REGULATION (part 2/3)

Transcriptional regulation

An investigation has been carried out of how Drosophila imaginal disc cells establish and maintain their appendage-specific determined states. Ectopic expression of wingless (wg) induces leg disc cells to activate expression of the wing marker Vestigial (Vg) and to transdetermine to wing cells. Ectopic wg expression non-cell-autonomously induces Vg expression in leg discs; activated Armadillo, a cytosolic transducer of the Wg signal, cell-autonomously induces Vg expression in leg discs, indicating that this Vg expression is directly activated by Wg signaling. Ubiquitous expression of wg in leg discs can induce only dorsal leg disc cells to express Vg and transdetermine to wing. Dorsal leg disc cells normally express high levels of decapentaplegic (dpp) and its downstream target, optomotor-blind (omb). Ectopic omb expression is sufficient to dorsalize leg cells but is not sufficient to induce transdetermination to wing. Dorsalization of ventral leg disc cells, through targeted expression of either dpp or omb, is sufficient to allow wg to induce Vg expression and wing fate. Leg cells dorsalized by omb are competent to transdetermine to wing, as shown when wingless is expressed ubiquitously. Under these circumstances Vg is expressed in both dorsal and ventral regions. A non-autonomous effect of omb was observed on wg-induced Vg expression, suggesting that omb induces the expression of another signal that acts with wg to induce Vg expression (Maves, 1998).

High levels of dpp expression, which are both necessary and sufficient for dorsal leg development, are required for wg-induced transdetermination. Thus, dpp and omb promote both dorsal leg cell fate as well as transdetermination-competent leg disc cells. In leg discs, antagonist interactions between Wg and Dpp normally prevent wg expression from overlapping with high levels of dpp expresssion and with omb expression. It is thought that the interaction between Wg and Dpp in transdetermination mimics the interaction between Wg and Dpp normally used to establish the wing disc primordium. It is suggested that Wg signaling directly activates the vg boundary enhancer during wing disc development, presumably in conjunction with Notch signaling through Suppressor of Hairless. Taken together, these results show that the Wg and Dpp signaling pathways cooperate to induce Vg expression and leg-to-wing transdetermination. A specific vg regulatory element, the vg boundary enhancer, is required for transdetermination. It is proposed that an interaction between Wg and Dpp signaling can explain why leg disc cells transdetermine to wing and that these results have implications for normal leg and wing development (Maves, 1998).

Dpp, a TGFbeta, organizes pattern in the Drosophila wing by acting as a graded morphogen, activating different targets above distinct threshold concentrations. Like other TGFbetas, Dpp appears to induce transcription directly via activation of Mad. However, Dpp can also control gene expression indirectly by downregulating the expression of the brinker gene, which encodes a putative transcription factor that functions to repress Dpp targets. The medial-to-lateral Dpp gradient along the anterior-posterior axis is complemented by a lateral-to-medial gradient of Brinker, and the presence of these two opposing gradients may function to allow cells to detect small differences in Dpp concentration and respond by activating different target genes (Campbell, 1999).

Dpp controls patterning along the A-P axis in the wing of Drosophila by activating a number of downstream targets, including sal, omb, and vg. These targets are activated cell autonomously by Dpp signaling, and there is evidence, at least for vg, that Dpp induces gene transcription directly through activation of the SMAD Mad, which may act as a transcription factor. Expression of these three targets is also regulated negatively by brk: loss-of-function mutations in brk lead to ectopic and inappropriate levels of expression of sal, omb, and vg. Wing discs from the pupal lethal hypomorph brkXA are greatly enlarged along the A-P axis, phenocopying the ubiquitous expression of Dpp. These discs show expanded domains of sal, omb, and vg expression in the expanded wing pouch. Null mutants, including brkXA, are embryonic lethal, but mutant clones can result in outgrowths in adult wings when the clone is located in the anterior or posterior extremes of the wing. These outgrowths are comprised entirely of mutant tissue but are similar to outgrowths produced by misexpression of Dpp. Examination of such clones in wing discs reveals autonomous activation of sal, omb, and vg outside of their normal expression domains. Thus, Brk functions in the developing wing to repress the expression of Dpp targets such as sal, omb, and vg (Campbell, 1999).

Why is the indirect method involving Brk used to activate expression of Dpp targets? In other words, if Dpp can directly activate these genes via Mad, then why is this not sufficient? It is speculated that it is directly related to Dpp acting as a morphogen. Activation of sal, omb, and vg is not simply all or none, but each is induced above a distinct threshold concentration of Dpp, with sal requiring the highest level and vg the lowest. The gradient of Dpp will be transduced into a gradient of activated Mad, but it is possible that cells cannot perceive small differences in activated Mad reliably enough to faithfully define the expression domains of Dpp targets and that the introduction of the Brk intermediary provides the necessary information. This type of dual control of gene expression may turn out to be a common feature of many morphogen systems. The possibility is raised that other TGFßs may also use indirect mechanisms to control expression of target genes, possibly even Brk-related proteins, especially if they also induce multiple targets in a concentration-dependent manner. One relevant observation in this regard is that brk is also expressed in the early embryo where its expression also appears to be regulated by Dpp; null mutant embryos are partially dorsalized, suggesting it has a similar function here as in the wing. Unlike the wing, control of D-V patterning by TGFßs is probably a conserved feature of almost all animal embryos and strengthens the possibility that brk homologs will be typical regulators of TGFß target genes (Campbell, 1999).

The role of the zinc finger transcription factor Schnurri (Shn) in mediating the nuclear response to Dpp during adult patterning has been investigated. Using clonal analysis, it has been shown that wing imaginal disc cells mutant for shn fail to transcribe the genes spalt, optomotor blind, vestigial, and Dad, that are known to be induced by dpp signaling. shn clones also ectopically express brinker, a gene that is downregulated in response to dpp, thus implicating Shn in both activation and repression of Dpp target genes. Loss of shn activity affects anterior-posterior patterning and cell proliferation in the wing blade, in a manner that reflects the graded requirement for Dpp in these processes. Furthermore, shn is expressed in the pupal wing and plays a distinct role in mediating dpp-dependent vein differentiation at this stage. The absence of shn activity results in defects that are similar in nature and severity to those caused by elimination of Mad, suggesting that Shn has an essential role in dpp signal transduction in the developing wing. These data are consistent with a model in which Shn acts as a cofactor for Mad (Torres-Vazquez, 2000).

A study by S. D. Weatherbee (1998) is arguably the best study yet published about how gene regulation differs in homologous structures, and points to future studies for how differential gene regulation will be shown to account for the structural differences between species. The differentiation of the Drosophila haltere from the wing through the action of the Ultrabithorax (Ubx) gene is a classic example of Hox regulation of serial homology. This study reveals several features of the control of haltere development by Ubx which, in principle, are likely to apply to the Hox-regulated differential development of other serially homologous structures in other animals. In the proximodistal axis, Ubx selectively represses one enhancer of the vestigial gene. vg is expressed and required in the cells that will give rise to the distal appendage fields of the wing and haltere imaginal discs. vg expression in the wing field is regulated by two distinct enhancers that are activated by different signaling pathways. vg expression is first activated along the DV boundary of the wing disc by the Notch pathway, through the boundary enhancer; later it is activated in the growing wing pouch by the Dpp and Wg signals, through the "quadrant" enhancer. Similarly, the boundary enhancer is activated in both the wing and haltere discs, however, the quadrant enhancer is silent in the haltere field. The repression of the quadrant enhancer in the haltere is sensitive to the dosage of Ubx and is partially derepressed in Ubx/+ haltere discs. More importantly, in Ubx clones in the haltere disc, the quadrant enhancer is fully activated. These results show that Ubx selectively represses a portion of the native vg wing expression pattern in the haltere disc through the quadrant enhancer (Weatherbee, 1998).

