo has a PHD and an N-terminal asparagine proline phenylalanine (NPF) motif, each essential for development and tissue patterning. Much is known about the PHD finger, which binds to Legless/BCL9 and to histone H3 tail methylated at lysine 4 via opposite surfaces that are connected by allosteric communication. By contrast, the NPF ligand is unknown, but two contrasting models have been proposed for its function (1">Figure 1) (Fiedler, 2015).

This study used a proteomics approach to discover that the NPF ligand is an ancient protein complex composed of Chip/LDB (LIM-domain-binding protein) and single-stranded DNA-binding protein (SSDP), also called SSBP. This complex controls remote Wnt- and Notch-responsive enhancers of homeobox genes in flies (Bronstein, 2011), and remote enhancers of globin and other erythroid genes in mammals, integrating lineage-specific inputs from LIM-homeobox (LHX) proteins and other enhancer-binding proteins. Using nuclear magnetic resonance (NMR) spectroscopy, this study demonstrated that Chip/LDB-SSDP (ChiLS) binds directly and specifically to Pygo NPFs, and also to NPF motifs in Runt-related transcription factors (RUNX) proteins and Osa (Drosophila ARID1), whose relevance is shown by functional analysis of Drosophila midgut enhancers. Furthermore, Groucho was identified as another new ligand of ChiLS by mass spectroscopy. This study thus define the core components of a Wnt enhanceosome assembled at TCF enhancers via Groucho/TLE and RUNX, primed for timely Wnt responses by ChiLS-associated Pygo. The pivotal role of ChiLS in integrating the Wnt enhanceosome provides a molecular explanation for the context-dependence of TCFs (Fiedler, 2015).

The discovery of ChiLS as the NPF ligand of Pygo proteins led to the definition of the core components of a multi-protein complex tethered to TCF enhancers via Groucho/TLE and RUNX, and slated for subsequent Wnt responses by Pygo (see Model of the Wnt enhanceosome). ChiLS also contacts additional enhancer-binding proteins via its LID, including lineage-specific and other signal-responsive factors, thereby integrating multiple position-specific inputs into TCF enhancers, which provides a molecular explanation for the context-dependence of TCF/LEF. This complex will be referred to as the Wnt enhanceosome since it shares fundamental features with the paradigmatic interferon β-responsive enhanceosome (Panne, 2007). Its components are conserved in placozoa, arguably the most primitive animals without axis and tissues with only a handful of signaling pathways including Wnt, Notch and TGF-β/SMAD, suggesting that the Wnt enhanceosome emerged as the ur-module integrating signal-responses (Fiedler, 2015).

Other proteins have been reported to interact with the Pygo N-terminus, but none of these recognize NPF. It is noted that this N-terminus is composed of low-complexity (intrinsically disordered) sequences that are prone to non-specific binding (Fiedler, 2015).

NPF is a versatile endocytosis motif that binds to structurally distinct domains, including eps15 homology (EH) domains in epsin15 homology domain (EHD) protein. Indeed, EHDs were consistently identified in lysate-based pull-downs with triple-NPF baits. EHDs are predominantly cytoplasmic, and do not interact with nuclear Pygo upon co-expression, nor are any of the Drosophila EHDs required for Wg signaling in S2 cells. ChiLS is the first nuclear NPF-binding factor (Fiedler, 2015).

NPF binding to ChiLS appears to depend on the same residues as NPF binding to EHD domains, that is, on the aromatic residue at +2, the invariant P at +1, N (or G) at 0 and NPF-adjacent residues, including negative charges at +3 and +4 (whereby a positive charge at +3 abolishes binding to EHD). Indeed, an intramolecular interaction between the +3 side-chain and that of N predisposes NPF to adopt a type 1 β-turn conformation, which increases its affinity to the EHD pocket, while the -1 residue undergoes an intermolecular interaction with this pocket. ChiLS also shows a preference for small residues at -1 and -2, similarly to N-terminal EHDs although RUNX seems to differ at -1 and -2 from Pygo and MACC1 (F/L A/E/D vs S A, respectively) (Fiedler, 2015).

Groucho/TLE is recruited to TCF via its Q domain, which tetramerizes. Intriguingly, the short segment that links two Q domain dimers into a tetramer is deleted in a dTCF-specific groucho allele that abolishes dTCF binding and Wg responses, suggesting that TCF may normally bind to a Groucho/TLE tetramer (Fiedler, 2015).

