Chip

Targets of Activity

Chip is present in all nuclei examined and at numerous sites along the salivary gland polytene chromosomes. Embryos without Chip activity lack segments and show abnormal gap and pair-rule gene expression, although no LIM domain proteins are known to regulate segmentation. The unsegmented phenotype of Chip mutant germ-line clone embryos resembles that displayed by embryos homozygous for null alleles of the pair-rule gene even-skipped. Eve is expressed in Chip mutants but the pattern is abnormal. There are stripes, but the number is often less than the seven found in wild type embryos. The first stripe to appear is often wider and stronger than in wild type, whereas later-appearing stripes are narrow, weaker, and more uneven than wild type. Pair rule genes are regulated by gap genes. Lack of active Chip affects Giant more severely than the other gap proteins. In wild-type precellular and early cellular blastoderm embryos, Gt is restricted to two broad bands, whereas in Chip germ-line clone embryos, Gt is expressed at low to moderate levels in the entire embryo, including the pole cells. At later stages, Gt expression is similar to wild type. Lack of Giant in embryos lacking active Chip can explain the decreased expression of Kruppel and Knirps because Gt represses Kr and kni. It is conceivable that abnormal Gt expression also weakens Eve stripe 2 (Morcillo, 1997).

Protein Interactions

Chip was cloned and found to encode a homolog of the recently discovered mouse Nli/Ldb1/Clim-2 and Xenopus Xldb1 proteins, which bind nuclear LIM domain proteins. Chip protein physically interacts with the LIM domains in the Apterous homeodomain protein, and Chip interacts genetically with apterous, showing that these interactions are important for Apterous function in vivo. Importantly, Chip also appears to have broad functions beyond interactions with LIM domain proteins (Morcillo, 1997).

Formation of the dorsal-ventral axis of the Drosophila wing depends on activity of the LIM-homeodomain protein Apterous (Ap). Ap activity levels are modulated by dLMO, the protein encoded by the Beadex (Bx) gene. Overexpression of dLMO in Bx mutants interferes with Apterous function. Conversely, Bx loss-of-function mutants fail to down-regulate Apterous activity at late stages of wing development. Biochemical analysis shows that dLMO protein binds to Chip, thus competing with Apterous binding to Chip. These data suggest that Apterous activity depends on formation of a functional complex with Chip and that the relative levels of dLMO, Apterous, and Chip determine the level of Apterous activity. The dominant interference mechanism of dLMO action may serve as a model for the mechanism by which LMO oncogenes cause cancer when misexpressed in T cells (Milan, 1998).

LIM domains are thought to mediate protein interactions and are found in a variety of different types of proteins, often in combination with other recognized protein domains, as in the LIM-homeodomain proteins. dLMO belongs to a class of LIM domain proteins that have two LIM domains and no other recognizable motifs (hence, the designation LMO, for LIM only). In view of the effects of dLMO on Ap function, experiments were designed to discover whether dLMO can interact with Ap, a LIM homeodomain protein and Chip, an LDB (LIM domain binding) protein. LDB proteins bind to the LIM domains of nuclear LMO proteins of the type encoded by Bx. Genetic interactions between chip and ap suggest that as with Ldb1 and XLim1, Chip binding might activate Ap function. When overexpressed, Bx appears to interfere with Ap function without affecting either Chip or Ap protein expression. This raises the possibility that dLMO might interfere with binding between Ap and Chip. This was tested using a coimmunoprecipitation assay in which the binding between constant amounts of Chip and Ap proteins was challenged by increasing concentrations of Bx protein. Chip protein can be immunoprecipitated with T7-epitope-tagged Ap protein and anti-T7 antibody, showing that Ap and Chip proteins bind in vitro. Binding between Chip and Ap was challenged by adding increasing amounts of in vitro-translated dLMO protein. A dose-dependent decrease in the amount of Chip immunoprecipitating with Ap is observed as the amount of dLMO protein is increased and a corresponding increase in the amount of Chip immunoprecipitating with dLMO is also observed. These observations indicate that dLMO can bind to Chip in vitro and can compete for binding between Chip and Ap in a concentration-dependent manner. As a further test, the LIM domains of Ap were expressed as a GST fusion protein and tested for binding to full-length dLMO, Chip, and Ap proteins. Ap binds to itself and to Chip but not to dLMO in the GST-pull-down assay. This suggests that dLMO interferes with formation of the active Ap-Chip complex by competing with Ap for binding to Chip (Milan, 1998).