The effects of ectopic expression of the vestigial gene was examined in the haltere and other tissues under the control of the GAL4/UAS system. Whereas vg expression in all other appendages and tissues causes wing-like outgrowths, in the haltere no significant change in adult appendage size or morphology was observed. However, striking differences between the morphology of the outgrowths formed on the second and third thoracic legs were observed. The former had clear wing-like morphology, whereas the latter had haltere-like morphology. The failure of ectopic vg expression to significantly alter haltere morphology and the distinct haltere-like character of the outgrowths formed in third thoracic legs suggests that Ubx acts on genes that are downstream of or parallel to vg in the genetic hierarchy. To test whether Ubx regulates genes downstream of vg, a search was carried out for candidate genes whose expression depends on Vg. The spalt and DSRF genes that are normally not expressed in leg imaginal discs are ectopically induced in first and second thoracic leg imaginal discs as a response to targeted expression of Vg, and may thus be activated in the developing wing through some mechanism that is dependent on Vg. The patterns of ectopic induction of sal and DSRF in T1 and T2 leg discs are reminiscent of their normal expression patterns in wing discs. In contrast to T3 leg discs, which also express Ubx, the central domains of ectopic induction of Sal and DSRF expression are suppressed. These results demonstrate that downstream targets of Vg are also regulated by Ubx, independent of the Ubx regulation of Vg itself. The repression of these and other targets by Ubx would then suggest why the deregulation of Vg expression in the developing haltere is insufficient to reprogram haltere development toward wing development and to alter the morphology of the adult haltere. Similarly, ectopic expression of the DSRF or Sal (M. Averof, personal communication to Weatherbee, 1998) transcription factors also does not alter haltere size, shape, or cell morphology. These results imply that there are genes downstream of DSRF and Sal whose expressions are necessary for the realization of a phenotype but which are repressed by Ubx in the haltere disc (Weatherbee, 1998)

During Drosophila wing development growth and patterning are mediated by signaling from the dorsoventral (D/V) organizer. In the metathorax, wing development is essentially suppressed by the homeotic selector gene Ubx to mediate development of a pair of tiny balancing organs, the halteres. Expression of Ubx in the haltere D/V boundary down-regulates the haltere's D/V organizer signaling compared to that of the wing D/V boundary. Somatic loss of Ubx from the haltere D/V boundary thus results in the formation of a wing-type D/V organizer in the haltere field. Long-distance signaling from this organizer was analyzed by assaying the ability of a Ubx minus clone induced in the haltere D/V boundary to effect homeotic transformation of capitellum cells (the capitellum is the main body of the haltere) away from the boundary. The clonally restored wing D/V organizer in mosaic halteres not only enhances the homeotic transformation of Ubx minus cells in the capitellum but also causes homeotic transformation of even Ubx plus cells in a genetic background known to induce excessive cell proliferation in the imaginal discs. In addition to demonstrating a non-cell-autonomous role for Ubx during haltere development, these results reveal distinct spatial roles for Ubx during maintenance of cell fate and patterning in the halteres. Ubx modulates the expression of wingless and cut in the haltere D/V boundary and represses vestigial in the capitellum, thereby suggesting a mechanism for the Ubx mediated down-regulation of the D/V organizer activity in the haltere. While the repression of wingless and cut expression is cell-autonomous, that of the quadrant vestigial-lacZ is non cell-autonomous: pouch cells farther away from the D/V boundary show more severe reduction in lacZ expression. Given the fact that quadrant vestigial-lacZ repression is dependent on the formation of the D/V boundary, the non-cell-autonomy in quadrant vestigial-lacZ repression by ectopic Ubx would not be surprising if Ubx function is to negatively regulate D/V signaling. Thus all the results provide strong evidence for the negative regulation of D/V organizer signaling by Ubx during haltere specification. It is likely that during haltere development, repression of wing patterning signals results in the specification of cell shape and volume that are unique to the haltere (Shashidhara, 1999).

The Notch receptor mediates cell interactions controlling the developmental fate of a broad spectrum of undifferentiated cells. By modulating Notch signaling in specific precursor cells during Drosophila imaginal disc development, it has been demonstrated that Notch activity can influence cell proliferation. The activation of the Notch receptor in the wing disc induces the expression of the wing margin patterning genes vestigial and wingless, and strong mitotic activity. However, the effect of Notch signaling on cell proliferation is not the simple consequence of the upregulation of either vestigial or wingless. On the contrary, Vestigial and Wingless display synergistic effects with Notch signaling, resulting in the stimulation of cell proliferation in imaginal discs (Go, 1998).

To explore the consequences of Notch signaling modulation during Drosophila development, the UAS-GAL4 system was used. Loss-of-function phenotypes were elicited through the expression of either a truncated, dominant negative form of the Notch receptor (d.n.N) lacking the intracellular domain, or the Hairless (H) protein, a negative regulator of Notch signaling. Gain-of-function phenotypes were induced by expressing a constitutively activated form of the Notch receptor (act.N) (Go, 1998).

To examine the link between the H misexpression phenotypes and Su(H)-dependent Notch activity, transgenic animals were generated carrying a lacZ reporter construct driven by the fusion between multimerized Su(H)-binding sites and an E(spl)m g promoter, a known Su(H) target. This construct consists almost exclusively of engineered Su(H)-binding sites. In a cell culture based reporter, the expression from the reporter construct is induced by the simultaneous expression of Su(H) and act.N, while the expression of any one construct alone fails to induce transcription. Strong lacZ expression is detected in the posterior part of the eye disc of late third instar transgenic larva. This expression is effectively suppressed by misexpression of H using the GAL4 line T113 and results in small eye discs, indicating that H overexpression can suppress Su(H)-dependent Notch signaling in vivo. The size of the eye is significantly affected and, in extreme cases, the eye is missing. In addition to small eyes, small wings and halteres are observed as well as more typical Notch loss-of-function phenotypes, such as extra thoracic bristles. The 'small eye' phenotype induced by H expression is not associated with severe eye roughness. This 'small eye' phenotype, together with the wing and haltere abnormalities, is reminiscent of Serrate loss-of-function mutations. To further explore the possibility that the observed eye phenotype reflects Ser-dependent Notch signaling, the genetic interactions were examined with Beaded of Goldschmidt (BdG), a dominant negative mutation of Ser known to affect wing margin development. In combination with BdG, strong synergistic effects are observed displaying phenotypes characteristic of Ser, such as small eyes, wings and halteres. Therefore, H misexpression can mimic Ser loss-of-function mutations, raising the possibility that Ser/Notch signaling may control eye morphogenesis (Go, 1998).

To further investigate the role Notch signaling plays in morphogenesis, the H and d.n.N transgenes were expressed at the d/v compartment boundary of the wing disc using the vestigial-GAL4 driver. Misexpression of either H or d.n.N results in similar phenotypes, which range from wing margin notches to rudimentary wings. The effect of H misexpression can be suppressed by expressing act.N and vice versa. For example, the lethality associated with misexpression of act.N is suppressed by simultaneous expression of H. Conversely, the phenotypes elicited by H misexpression are largely suppressed by act.N. This mutual suppression is observed with other GAL4 lines as well. Given that the actions of act.N and H seem to be manifested through Su(H), it is likely that the mutual suppression of act.N and H is also mediated by Su(H). It is noteworthy that, even though both H and d.n.N act as antagonists of Notch signaling and the phenotypes associated with their expression are similar, their interactions with act.N are different. While act.N is an effective suppressor of the phenotypes induced by H misexpression, it fails to suppress the effects of d.n.N (Go, 1998).

The relationship between Notch signaling and the expression of vg and wg was examined, since the induction of both genes is considered to be essential for wing morphogenesis. When either d.n.N or H is misexpressed along the anterior/posterior (a/p) boundary using the ptc-GAL4 line, expression from the vg d/v boundary enhancer, as well as the wg enhancer, is effectively repressed near the intersection between the a/p and d/v boundaries. In contrast, the vg quadrant enhancer, which is normally silent at the intersection between a/p and d/v boundaries, is induced by the identical constructs. Essentially the opposite effect is observed when act.N is misexpressed, demonstrating that Notch signaling has opposite effects on two distinct enhancers of vg (Go, 1998).

The wing phenotypes elicited by misexpression of act.N are similar to those induced by Abruptex (Ax) mutations, which are Notch gain-of-function alleles associated with point mutations in the extracellular domain of the protein. A heteroallelic Ax combination results in the activation of Notch signaling and the expression of Notch downstream genes is induced. For instance, ectopic wg expression is found around the d/v boundary. Induction of the vg d/v boundary enhancer and repression of the vg quadrant enhancer around the d/v boundary are found, similar to the effect of expression of act.N. Activation of Notch signaling around the d/v boundary of the wing disc through either misexpression of act.N or the Ax mutations results in a substantial enlargement of the disc. BrdU incorporation experiments indicate that these phenotypes are associated with an elevated mitotic activity. BrdU incorporation is stimulated and is particularly obvious in the peripheral region of the wing pouch, suggesting that the periphery Is more responsive than other regions. Misexpression of act.N in other parts of the wing disc also results in the stimulation of mitotic activity. When act.N is expressed in the wing pouch, the disc grows in such a way that the characteristic folded structures of the wing pouch are pushed to the periphery. Conversely, the same structures are 'pushed' toward the d/v boundary when act.N is expressed in the periphery. When act.N is misexpressed in a discrete pattern in the periphery using the GAL4 line 766, a regional correspondence is observed between Notch signaling activation and high mitotic activity, demonstrating a local effect of Notch activity on cell proliferation in the periphery. However, as is particularly evident when the Notch receptor is activated around the d/v boundary, the region of Notch signaling activation does not coincide with the region of the highest mitotic activity. It is therefore concluded that the effect of Notch signaling on cell proliferation must be indirect (Go, 1998).