Groucho/TLE uses its second domain, the WD40 propeller, to bind to other enhancer-binding proteins on Wnt-responsive enhancers, most notably to the C-terminal WRPY motif of RUNX proteins (common partners of TCFs in Wnt-responsive enhancers). This interaction can occur simultaneously with the WD40-dependent binding to ChiLS, given the tetramer structure of Groucho/TLE. In turn, RUNX uses its DNA-binding Runt domain to interact with HMG domains of TCFs, and to recruit ChiLS. RUNX thus appears to be the keystone of the Wnt enhanceosome since it binds to the enhancer directly while undergoing simultaneous interactions with Groucho/TLE (through its C-terminal WRPY motif), TCF and ChiLS (though its Runt domain) (Fiedler, 2015).

In line with this, Runt has pioneering functions in the early Drosophila embryo, shortly after the onset of zygotic transcription, and in the naïve endoderm as soon as this germlayer is formed, in each case prior to the first Wg signaling events. RUNX paralogs also have pioneer-like functions in specifying cell lineages, that is, definitive hematopoiesis in flies and mammals (Fiedler, 2015).

The model predicts that ChiLS, once tethered to the enhanceosome core complex, recruits Pygo via NPF to prime the enhancer for Wnt responses (see Model of the Wnt enhanceosome). Given the dimer-tetramer architecture of ChiLS, its binding to Pygo can occur simultaneously to its NPF-dependent binding to RUNX. In turn, tethering Pygo to the Wnt enhanceosome may require Pygo's binding to methylated histone H3 tail, similarly to Groucho/TLE whose tethering to enhancers depends on binding to hypoacetylated histone H3 and H4 tails. Interestingly, Pygo's histone binding requires at least one methyl group at K4-the hallmark of poised enhancers. Indeed, Drosophila Pygo is highly unorthodox due to an architectural change in its histone-binding surface that allows it to recognize asymmetrically di-methylated arginine 2-a hallmark of silent chromatin. Thus, the rare unorthodox Pygo proteins may recognize silent enhancers even earlier, long before their activation, consistent with the early embryonic function of Pygo, prior to Wg signaling (Fiedler, 2015).

Overcoming the OFF state imposed on the enhancer by Groucho/TLE involves Pygo-dependent capturing of β-catenin/Armadillo, which recruits various transcriptional co-activators to its C-terminus. Although these include CREB-binding protein (CBP), a histone acetyl transferase, its tethering to TCF enhancers is likely to co-depend on CRE-binding factors (CREB, c-Fos) and SMAD which synergize with Armadillo to activate these enhancers-similarly to the interferon-β enhanceosome where CBP recruitment also co-depends on multiple enhancer-binding proteins (Panne, 2007). The ensuing acetylation of the Wnt enhancer chromatin could promote the eviction of Groucho/TLE whose chromatin anchoring is blocked by acetylation of histone H3 and H4 tails, thus initiating the ON state (Fiedler, 2015).

Osa antagonizes Wg responses throughout development, and represses UbxB through its CRE, which also mediates repression in response to high Wg signaling. Osa could therefore terminate enhancer activity, by displacing HAT-recruiting enhancer-binding proteins such as CREB and c-Fos from CREs and by cooperating with repressive enhancer-binding proteins such as Brinker (a Groucho-recruiting repressor that displaces SMAD from UbxB) to re-recruit Groucho/TLE to the enhancer, thereby re-establishing its OFF state. Notably, Osa binds Chip, to repress various Wg and ChiLS targets including achaete-scute and dLMO (Fiedler, 2015).

Therefore, ChiLS is not only a coincidence detector of multiple enhancer-binding proteins and NPF proteins, but also a switch module that exchanges positively- and negatively-acting enhancer-binding proteins (through LID) and NPF factors, to confer signal-induced activation, or re-repression. Its stoichiometry and modularity renders it ideally suited to both tasks. It is noted that the interferon-β enhanceosome does not contain a similar integrating module, perhaps because it is dedicated to a single signaling input (Fiedler, 2015).

ChiLS is essential for activation of master-regulatory genes in the early embryo, similarly to DNA-binding pioneer factors such as Zelda (in the Drosophila embryo) or FoxA (in the mammalian endoderm) which render enhancers accessible to enhancer-binding proteins. Moreover, ChiLS maintains HOX gene expression throughout development, enabling Wg to sustain HOX autoregulation, a mechanism commonly observed to ensure coordinate expression of HOX genes in groups of cells (Fiedler, 2015).