Two distinct functional domains have been identifed within NLI, the vertebrate homolog of Chip; the amino-terminal 200 amino acids (aa) mediate homodimerization, and 38 aa near the carboxyl terminus are required for high-affinity binding to LIM domains of LIM-HD proteins. These two domains within Chip are predicted to be present by primary sequence similarities between NLI, Chip, and orthologs from other species. To perform biochemical assays in vitro and misexpression studies in vivo, recombinant mutant Chip proteins were prepared in which either the predicted dimerization domain (DD) or the predicted LIM interaction domain (LID) was deleted. In addition, a control full-length Chip with no mutations was used. Five copies of the c-myc epitope were fused to the carboxyl terminus of each protein to monitor each protein's expression (van Meyel, 1999).

To determine whether the predicted DD and LID of Chip are indeed required for self-dimerization and LIM-HD interaction, respectively, double-immunoprecipitation assays were performed to biochemically detect high-affinity protein interactions. A 32-amino acid LID has been shown to be required for Chip to interact with Apterous, Islet, and Lim3. Deletion of 156 amino acids of the putative DD severely impairs the ability of Chip to dimerize. The DD is dispensable for LIM-HD interactions, and the LID is not required for Chip self-dimerization, indicating that these two domains of Chip function independently. Because deletion of the DD (aa 221-376) does not completely abolish the ability of Chip to self-dimerize, it is suspected that in conservative efforts to avoid disturbing nearby motifs, not all components of the dimerization domain were removed. Chip contains nearly 200 unique N-terminal amino acids not present in NLI that could participate in Chip dimerization (van Meyel, 1999).

Ap is the only known LIM-HD family member expressed in the developing wing disc. Chip interacts genetically with ap to cause disruptions of the wing margin, suggesting that these two genes act in the same pathway (Morcillo, 1997). In vitro analysis of Chip interactions with LIM-HD factors further suggests that the functionally relevant complexes required for transcriptional regulation in the wing disc are comprised of two Ap molecules bridged by two dimerized molecules of Chip. Alternatively, it is possible that Ap function is independent of Chip and that Chip, although expressed coincidentally with Ap, participates in an independent pathway and modulates Ap function indirectly by sequestering it from alternative complexes (van Meyel, 1999).

To distinguish whether Ap and Chip form functional complexes in vivo, or whether Ap works independently of Chip, the GAL4-UAS system was used to express Chip (ChipFL) and mutant Chip proteins (ChipdeltaDD and ChipdeltaLID). ap^GAL4, a P[GAL4] insertion in the ap locus, was used to direct reproducibly high levels of UAS transgene expression in the ap cells of the wing disc. As assayed by staining for the c-myc epitope, all Chip variants were expressed and were localized to the nucleus. Control ap^GAL4/+ heterozygotes display no wing defects, nor do they in the presence of a UAS-ap transgene. The latter observation is consistent with the hypothesis that a Chip/Ap tetrameric complex is functionally relevant, since it would not be compromised by Ap overexpression. In contrast, overexpression of ChipFL in the wing discs of ap^GAL4/+ heterozygotes results in defects resembling hypomorphic ap mutants. Wings were severely compromised in size and structure, and the wing margin was poorly demarcated. These wing defects are suppressed by simultaneous overexpression of Ap using a UAS-ap transgene. This indicates that the stoichiometry between Ap and Chip is an important factor in wing development and further suggests that overexpression of ChipFL titrates endogenous Ap to form incomplete complexes in which LID domains of Chip molecules remain vacant. The suppression of wing defects caused by co-overexpression of Ap is consistent with the idea that these incomplete complexes can be rendered fully functional by providing additional Ap to occupy vacant LID sites (van Meyel, 1999).

Overexpression of ChipdeltaDD results in severe wing defects that mimic those of extreme ap mutants. Like ChipFL, ChipdeltaDD is predicted to bind and sequester endogenous Ap but is unable to dimerize. The relative severity of ChipdeltaDD suggests that it renders Ap completely nonfunctional and blocks it from forming further Chip-bridged multimolecular interactions. Simultaneous overexpression of Ap suppresses the wing defects induced by ChipdeltaDD, suggesting that a pool of endogenous, dimerized Chip molecules exists to which ectopic Ap molecules bind and ''fill in'' unoccupied LID sites to form functional complexes. The suppression was not to the same extent as that achieved by Ap coexpressed with ChipFL and suggests that the poisoned ChipdeltaDD/Ap complexes can compete with fully functional complexes for binding to control elements in target genes (van Meyel, 1999).