The effect of Notch activity on cell proliferation is not the simple consequence of vg induction Since vg is a direct target of Su(H)-dependent Notch signaling, it is possible that the mitogenic effect of Notch is mediated by the upregulation of vg. In this case, misexpression of Vg would be expected to result in phenotypes similar to those elicited by act.N. Misexpression of act.N in the dorsal side of the wing pouch, using the GAL4 line A9, induces expression from the vg d/v boundary enhancer as well as the wg enhancer The dorsal side of the wing pouch region appears enlarged. In contrast, when Vg is misexpressed in the same region, the dorsal side of the wing pouch becomes much smaller than the ventral side, while wg expression in the periphery of the dorsal side was suppressed. The loss of dorsal wing pouch induced by Vg misexpression is significantly rescued by expressing Wg simultaneously. This is consistent with the notion that the observed phenotype caused by misexpression of Vg is due to the repression of wg, whose expression in the wing pouch is more uniform at earlier stages. Misexpression of Wg alone in the dorsal side, unlike the misexpression of act.N, does not have a significant effect on cell proliferation in the wing pouch. These results indicate that the effect of act.N expression on mitosis is separable from vg induction. In addition, they indicate that Vg is capable of repressing wg expression in the wing pouch, but not at the d/v boundary (Go, 1998).

Misexpression of Vg compared to act.N has opposite effects in the wing disc. Thus, Vg misexpression in the wing disc induces wg downregulation and small discs. In contrast, misexpression of Vg in the eye discs upregulates wg and results in a clear enlargement of the discs, demonstrating that Vg can either repress or induce wg expression in a context-dependent manner. The observed context-dependent effect of Vg on wg expression raises the possibility that Notch signaling may be capable of modulating the way Vg affects wg expression. This is of particular interest in view of the possibility that Vg does not suppress wg expression at the d/v boundary because of the existing high level of Notch signaling activity. In fact, the simultaneous expression of act.N and Vg reveals a striking synergistic effect on cell proliferation. The most notable effects are in the eye discs, where tissue expressing the two proteins shows striking overgrowth associated with strong wg induction. The other discs are also clearly affected, displaying cellular overgrowth, but the effects are far less dramatic than the eye discs. This overgrowth phenotype is also evident when act.N and Vg misexpression are driven by dpp-GAL4, even though the synergistic effect is less dramatic. In contrast, the effect of misexpression of Vg with dpp-GAL4 on wg induction and cell proliferation in the eye discs is, in some cases, significantly suppressed by simultaneous expression of H. These experiments demonstrate that the proliferative potential of certain tissues can be modulated by the synergistic action of Notch with other genes. Moreover, they identify Notch signaling as an important factor in the way Vg affects wg expression and cell proliferation at the d/v boundary during wing morphogenesis (Go, 1998).

Several different mechanisms have been proposed to account for the activation of Su(H) by Notch. To further investigate how Su(H) activity is regulated, misexpression assays were used with wild-type Su(H) and with modified forms of Su(H) containing either a nuclear localization signal [Su(H)NLS], a transcriptional activation domain [Su(H)VP16], or a deletion of the domain required for interaction with the antagonist Hairless [Su(H)DH]. Only Su(H)VP16 is able to mimic Notch activation effectively in the Drosophila wing, in agreement with the model that Notch activity normally confers coactivator function on Su(H). Neither nuclear localization nor elimination of Hairless binding is sufficient for activation. The phenotypes produced by overexpression of Su(H)wt and Su(H)NLS indicate a mixture of both increased and reduced Notch pathway activity and point to a role for Su(H) in both activation and repression of gene expression, as has been proposed for the mammalian homolog CBF1. Some phenotypes are equivalent to Notch loss-of-function, with wing-nicks and inhibition of a subset of target genes, which is most consistent with the ectopic proteins displacing a Su(H)-coactivator complex. Conversely, other phenotypes are equivalent to Notch gain-of-function, with wing-overgrowths and ectopic target-gene expression. These effects can be explained by the ectopic Su(H)/Su(H)NLS titrating a repressor complex. The wing-overgrowth phenotype is sensitive to the dose of Hairless and the phenotypes produced by coexpressing Su(H) and Hairless suggest that Hairless could form a component of this repressive complex (Furriols, 2000).

The phenotypes produced by misexpressing Su(H)wt and Su(H)NLS appear to combine activation and repression of Notch activity. To confirm whether this interpretation is correct, the effects on genes whose expression at the dorsal/ventral boundary is dependent on Notch were analyzed. Two assays used fragments that are directly responsive to Su(H) and Nicd: mbeta1.5 (a fusion between the E(spl)mbeta regulatory sequences and lacZ) and vgBE-lacZ (a fusion between the vestigial boundary enhancer and lacZ). Expression of wingless and the entire vestigial gene, which may involve indirect as well as direct regulation by the Notch pathway, were also examined. All are ectopically activated by Su(H)VP16, consistent with it mimicking the effects of Nicd. As anticipated, the wing-nick phenotypes produced by the modified Su(H) proteins in combination with ptc-Gal4 correlate with a reduction in the levels of mbeta1.5 and wingless expression at the d/v boundary. mbeta1.5 appears to be more sensitive and is strongly repressed by all three proteins, with Su(H)NLS the most effective as suggested by the wing-nicking phenotypes. The effects on wingless were milder and only Su-(H)NLS strongly represses expression. In contrast, vgBE-lacZ shows a very different response and is ectopically activated by both Su(H)wt and Su-(H)NLS. The activation of vgBE-lacZ is also observed when the levels of misexpressed proteins are lower (using dpp-Gal4) although in this combination Su(H)NLS and Su(H)wt still repress mbeta1.5. The two enhancers therefore appear intrinsically different in the way they respond to Su(H), suggesting that their regulation may involve different thresholds of activating and repressing Su(H) containing complexes (Furriols, 2000).

The mixed loss and gain of Notch function phenotypes produced in the wing by ectopic expression of Su(H)wt and Su(H)NLS suggests that Su(H) has a dual function, acting in some contexts as an activator and in others as a repressor. The simplest model is that Su(H) can exist in at least two complexes -- one where it interacts with a coactivator(s) and the other where it interacts with a corepressor(s). In cells where there is no/low Notch signaling the primary function of Su(H) would be to keep the target genes repressed by interacting with a corepressor complex. One of these corepressors could be Hairless; another could be the HDAC complex described in vertebrates. In contrast, in the cells where Notch is active, Su(H) would be complexed with coactivator(s) (e.g., Nicd so that the transcription of Notch target genes would be initiated). Depending on the relative levels or activity of the components, the equilibrium would shift in favor of one or the other complex (Furriols, 2000).

Cell interactions mediated by Notch-family receptors have been implicated in the specification of tissue boundaries in vertebrate and insect development. Although Notch ligands are often widely expressed, tightly localized activation of Notch is critical for the formation of sharp boundaries. Evidence is presented that the POU domain protein Nubbin contributes to the formation of a sharp dorsoventral (DV) boundary in the Drosophila wing. Nubbin represses Notch-dependent target genes and sets a threshold for Notch activity that defines the spatial domain of boundary-specific gene expression (Neumann, 1998).

Certain features of the abnormal wings in flies mutant for nubbin suggest a possible role for Nubbin protein in spatially limiting Notch activity at the DV boundary of the wing. The row of sensory bristles that makes up the wing margin is disorganized in nubbin wing mutants, suggesting a defect in Wingless or Notch activity. In preparations where the wing margin is viewed edge on, this disorganization reflects a broadening of the region where bristles form. Margin bristles are normally specified in cells very close to the DV boundary, reflecting a requirement for high levels of Wingless signaling activity. The broadening of the margin suggests that Wingless might be ectopically expressed in nubbin mutant wing discs. Wingless is normally expressed in a stripe of two to three cells straddling the DV boundary. In nubbin mutant discs, this stripe is widened considerably. Expression of the Notch targets vestigial and cut are similarly expanded at the DV boundary in nubbin mutants (Neumann, 1998).