Another hallmark of pioneer factors is that they initiate communication with the basal transcription machinery associated with the promoter. Chip is thought to facilitate enhancer-promoter communication, possibly by bridging enhancers and promoters through self-association. Indeed, Ldb1 occupies both remote enhancers and transcription start sites (e.g., of globin genes and c-Myb), likely looping enhancers to the basal transcription machinery at promoters which requires self-association, but possibly also other factors (such as cohesin, or mediator) (Fiedler, 2015).

It is noted that the chromatin association of Ldb1 has typically been studied in erythroid progenitors or differentiated erythroid cells, following activation of erythoid-specific genes. It would be interesting (if technically challenging) to examine primitive cells, to determine whether ChiLS is associated exclusively with poised enhancers prior to cell specification or signal responses (Fiedler, 2015).

Previous genetic analysis in Drosophila has linked chip predominantly to Notch-regulated processes. Likewise, groucho was initially thought to be dedicated to repression downstream of Notch, before its role in antagonizing TCF and Wnt responses emerged. Moreover, Lozenge facilitates Notch responses in the developing eye, and in hematocytes. Indeed, the first links between Groucho/TLE, RUNX and nuclear Wnt components came from physical interactions, as in the case of ChiLS. The current work indicates that these nuclear Notch signaling components constitute the Wnt enhanceosome. Although the most compelling evidence for this notion is based on physical interactions, the genetic evidence from Drosophila is consistent with a role of ChiLS in Wg responses (Bronstein, 2010). Indeed, mouse Ldb1 has been implicated in Wnt-related processes, based on phenotypic analysis of Ldb1 knock-out embryos and tissues. Notably, Ldb1 has wide-spread roles in various murine stem cell compartments that are controlled by Wnt signaling (Fiedler, 2015).

An interesting corollary is that the Wnt enhanceosome may be switchable to Notch-responsive, by NPF factor exchange and/or LMO-mediated enhancer-binding protein exchange at ChiLS. Hairy/Enhancer-of-split (HES) repressors could be pivotal for this switch: these bHLH factors are universally induced by Notch signaling, and they bind to ChiLS enhancers to re-recruit Groucho/TLE via their WRPW motifs. HES repressors may thus be capable of re-establishing the OFF state on Wnt enhancers in response to Notch (Fiedler, 2015).

Notably, restoring a high histone-binding affinity in Drosophila Pygo by reversing the architectural change in its histone-binding surface towards human renders it hyperactive towards both Wg and Notch targets even though pygo is not normally required for Notch responses in flies. Humanized Pygo may thus resist the Notch-mediated shut-down of the Wnt enhanceosome, owing to its elevated histone affinity that boosts its enhancer tethering, which could delay its eviction from the enhanceosome by repressive NPF factors. The apparent Notch-responsiveness of the Wnt enhanceosome supports the notion that orthodox Pygo proteins (as found in most animals and humans) confer both Wnt and Notch responses (Fiedler, 2015).

Previous genetic studies have shown that the components of the Wnt enhanceosome (e.g., TCF, RUNX, ChiLS and LHX) have pivotal roles in stem cell compartments, as already mentioned, suggesting a universal function of this enhanceosome in stem cells. It is therefore hardly surprising that its dysregulation, that is, by hyperactive β-catenin, is a root cause of cancer, most notably colorectal cancer but also other epithelial cancers. Indeed, genetic evidence implicates almost every one of its components (as inferred from the fly counterparts) in cancer: AML1 and RUNX3 are tumour suppressors whose inactivation is prevalent in myeloid and lymphocytic leukemias, and in a wide range of solid tumors including colorectal cancer, respectively. Likewise, ARID1A is a wide-spread tumor suppressor frequently inactivated in various epithelial cancers. Furthermore, many T-cell acute leukemias can be attributed to inappropriate expression of LMO2. Intriguingly, AML1 and ARID1A behave as haplo-insufficient tumor suppressors, consistent with the notion that these factors compete with activating NPF factors such as Pygo2, RUNX2 and possibly MACC1 (predictive of metastatic colorectal cancer) for binding to ChiLS, which will be interesting to test in future. The case is compelling that the functional integrity of the Wnt enhanceosome is crucial for the avoidance of cancer (Fiedler, 2015).

Home page: The Interactive Fly © 1997 Thomas B. Brody, Ph.D.

The Interactive Fly resides on the
Society for Developmental Biology's Web server.