Overexpression of ChipdeltaLID causes wing defects resembling those induced by overexpression of ChipFL. This experiment demonstrates a key role for Chip in wing development, since ChipdeltaLID is unable to interact with Ap and thus cannot simply sequester it from biological function. Importantly, these defects are not suppressed by simultaneous overexpression of Ap, as are those induced by ChipFL or ChipdeltaDD. Therefore, Ap cannot reconstitute function of complexes containing ChipdeltaLID, strongly supporting a role for direct physical interaction between Ap and Chip for function in vivo (van Meyel, 1999).

These results suggest that a tetrameric Chip/Ap complex mediates Chip and Ap function in the wing. In a simple model, the Ap LIM domains and the Chip LID domain bind one another and permit the homeodomains of two Ap molecules to be bridged by Chip dimerization. To test this model directly, the domains responsible for the interaction were removed and the remainder of each protein was tethered by fusing ChipdeltaLID directly to a LIM domain deleted version of Ap. This chimeric molecule should allow reconstitution of the Chip/Ap complex independently of the LID and LIM domains and rescue ap mutant wing defects if Chip and Ap are indeed required to bind one another. ap^GAL4 has been shown to act as a strong hypomorphic allele of ap. Using ap^GAL4 as a driver in an ap mutant background, it was found that ApdeltaLIM and ChipdeltaLID are incapable of rescuing any ap mutant wing defects. In contrast, the ChipdeltaLID:ApdeltaLIM chimera does rescue ap mutants. While the extent of rescue is not as complete as for Ap, flies expressing ChipdeltaLID:ApdeltaLIM exhibit significant wing outgrowth and a clearly demarcated margin with a triple row of sensory bristles. It is concluded from this result that physical interactions between Ap and Chip are required for appropriate margin formation and wing outgrowth (van Meyel, 1999).

In experiments to be reported elsewhere, it has been found that loss-of-function Chip mutations reveal ap-like defects in axon pathfinding and neurotransmitter production by Ap interneurons. Having implicated a role for Chip in three distinct functions of Ap, and given the widespread expression of Chip, it is speculated that Chip is an obligate cofactor for other LIM-HD activities. In fact, it would appear that the only way to exclude Chip from participating in complexes with LIM-HD factors is to selectively render it ineffective, perhaps by sequestration or by the formation of more favorable interactions between LIM-HDs and other factors. Indeed, both of these processes may act to limit the range of activity of Chip. (1) Chip may be sequestered in cells that coexpress dLMO, a member of the LIM-only subclass of nuclear LIM-containing proteins that have no homeodomain. dLMO can selectively inhibit Chip interactions with Ap in vitro and can modulate Ap function in the wing. (2) LIM-HD proteins may directly interact with proteins other than Chip. For example, in the absence of NLI, the LIM domains of Lhx-3 mediate direct heterodimerization with Isl-1 and Isl-2 but not with other LIM-HD family members. Drosophila Lim3 and Isl are likewise capable of forming heterodimers in the absence of Chip, suggesting that they can participate in both Chip-dependent and Chip-independent heterodimeric complexes in cells in which they are coexpressed. In fact, the data show that direct Islet/Lim3 interactions may be of higher affinity than Islet/Chip interactions, raising the possibility that under certain conditions Islet/Lim3 interactions may be favored over interactions with Chip. Recent studies have shown that Islet and Lim3 function in a combinatorial manner to specify motor axon pathway selection in flies (Thor, 1999), and analyses of Lhx-3/4 mutant mice indicate that a similar code operates in vertebrates. The implementation of any LIM-HD combinatorial code relies not only on the availability, concentration, and relative affinities of LIM-HDs; in addition, Chip/NLI and possibly LMO cofactors are also involved (van Meyel, 1999).