To determine whether the effect on bristle specification is a direct consequence of removing nubbin activity, clones of nubbin mutant cells were generated in a wild-type background. Ectopic wing margin bristles are found in nubbin mutant clones located near the endogenous wing margins. The nubbin mutant clones show ectopic expression of neuralized, a molecular marker for precursors of the sensory neurons that innervate the bristles. The nubbin mutant clones misexpress wingless and vestigial. The largely autonomous effect of nubbin mutant clones on bristle specification may be due to the relatively low levels of Wg protein expressed in nubbin mutant clones. Together with the results on cut expressioon, these observations suggest that Notch target genes are transcriptionally up-regulated in nubbin mutant cells near the DV boundary (Neumann, 1998).

To test whether ectopic activation of these genes in nubbin mutant clones directly depends on Notch signaling activity, clones of cells were generated that were simultaneously mutant for nubbin and Suppressor of Hairless [Su(H)]. Su(H) encodes a DNA binding protein that mediates transcriptional activity of Notch target genes. Su(H) is autonomously required for the expression of wingless, vestigial, and cut at the DV boundary and binds directly to the vestigial DV boundary enhancer. Clones of cells mutant for both nubbin and Su(H) do not ectopically activate wingless, demonstrating that ectopic expression of wingless in nubbin mutant cells depends on activity of the Notch pathway. To confirm that Nubbin acts downstream of Notch, a test was performed to see whether overexpression of Nubbin could suppress the effects of a ligand-independent form of Notch. When Nubbin is coexpressed with such a constitutively active Notch, ectopic Wingless expression is strongly reduced. Together, these observations suggest that Nubbin may act as a direct repressor of Notch-dependent target gene expression. These findings argue that the effects of Nubbin are unlikely to be mediated by indirect effects on the expression of Notch ligands (Neumann, 1998).

Two different activities of Suppressor of Hairless during wing development in Drosophila

The Notch pathway plays a crucial and universal role in the assignation of cell fates during development. In Drosophila, Notch is a transmembrane protein that acts as a receptor of two ligands, Serrate and Delta. The current model of Notch signal transduction proposes that Notch is activated upon binding its ligands and that this leads to the cleavage and release of its intracellular domain (also called Nintra). Nintra translocates to the nucleus where it forms a dimeric transcription activator with the Su(H) protein. In contrast with this activation model, experiments with the vertebrate homolog of Su(H), CBF1, suggest that, in vertebrates, Nintra converts CBF1 from a repressor into an activator. The role of Su(H) in Notch signaling during the development of the wing of Drosophila has been assessed. The results show that, during this process, Su(H) can activate the expression of some Notch target genes and that it can do so without the activation of the Notch pathway or the presence of Nintra. In contrast, the activation of other Notch target genes requires both Su(H) and Nintra, and, in the absence of Nintra, Su(H) acts as a repressor. The Hairless protein interacts with Notch signaling during wing development and inhibits the activity of Su(H). These results suggest that, in Drosophila, the activation of Su(H) by Notch involves the release of Su(H) from an inhibitory complex, which contains the Hairless protein. After its release Su(H) can activate gene expression in the absence of Nintra (Klein, 2000).

The loss of H function seems to elicit Su(H)-dependent target gene expression in the wing pouch, a region probably devoid of Notch activity. This suggests that the inactivation of H is sufficient to activate Su(H). To test further this conclusion, an examination was performed to see whether the activity of the vgBE is maintained in H mutant wing pouches if Notch is concomitantly removed. For this, Notch mutant clones were induced in H mutant wing discs. In H mutant wing pouches, weak ubiquitous expression of the vgBE is observed throughout the whole area of the wing, confirming the clonal analysis. vgBE is also active in several Notch mutant clones near the DV and anteroposterior (AP) boundary, but the activity is not maintained in all clones. One explanation for this might be again the requirement of other so far unidentified factors emanating from the two compartment boundaries. In agreement with this, the vgBE enhancer has a late expression domain along the AP boundary, suggesting an input from these areas for its proper expression. However this domain is also dependent on Notch during normal development. The removal of the Su(H)-binding site in the enhancer leads to the loss of all expression domains in the wing pouch, suggesting that Su(H) is required (Klein, 2000).

Therefore, the fact that the cells of several mutant clones do express the vgBE suggest that the vgBE can be activated in the complete absence of Notch activity and that the inactivation of H is sufficient to activate Su(H). No activation of the vgBE was ever found in Notch mutant clones induced in wild-type wing pouches, suggesting that during wild-type development, the activity of Notch is required to activate the vgBE. Hence, Notch probably activates Su(H) through inactivation of H. An examination was performed to see whether the degree of endogenous Su(H) activation that results from the removal of H is sufficient to elicit a biological effect. To assay this, it was asked whether or not removal of H activity can induce Su(H)-dependent development of the pouch in wing discs in which Notch signaling is absent, such as apterous and Presenilin mutant wing discs. Loss of H function rescues the loss of wing development of ap mutants: whereas ap mutants have no wing pouch, ap;H double mutants have large wing pouches with no margin structures. The enlarged pouch of the double mutant discs expresses spalt (sal) and the two vg reporters, vgQE and vgBE, all of which are expressed specifically in the wing pouch in a Notch/Su(H)-dependent manner and are not expressed in ap mutants. In contrast, no wg expression is induced in these double mutant discs, suggesting that the observed rescue is likely to be due to the activation of Su(H) in the double mutants. This is strongly supported by the fact that Su(H);H double mutants exhibit a small wing rudiment identical to that of Su(H) mutants. Expression of UAS-vg by dpp-Gal4 in ap mutant discs can recover the pouch-specific expression domain of sal, suggesting that the activation of vg expression by Su(H) is responsible for the recovered sal expression in the ap;H double mutant wing discs. Similar to overexpression of UAS-Su(H) in ap mutant wing discs, the pouch in ap;H mutant discs develops near the residual wg expression in the remaining hinge. As expected from the analysis of the wing discs, the pharate adult ap;H double mutants have large wing pouches, which are devoid of any margin like structure such as innervated bristles (Klein, 2000).

Overexpression of Su(H) leads to three different responses: (1) activation, as is the case for vg, some E(spl) genes, Dl and Ser; (2) inactivation, as shown for cut and E(spl)m8; or (3) no effect, as is the case for wg. This differential behavior is, at least in some cases, a consequence of direct binding of Su(H) to the promoters: the vgBE as well as the E(spl) genes contain Su(H)-binding sites to which Su(H) binds; such sites are necessary for the activation of these genes in vivo. Despite that, they react differently towards Su(H) overexpression. Since E(spl)m8, which is suppressed by Su(H) overexpression, can be activated by expression of Su(H)VP16 or Nintra, it is concluded that Nintra is required in addition to Su(H) to activate E(spl)m8 expression. The results suggest that, in this case, Nintra probably acts as an activation domain of a dimeric transcription factor containing Su(H), as has been proposed. From this, it follows that Nintra might have two function during a Notch signaling event: first it inactivates H, which leads to the release of Su(H) and then, in some instances, it provides the transactivation domain for free Su(H) to activate the expression of target genes (Klein, 2000).

Involvement of winged eye encoding a chromatin-associated bromo-adjacent homology domain protein in disc specification

How organ identity is determined is a fundamental question in developmental biology. In Drosophila, field-specific selector genes, such as eyeless (ey) for eyes and vestigial (vg) for wings, participate in the determination of imaginal disc-specific identity. Gain-of-function screening was performed and a gene named winged eye (wge) was identified, that encodes a bromo-adjacent homology domain protein that localizes at specific sites on chromosomes in a bromo-adjacent homology domain-dependent manner. Overexpression of wge induces ectopic wings with antero-posterior and dorso-ventral axes in the eye field in a region-specific Hox gene- (Antennapedia) independent manner. Overexpression of wge is sufficient for ectopic expression of vg in eye discs. A context-dependent requirement of wge was demonstrated for vg expression in wing discs and for expression of eyes absent (eya), a control gene for eye development downstream of ey, in eye discs. In contrast to vg, however, overexpression of wge inhibits Eyeless-mediated expression of eya. Consistent with colocalization on polytene chromosomes of Wge and Posterior sex combs (PSC -- a Polycomb group gene product), an antagonistic genetic interaction between wge and Psc was demonstrated. These findings suggest that wge functions in the determination of disc-specific identity, downstream of Hox genes (Katsuyama, 2005).

Artificial activation of Notch signaling induces various ectopic appendages, such as ectopic eyes, antennae, wings, and legs, in the eye field in a context-dependent manner. In this system, the ey enhancer-dependent activation of Notch signaling and gene expression is crucial for transdetermination within a limited time window and for shutting off the forced expression once the transdetermination is induced. To identify the genes capable of changing disc-specific identity, a system with a P element-based GS vector was used. For a pilot experiment, 106 lines harboring the GS vector (GS lines) were crossed with flies carrying the ey enhancer-GAL4 (ey-GAL4) and a construct for the constitutively active Notch receptor under an upstream-activating sequence for GAL4 (UAS-Nact). Ectopic structures were induced in the eye field in combination with Notch signaling activation in five lines. For example, ectopic wings were induced in GS 1068 in which ey-GAL4 drives the expression of PGRP-LE, CG8509, and sd, encoding a cofactor of Vg. In the absence of Notch signaling activation, all five lines had reduced eye phenotypes. Therefore, to increase the screening efficiency, a prescreening step was introduced in which GS lines were crossed with the ey-GAL4 driver, and the resulting lines with reduced eye phenotypes were crossed with the ey-GAL4, UAS-Nact line. To confirm the effects of the prescreening, 74 negative GS lines were crossed with normal eyes in a prescreening with an ey-GAL4, UAS-Nact line. None of the lines had ectopic structures in the eye field, indicating that the prescreening was effective. In the prescreening, 8,486 lines had normal eyes and 1,202 lines had reduced eye phenotypes. In 9 of 9,710 lines, ectopic structures were induced in the eye field, e.g., ectopic wings in the GS 15923, even in the absence of Notch signaling activation. The resulting 729 lines with reduced eye phenotypes or ectopic structures in the eye field in the prescreening and 22 lines that were selected without prescreening were then crossed with the ey-GAL4, UAS-Nact line, and ectopic structures were induced in 45 lines in combination with Notch signaling activation (wing, 3 lines; antenna, 26 lines; leg, 16 lines) (Katsuyama, 2005).

This paper focusses on the GS15923 line, because well organized wings with antero-posterior and dorso-ventral axes were induced in the eye field when GS15923 was crossed with ey-GAL4. Using the inverse PCR method, the insertion site of the GS vector in GS15923 was determined and a predicted gene, CG31151, was found adjacent to the insertion site. CG31151 was the only gene expressed in a GAL4-dependent manner in GS15923. After cloning cDNAs corresponding to CG31151, an ORF of a previously uncharacterized gene was identified that was named winged eye (wge), which has a 345-bp extension at the 5' end from the estimated translation initiation site of expressed sequence tag (EST) clones corresponding to CG31151. The sequence of an EST clone, LP24488, overlapped by 693 bp with the 5' end of the cloned cDNA and 20 bp of the 5' end of the RA transcript, confirming the identified ORF. wge encodes a 1,658-aa protein with two Gln-rich, one Ala-rich, and one Ser-rich domain at the N-terminal half and a bipartite nuclear localization signal and a BAH domain in the C-terminal half, implying that Wge is involved in epigenetic regulation of gene expression. This prediction is consistent with the specific localization of Wge on polytene chromosomes. Ectopic wing induction by forced wge expression was confirmed by using UAS-wge transgenic flies. The Wge-mediated ectopic wings had the costa with spine bristles, the triple row of bristles, and the double row of bristles that are formed at the anterior-proximal part of the wing, at the anterior wing margin, and at the distal wing margin in wild-type wing, respectively, indicating that Wge-mediated ectopic wings are correctly organized along the antero-posterior and dorso-ventral axes. In contrast, Vg-mediated ectopic wings in the eye field are outgrowths with wing hair that do not have any wing margins, suggesting that Wge is involved in the wing formation upstream of Vg (Katsuyama, 2005).

Database analysis revealed a genome sequence of an Anopheles gambiae gene with striking similarity to wge; for example, there was a 77% amino acid identity in the BAH domain. Comparison of the two sequences led to the identification of a previously uncharacterized protein domain named highly corresponding region (HCR: 126 aa, amino acids 1130–1255) with similarity with the A. gambiae gene product (57% amino acid identity), KIAA1447 human protein (41%), CAGL79 human protein (33%), and BC060615 mouse protein (42%). These results suggest that wge is evolutionarily conserved in insects and mammals (Katsuyama, 2005).

It has been demonstrated that, in combination with activation of Notch signaling, expression of Antp by ey-GAL4 induces ectopic wings with antero-posterior and dorso-ventral axes, that are similar to Wge-mediated ectopic wings. To investigate the requirement of Antp for the Wge-mediated induction of ectopic wings, transgenic flies were generated possessing an inverted repeat (IR) expression construct of Antp cDNA that specifically inhibits Antp expression in a GAL4-dependent manner due to RNA interference. Antp-IR expression inhibited the formation of ectopic wings induced by the expression of Antp and Notch signaling activation, indicating that the construct works as a specific inhibitor of Antp expression. Antp-IR expression, however, does not inhibit the formation of ectopic wings when coexpressed with wge. Consistent with this result, forced expression of wge does not induce ectopic expression of Antp in eye discs of ey-GAL4;UAS-wge larvae. These results indicated that wge overexpression induces ectopic wings in an Antp-independent manner. Moreover, Wge-IR expression suppresses the formation of ectopic wings induced by the expression of Antp and Notch signaling activation. These results indicate that Wge acts downstream of Antp and Notch signaling in the ectopic wing induction (Katsuyama, 2005).

Ectopic expression of vg in various imaginal discs induces ectopic wing-like outgrowth. In combination with Wg signaling, however, it induces wings with wing margins. Wg signaling is also suggested to participate in the determination of wing discs. Whether wge overexpression induces ectopic expression of Vg and Wg was investigated in eye imaginal discs when it induces ectopic wings with wing margins. In wild type, Vg is expressed in the wing discs but not in eye discs, whereas, in ey-GAL4;UAS-wge, there was significant expression of Vg in the eye discs. The forced expression of wge also activates vg D/V boundary enhancer-lacZ (vgBE-lacZ). These results indicate that wge overexpression is sufficient to induce the ectopic expression of vg, a control gene for wing formation. In third-instar larvae, Wg was expressed at the peripheral edge of the dorsal and ventral sides of wild-type eye discs. Overexpression of wge by ey-GAL4 enhances the dorsal expression of Wg and suppresses the ventral expression of Wg in eye discs. These results are consistent with the findings that wge overexpression induces ectopic wings at the dorsal part of the eye field. Therefore, Wge is suggested to induce ectopic wings upstream of Vg and Wg (Katsuyama, 2005).

The effects of wge overexpression on the expression of genes involved in the determination of eye identity, such as ey and eya, was investigated. Overexpression of wge was seen to repress Eya expression at the dorsal side of the eye discs, in contrast to Wg expression, which is up-regulated at the dorsal side by Wge. Similar results were obtained with eya-lacZ transgenic larvae. The enhancer trap line of eya-lacZ reflects endogenous expression of Eya. Double staining revealed that the ectopic induction of Vg does not overlap with Eya expression when wge was overexpressed in eye discs. These results suggest that Wge-mediated down-regulation of eya is involved in the ectopic induction of vg. However, it was not possible to examine the effects of eya overexpression on Wge-mediated ectopic vg expression, because eya overexpression in eye discs inhibits eye disc development. Further analysis is required to determine the relationship between the down-regulation of eya and the ectopic induction of vg (Katsuyama, 2005).

Clonal activation of Wg signaling represses eya expression in eye discs. To examine the effects of clonal induction of wge overexpression on eya and wg expression, a cell lineage tracer technique was applied by using a combination of the flp/FRT and GAL4/UAS recombinase systems. In this system, in which wge-overexpressing cells are labeled with GFP, eya-lacZ expression is repressed by clonal induction of wge overexpression in a cell-autonomous manner, whereas Wg expression is not induced when the wge-expressing clone is induced, even on the dorsal side of the eye discs. These results indicate that overexpression of wge represses expression of eya in eye discs in a cell-autonomous manner, independent of the up-regulation of wg. Overexpression of wge-inhibits expression of eya, a gene downstream of ey; however, ey-lacZ expression in eye discs is not repressed by clonal induction of wge overexpression. Consistent with these results, Ey-mediated ectopic induction of eya is inhibited by coexpression of wge. Ectopic induction of ey under the control of dpp-GAL4 induces ectopic induction of eya-lacZ in the leg, wing, and antennal discs, but the ectopic expression of eya-lacZ is totally suppressed by the coexpression of wge. Coexpression of GFP does not suppress the Ey-mediated induction of eya, indicating a specific effect of wge overexpression on eya expression. These results suggest that overexpression of wge suppresses the eye development program downstream of ey. Eya repression occurs only on the dorsal side of the eye disc when wge is overexpressed with ey-GAL4, although clonally overexpressed wge also represses Eya on the ventral side. Further analysis is required to explain these phenomena (Katsuyama, 2005).

Although ey-GAL4-dependent overexpression of wge-induced Wg expression on the dorsal side of the eye discs, clonal induction of wge overexpression did not induce Wg expression in eye discs. Consistent with these results, clonal induction of wge overexpression did not induce either ectopic expression of Vg in eye discs or ectopic formation of wings in the eye field on the head, suggesting that overexpression of wge in a relatively large field is required for the transformation of eye to wing. Repression of eya but absence of vg expression in clonal overexpression of wge might represent an intermediate step toward wing transformation, which has to be analyzed further. In wing discs, clonal induction of wge overexpression did not affect expression of either Vg or Wg, indicating a specific effect of clonal induction of wge overexpression on eya expression in eye discs (Katsuyama, 2005).

To investigate the requirement of wge for wing development, a wge-deficient mutant was generated by mobilizing the P element. The deletion of wge was screened by using genomic PCR in 126 excision lines. In one line, Wge40, sequencing analysis after genomic PCR revealed that a 2,215-bp sequence, including the wge first exon, was deleted. In Wge40, there was no wge expression, and the expression of a gene neighboring Wge, Irp-1A, was not affected, indicating that Wge40 is a wge null mutant. The embryogenesis of Wge40 was quite normal, but the development of Wge40 gradually stopped after the first-instar larval stage, suggesting crucial roles of wge in larval development or growth. A rare rescue (3%) of larval lethality was observed by wge overexpression by using a heat-shock promoter, reflecting the context-dependent requirement of wge for development. Rescue was never observed, however, in the absence of heat shock (29°C for 2 h every 24 h). wge mutant clones were then induced by using the flp/FRT system with the Minute technique. When the wge mutant clones were introduced in wing discs 48 h AED by heat shock, no Vg expression was observed in many cells within the clones, but some cells in the clones still expressed Vg. The expression of dpp-lacZ was observed in the wge mutant clones, indicating a specific effect of the wge mutation on vg expression. The size of these clones is relatively small, whereas Vg expression is not affected in all but a few small-sized clones when the wge mutant clones are introduced 72 h AED. These results indicate that wge is required for vg induction in wing discs in a context-dependent manner. Compared with the wge mutant clones that were introduced 72 h AED, the wge mutant clones that were introduced 48 h AED were small in size, suggesting stage-specific involvement of wge on the growth of disc cells (Katsuyama, 2005).

Clonal induction of wge overexpression represses eya-lacZ expression in eye discs. Derepression of eya was investigated in the wge mutant clones. Contrary to prediction, there was no misexpression of Eya in wge mutant clones that were introduced at both 48 and 72 h AED in imaginal discs such as antennal, leg, and wing discs. In eye discs, Eya was not expressed in many cells within the wge mutant clones but was expressed in some cells when the clones were introduced 48 h AED, whereas the expression of Eya was not affected in the clones when the wge mutant clones were introduced 72 h AED. These results indicate that wge is required for eya expression in eye discs in a context-dependent manner, which is similar to the function of wge in the regulation of vg expression in wing discs. Consistent with the requirement of wge for the function of vg and eya, adult structures, such as eyes, wings, and legs, were malformed when wge mutant clones were introduced 72 h AED. The mutant clones were distinguished by the absence of Ubi-GFP. Pupal lethality was induced when wge mutant clones were introduced 48 h AED. The participation of wge in the development and growth of various appendages confirmed the ubiquitous expression of wge. RT-PCR findings revealed constitutive expression of wge throughout the larval and pupal developmental stages and ubiquitous expression of wge in larval tissues and imaginal discs. The ubiquitous expression of wge in various tissues was confirmed by in situ hybridization experiments (Katsuyama, 2005).

Wge has a BAH domain that is frequently found in proteins participating in the epigenetic regulation of gene expression. To investigate the nuclear localization and chromatin association of Wge, FLAG-tagged wild-type Wge and FLAG-tagged mutant protein (deltaBAH) lacking the BAH domain was expressed and salivary glands were stained with anti-FLAG antibody. Both wild-type Wge and DeltaBAH were localized in the nuclei of the salivary glands, whereas only wild-type protein, and not DeltaBAH, localized at specific sites on polytene chromosomes. These results indicate that Wge associates with chromatin in a BAH domain-dependent manner. The association of Wge with chromatin seems to be crucial for Wge function, because DeltaBAH does not induce ectopic wings in the eye field when it is expressed by ey-GAL4 (Katsuyama, 2005).

The binding sites of Wge were analyzed with DAPI staining (DNA) under higher magnification. Some signals overlapped with DAPI staining and others did not overlap with DAPI staining, suggesting that Wge localizes in both bands and interbands of polytene chromosomes. The Wge binding sites on polytene chromosomes were compared to that of Posterior sex combs (PSC). Almost all PSC binding sites were coincident with some Wge binding sites, suggesting that some Wge function is related to PSC function. The genetic interaction between wge and Psc was investigated with Wge40 and Psc1. The extra sex comb phenotype of Psc1 was suppressed by the loss of one dose of wge, suggesting a wge function similar to that of trithorax-group (trxG) genes, which antagonize PcG genes. There was a maternal effect in Wge40 single heterozygotes, as indicated by the appearance of an additional sex comb on the second tarsomere of the first leg. Such a transformation of the second to the first tarsomere of the leg occurs in some PcG mutants such as multi sex combs and cramped. The additional sex comb phenotype of Wge40 was suppressed by a partial loss-of-function of Psc. In both Wge40 and Psc1 single heterozygotes, the number of sex comb teeth on the first tarsomere of the first legs was increased, whereas there was no significant modification of the number of sex comb teeth in double heterozygotes. These results indicate that wge and Psc have antagonistic roles in both transformation from the second thoracic legs to the first thoracic legs and transformation from the second tarsomere to the first tarsomere of the first leg but not in the increase in the number of sex comb teeth (Katsuyama, 2005).

Thus, Wge localizes at specific sites on polytene chromosomes in a BAH domain-dependent manner, and wge overexpression induces a gain-of-function transformation of eyes to wings. The extra sex comb phenotype of Psc1 is suppressed by the loss of one dose of wge. The characteristics of wge are similar to that of trxG genes. The trxG genes act as suppressors of the Polycomb phenotype and are implicated in the activation of Hox selector genes. Therefore, similar to loss-of-function mutations in the PcG genes, gain-of-function mutations in trxG genes cause ectopic expression of Hox selector genes and homeotic transformations. TrxG proteins also localize at specific sites on polytene chromosomes, and one of the trxG proteins, ASH1, has a BAH domain. There are several functional differences, however, between wge and trxG genes. One major functional difference is Hox selector gene independence on homeotic transformation. Mutations of the trxG genes cause homeotic transformations through the modulation of transcriptional regulation of the Hox selector genes. In contrast, overexpression of wge induces ectopic wings in an Antp-independent manner. Although endogenous wing development is considered to be independent of Antp, Antp is the only Hox selector gene examined that induced eye-to-wing transformation in the system. In addition, wge overexpression does not induce ectopic expression of Antp in eye discs. Therefore, wge induces eye-to-wing transformation in an independent Hox selector gene. Moreover, wge is required for the ectopic wing formation that is induced by the expression of Antp and the activation of Notch signaling. These results suggest that wge is involved in the regulation of field-specific selector gene expression but not in the regulation of region-specific Hox selector gene expression. There is probably a regulatory mechanism that determines the field-specific identity after determination of region-specific identity by Hox selector genes. Another difference between wge and trxG is that the trxG functions as activators, whereas wge overexpression also represses eya (Katsuyama, 2005).

wge is required for the expression of both vg in wing discs and eya in eye discs in a context-dependent manner. Overexpression of wge, however, induces ectopic expression of vg and represses eya expression in eye discs. In wing discs, wge overexpression does not induce either ectopic expression of vg or repression of vg. Therefore, wge regulates expression of vg and eya in a context-dependent manner. Consistent with the context-dependent function of wge, wge is expressed ubiquitously throughout larval to pupal development and in various tissues. The field-specific identity should be determined from an equivalent group of cells. This characteristic is observed not only in normal development but also in artificial situations of imaginal discs called transdetermination, in which, after regenerative cell growth, disc cells change their determined state to another determined state, e.g., a leg disc transdetermines to a wing disc. Transdetermination is a polyclonal event and not the result of either differentiation of reserve cells or somatic mutations. Context-dependent regulation of gene expression by a ubiquitously expressed gene might explain how differences are created within a group of equivalent cells (Katsuyama, 2005).

A novel Pzg-NURF complex regulates Notch target gene activity

Drosophila putzig was identified as a member of the TRF2-DREF complex that is involved in core promoter selection. Additionally, putzig regulates Notch signaling, however independently of DREF. This study shows that Putzig associates with the NURF complex. Loss of any NURF component including the NURF-specific subunit Nurf 301 impedes binding of Putzig to Notch target genes, including cut, Enhancer of split and vestigial, suggesting that NURF recruits Putzig to these sites. Accordingly, Putzig can be copurified with any NURF member. Moreover, Nurf 301 mutants show reduced Notch target gene activity and enhance Notch mutant phenotypes. These data suggest a novel Putzig-NURF chromatin complex required for epigenetic activation of Notch targets (Kugler, 2010).

Putzig is a component of a large multiprotein complex that includes the TATA-box-binding-protein-related factor 2 (TRF2) and the DNA-replication related element (DRE) binding factor DREF. The TRF2-DREF complex has been associated with the transcriptional regulation of replication-related genes that contain DREF binding sites. Accordingly, Pzg acts as a positive regulator of cell cycle and replication-related genes. In addition to this, Pzg is also required for Notch target gene activation in a DREF-independent manner. Presumably, Pzg functions at the level of chromatin activation, because the open chromatin structure typical of active Notch target genes is no longer detectable in a pzg mutant background (Kugler, 2010).

The TRF2-DREF complex consists of more than a dozen of proteins and the biochemical function of most of them remains still elusive. Interestingly, it also contains three members of the nucleosome remodeling factor (NURF), imitation switch (ISWI), Nurf 55 and Nurf 38. NURF is a multisubunit complex that has been associated with chromatin activation and repression. NURF triggers nucleosome sliding thereby provoking changes in the dynamic properties of the chromatin. The subunit ISWI is a member of the SWI2/SNF ATPase family and is thought to provide energy for nucleosome remodeling. Nurf 38 encodes an inorganic pyrophosphatase, which catalyzes the incorporation of nucleotides into a growing nucleic acid chain during transcription, replication, and DNA repair mechanisms. Nurf 55 harbors WD-40 repeats, which allow interaction with other proteins and protein complexes. The fourth and largest subunit Nurf 301 is specific to the NURF complex, whereas all other members are shared with other chromatin modifying complexes. Accordingly, Nurf 301 is not a component of the TRF2-DREF complex. Nurf 301 exhibits a number of protein motifs that typify transcription factors and other chromatin modifying proteins. In addition, the N-terminal region of Nurf 301 shows homology to the DNA-binding protein HMGA (high mobility group A) implying that Nurf 301 mediates the contact with the DNA or provides a platform to recruit other transcription factors. In this context it has already been shown that Nurf 301 is required for the transcriptional activation for example of homeotic genes and notably of Ecdyson-receptor (EcR) and Wingless target genes (Kugler, 2010).

The DREF independence of Pzg during the activation of Notch target genes raised the possibility that it may instead involve the NURF complex for chromatin activation. This study provides evidence for a functional interplay between Pzg and the NURF complex with regard to Notch target gene activation. Coimmunoprecipitations revealed that Pzg is present in protein complexes containing the known NURF subunits. Moreover, Pzg binding on Notch target genes is neither detectable in mutants of the NURF-specific subunit Nurf301, nor in mutants affecting other subunits of NURF. In addition, Nurf301 is required for Notch target gene expression, which is impaired in Nurf301 mutant cell clones. Consistent with this, Nurf301 mutants enhance the Notch mutant wing phenotype, strongly arguing for an involvement of the NURF complex in Pzg-mediated epigenetic Notch target gene activation (Kugler, 2010).

This work shows that Pzg is associated with at least two different types of protein complexes that are involved in transcriptional activation: the TRF2-DREF complex and the NURF complex. Interestingly, these two complexes share several members apart from Pzg despite their different roles in core promoter selection versus nucleosome sliding and chromatin activation. However, the specific role for Pzg in the promotion of Notch target gene transcription involves NURF and not the TRF2-DREF complex. Notably, NURF also promotes efficient expression of a subset of Wingless target genes. In this case, a direct interaction between ISWI and Armadillo, the major transcriptional coactivator of Wingless targets, was shown. There is no indication however, that pzg is involved in the regulation of wg, suggesting that the NURF complex recruits Pzg only onto specific promotors. Furthermore, the NURF subunit Nurf 301 contacts the Ecdysone receptor (EcR), thereby modulating the activity of ecdysone signaling during the larval and pupal stages of Drosophila development. How is NURF recruited to Notch target sites? Notch target gene activation involves a ternary complex containing the DNA-binding protein Suppressor of Hairless [Su(H)], intracellular Notch, and Mastermind, plus other more general coactivators. There is no indication of a direct contact of Pzg to either Notch or Su(H), tested by coimmunoprecipitations as well as yeast two-hybrid assays. However, contacts between the other components, notably Mastermind or ISWI cannot be excluded. Mastermind has been shown to interact with several chromatin modifying proteins, for example, with the histone acetyltransferase p300 or with cyclin-dependent kinase 8 (Kugler, 2010).

Several studies in Drosophila and vertebrates have shown that many Notch-responsive target genes are regulated by combinatorial signal inputs, which need the Notch ternary complex and additional cooperators bound to sites nearby. In contrast to cofactors within the transactivation complex, these other factors do not physically interact with the Notch ternary complex but instead synergize during transcriptional activation at Notch target gene promoters. It is conceivable, that a Pzg-NURF complex is likewise needed in conjunction with the Notch transactivator complex for full Notch target gene expression (Kugler, 2010).

It is well established, that chromatin modification complexes share several components. For example, ISWI is not only contained in NURF and TRF2-DREF complexes but also in chromatin-remodeling and assembly factor (CHRAC) and ATP-utilizing chromatin-remodeling and assembly factor (ACF) in Drosophila, where it serves to increase the accessibility of nucleosomal DNA. Nurf 55, also known as CAF-1, forms a stable complex with Drosophila Myb and E2F2/RBf and regulates the transcription of several developmentally important genes. Like ISWI and Nurf 55, also Nurf 38 is present in the TRF2-DREF complex. Pzg is contained within the TRF2-DREF and within the NURF complex serving the activation of proliferation related genes and N target genes, respectively. Not all NURF complexes, however, require pzg, for example, as during the activation of Wg target genes. Sharing components raises the question, how specificity of the different complexes is achieved. Obviously, specificity is mediated either by unique subunits or by certain combinations of shared subunits. These subunits may specifically modulate the activity of the ATPase subunit or, more likely, may help to target the remodeling complexes to particular promoters. Two members of the NURF complex, ISWI and Nurf 301, have been shown to directly target transcription factors. It is tempting to speculate, that Pzg might be a specific cofactor needed to realize some of the operation spectrum of NURF, notably during the epigenetic regulation of Notch target genes (Kugler, 2010).

Drosophila araucan and caupolican integrate intrinsic and signalling inputs for the acquisition by muscle progenitors of the lateral transverse fate

A central issue of myogenesis is the acquisition of identity by individual muscles. In Drosophila, at the time muscle progenitors are singled out, they already express unique combinations of muscle identity genes. This muscle code results from the integration of positional and temporal signalling inputs. This study identified, by means of loss-of-function and ectopic expression approaches, the Iroquois Complex homeobox genes araucan and caupolican as novel muscle identity genes that confer lateral transverse muscle identity. The acquisition of this fate requires that Araucan/Caupolican repress other muscle identity genes such as slouch and vestigial. In addition, Caupolican-dependent slouch expression depends on the activation state of the Ras/Mitogen Activated Protein Kinase cascade. This provides a comprehensive insight into the way Iroquois genes integrate in muscle progenitors, signalling inputs that modulate gene expression and protein activity (Carrasco-Rando, 2011).

The study of myogenesis in Drosophila has increased the understanding of how the mechanisms that underlie the acquisition of specific properties by individual muscles are integrated within the myogenic terminal differentiation pathway. Thus, the current hypothesis proposes that distinct combinations of regulatory inputs leads to the activation of specific sets of muscle identity genes in progenitors that regulate the expression of a battery of downstream target genes responsible for executing the different developmental programmes. However, the analysis of the specific role of individual muscle identity genes and of their hierarchical relationships is far from complete since the characterisation of direct targets for these transcriptional regulators is very scarce (Carrasco-Rando, 2011).

ara and caup, two members of the Iroquois complex, have been identified as novel type III muscle identity genes. The homeodomain-containing Ara and Caup proteins are necessary for the specification of the lateral transverse (LT) fate. ara/caup appear to be bona fide muscle identity genes. Indeed, similarly to the identity genes Kr and slou, absence of ara/caup does not interfere with the segregation of muscle progenitors or their terminal differentiation, but modifies the specific characteristics of LT1-4 muscles, which are transformed towards VA1, VA2, LL1 and LL1 sib fates. These transformations may be due in part to the up-regulation of slou and vg in the corresponding muscles. Thus, a recent report (Deng, 2010) shows that forced expression of vg in LT muscles induces changes in muscle attachments similar to the ones observed in LT1 in ara/caup mutant embryos. However, it should be stressed that although in ara/caup mutants LT muscles are lost in more than 95% of cases, they are not completely transformed into perfect duplicates of the newly acquired fates. For instance, while the specific LT marker lateral muscles scarcer (lms) is lost in 91% of cases, ectopic slou expression is detected in only 75% of cases. These partial transformations might be due to differences in the signalling inputs acting in the mesodermal region from where these muscles segregate. Unpublished data also showed that forced pan-mesodermal expression of ara/caup alter the fates of many muscles both in dorsal and in ventral regions without converting them into LT muscles (i.e., they do not ectopically express lms). Similarly, Kr and slou ectopic expression is not sufficient to implement a certain muscle fate. The failure to recreate a given muscle identity by adding just one of the relevant muscle identity proteins reveals the importance that cell context, that is, the specific combination of signalling inputs and gene regulators present in each cell, have in determining a specific muscle identity (Carrasco-Rando, 2011).

Analysis of the myogenic requirement of ara/caup has revealed several features about how these genes act to implement LT fates. Thus, although they are expressed in six developing embryonic muscles, only four of them, LT1-4, are miss-specified in the absence of Ara/Caup. The remaining two, DT1 and SBM, seem to develop correctly, according to morphological as well as molecular criteria. It is worth noting that the requirement for ara/caup genes in these six muscles correlates with the onset of their expression. Thus, in the affected LT1-4 muscles Ara/Caup can be first detected at the earliest step of muscle lineages, that is in the promuscular clusters. In contrast, in the unaffected muscles ara/caup start to be expressed later, in the DT1/DO3 progenitor and the SBM founder. This suggests that in muscle lineages ara/caup have to be expressed very early to repress slou and vg to implement the LT fate. Several data support this interpretation. For instance, the observation that ara/caup are co-expressed with slou in DT1, whereas they repress slou in LT3-4, may be related to the fact that slou expression precedes that of ara/caup in the DT1 lineage. Should this be so, one would expect that ectopic expression of ara using the early driver mef2-GAL4, would repress slou in DT1, as it actually does, whereas this repression is not evident using the late driver Con-GAL4. Furthermore, the hypothesis of the relevance of the timing of muscle identity gene expression for muscle fate specification might also apply to the case of slou, where a similar correlation between the strength of the loss-of-function slou phenotypes in specific muscles and the onset of slou expression has also been found (Carrasco-Rando, 2011).

It should be stressed that the generation of the LT code depends not only on the early presence of Ara/Caup on the promuscular clusters but also on the absence (or strong reduction) of DER/Ras activity at that precise developmental stage and location. There is a dynamic regulation of MAPK signalling in the lateral mesoderm. Caup-expressing muscles develop from DER-independent clusters whereas the duplicated muscles observed in ara/caup mutants derive from progenitors that segregate very near the LT progenitors, but originate in DER-dependent promuscular clusters that are specified slightly later in development. Furthermore it was observed both by in vivo and in cell culture that low MAPK activity is required for Caup-dependent slou repression. Therefore, the role of Ara/Caup in the implementation of LT fate is interpreted as follows. At mid stage 11 in the myogenic mesoderm, groups of mesodermal cells acquire myogenic competence as a result of interpreting a combinatorial signalling code that reflects their position along the main body axes, as well as the state of activation of different signalling pathways. Accordingly, these clusters initiate the expression of lethal of scute and a unique code of muscle identity genes, as has been shown in great detail for eve expression in the dorsal mesoderm. In the case of the dorso-lateral mesoderm this code includes ara/caup and Kr and implements the LT fate. Since the level of activation of the Ras/MAPK cascade is low in these clusters, Ara/Caup will behave as transcriptional repressors, preventing the activation of slou or vg in LT1-2 and LT3-4 clusters, which would be otherwise activated in this location. Thus, Ara/Caup implement the LT fate by repressing the execution of the alternative fates (Kr+, Slou+, Con+, Poxm+ and Kr+, Vg+) that would give rise to duplicates of PVA1/VA2 and PLL1/LL1sib, respectively, and by allowing a different identity gene code (Kr+, Caup+, Con+, lms+) that generates the LT fate (Carrasco-Rando, 2011).

Slightly later the Ras/MAPK pathway becomes active at the dorsolateral region. This changes the combinatorial signalling code and coincides with a change in the muscle identity genes expressed by the promuscular clusters that segregate from this position, which now accumulate Kr but not Ara/Caup. Progenitors born from them will express either slou or vg and give rise to VA1-2 and LL1/LL1sib fates, all DER-dependent (Carrasco-Rando, 2011).

The data suggested that Ara/Caup might act as repressors of slou in the Drosophila mesoderm. Therefore whether slou might be a direct target of Ara/Caup was investigated. An 'in silico' search of a previously reported slou cis-regulatory region identified two putative Iro binding sites (BS) at positions +129 (BS1) and -1642 (BS2), relative to the transcription start site, which match the consensus ACAN2-8TGT. This regulatory region was cloned in a Luciferase reporter vector and Luciferase activity was measured in Drosophila Schneider-2 (S2) cells transiently transfected with this construct and increasing amounts of HA-tagged Caup. Contrary to expectations, it was found that addition of Caup-HA increased the basal Luciferase activity driven by the slou regulatory region in a dose dependent manner, indicating that Caup acts as a transcriptional activator of slou under these conditions. The reported regulation of the chicken Irx2 factor by MAPK (that switches it from repressor to activator) could explain this result. Since Western Blot analysis of S2 lysates using an antibody against diphospho-extracellular-signal related kinase (dpErk) showed the MAPK pathway to be active in S2 cells, and experimental evidence has been obtained showing the presence of phosphorylated Caup in S2 cells with constitutively active MAPK pathway, it was hypothesized that the activation effect of Caup in S2 cells could be due to the Ras/MAPK cascade turning Caup from transcriptional repressor into activator. Indeed, the inhibition of the Ras/MAPK pathway by the PD98059 MAP-erk kinase-1 (MEK1) inhibitor induced a Caup-dose dependent decrease in Luciferase activity driven by the slou regulatory sequences. This result could not be attributed to a direct effect of the inhibitor over the slou promoter, since its addition did not modify the basal Luciferase activity of the construct (Carrasco-Rando, 2011).

Thus S2 cell experiments suggest a molecular mechanism by which the Ras/MAPK pathway modulates the transcriptional activity of Ara/Caup on slou. Low MAPK activity and direct binding of Caup to BS1 site of the slou gene would favour strong repression of slou. BS1 could be embedded in a silencer regulatory element or its binding to Caup may block transcription of the downstream located luciferase gene. On the contrary, Caup-dependent activation of slou would be dependent on MAPK signalling. It is hypothesized that MAPK-dependent Caup phosphorylation could modulate its interaction with different transcriptional co-factors or/and its binding site affinity (Carrasco-Rando, 2011).

Furthermore, in vivo evidence indicates a repressor function of presumably non-phosphorylated Caup on slou since forced activation of the Ras pathway allows co-expression of slou and caup. On the other hand, the ectopic expression of slou induced by caup-over-expression is suggestive of a possible activator function of phosphorylated Caup (Carrasco-Rando, 2011).

The role of IRO proteins in cell fate specification is conserved in both vertebrates and invertebrates. This study has shown that the interplay between MAPK signalling and IRO activity found in vertebrate neuroepithelium is also at work in Drosophila myogenesis. This study has identifed potential direct target of Ara/Caup, slou and has proposed vg as a candidate gene to be regulated by Ara/Caup. In both cases the genes subordinated to ara/caup encode transcription factors that might in turn regulate the expression of other genes, genes that must be repressed in LT muscles in order to acquire the LT fate. These results, therefore, provide insights into the way Ara/Caup control lateral muscle identity and on the role of signalling pathway inputs to modulate the activity of these transcription factors, with consequences in their downstream targets. It also highlights the importance that the specific combination of muscle identity genes, their hierarchical relationships and their temporal activation have in determining the identity of a given muscle cell, very alike to what is at work during the acquisition of neural fates (Carrasco-Rando, 2011).

Continued vestigial Regulation part 3/3 | back to part 1/3

vestigial: Biological Overview | Evolutionary Homologs | Targets of Activity and Protein Interactions | Developmental Biology | Effects of Mutation | References

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

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