To test whether the active form of Ap is a complex involving two molecules of Ap and two molecules of Chip, the effects of expressing dominant-negative forms of Chip that differ in their ability to bind Ap were compared. Overexpression of wild-type Chip has dominant-negative activity in vivo. It has been suggested that this could be due to formation of trimeric complexes (Ap:Chip:Chip) that lack a second Ap molecule because the dominant-negative activity of Chip can be suppressed by overexpression of Ap. It was reasoned that a truncated form of Chip lacking the LIM interaction domain (ChipdeltaLID) would also serve as a dominant negative but that its activity should not be suppressed by overexpression of Ap. Before testing the activity of the ChipdeltaLID construct in vivo, it was verified that the molecular interactions between Ap and Chip in vitro are consistent with the expectations based on analysis of the human LDB protein, NLI (Milan, 1999).

Complex formation between Ap, Chip, and ChipdeltaLID was assayed using in vitro translated proteins. Ap was expressed with a T7-epitope tag, incubated with 35S-labeled Chip or ChipdeltaLID, and immunoprecipitated with anti-T7. Full-length Chip coprecipitates with T7-Ap. ChipdeltaLID is not recovered above background levels when incubated with T7-Ap, confirming that Chip needs the LID to bind effectively to Ap. ChipdeltaLID coprecipitates when incubated with T7-Ap and full-length Chip, demonstrating formation of a three part Ap:Chip:ChipdeltaLID complex. To verify that a Chip dimer can bridge two Ap molecules, a tagged version of Ap (Ap-TAP) was used. The biological activity of Ap-TAP is comparable to that of wild-type Ap when ectopically expressed in vivo under GAL4 control. Ap-TAP was in vitro translated and bound to IgG beads. The beads were washed and incubated with labeled Ap with or without unlabeled Chip. Without Chip, only background levels of 35S-Ap are recovered. When Chip is present, Ap-TAP beads bind 35S-Ap, indicating formation of the tetrameric complex in vitro (Milan, 1999).

Overexpression of Chip represses the Ap targets fringe-lacZ and dLMO (Beadex). Overexpression of ChipdeltaLID using patched GAL4 also represses fringe-lacZ and dLMO, indicating that both forms of Chip suppress Apterous activity when overexpressed. Overexpression of Chip under control of ap-GAL4 interferes with wing formation, producing a phenotype resembling the lack of ap function. This can be suppressed by coexpression of UAS-Chip and UAS-Ap. Overexpression of ChipdeltaLID using ap-GAL4 causes a mild apterous phenotype: distal notching of the wing margin and dorsal-to-ventral transformation of the alula. Although the phenotype suggests only partial reduction of Ap activity, the defects caused by overexpression of ChipdeltaLID cannot be suppressed by coexpression of Ap. These results suggest that ChipdeltaLID acts as a dominant negative for Ap activity in vivo by promoting the formation of a trimeric Ap:Chip:ChipdeltaLID complex that cannot bind another Ap molecule (Milan, 1999).

The LIM domain protein dLMO can compete with Ap for binding to its cofactor Chip. If the active form of Ap in vivo is a LIM-HD dimer bridged by a dimer of cofactor, dLMO might compete for Ap activity by displacing an Ap molecule from the Ap:Chip complex to form an Ap:dLMO heterodimer bridged by the cofactor. This model was tested by preparing a form of Ap that could dimerize but that could not be displaced by dLMO. To do so, a fusion protein consisting of the N-terminal dimerization domain of Chip linked to a C-terminal fragment of Ap containing the homeodomain (aa 270-469) was expressed. The new protein, called ChAp, lacks the LIM interaction domain of Chip and the LIM domains of Ap. Its structure should allow it to form an Ap dimer (Milan, 1999).

A test was performed to see whether ChAp has activity comparable to Ap in vivo. When expressed along the anteroposterior compartment boundary under control of dpp-GAL4, UAS-ap and UAS-ChAp produce essentially identical phenotypes. In both cases, ectopic wing margins are induced on both sides of the dpp-GAL4 stripe in the ventral compartment. The ectopic wing margin is due to the ectopic expression of Wingless in the ventral compartment. This correlates with ectopic induction of the dorsally expressed Ap targets fringe-lacZ and dLMO in ventral cells. These observations show that ChAp can mimic the effects of Ap in ectopic expression assays. A rescue assay was used to ask whether ChAp can functionally substitute for Ap in vivo. The wing defect in ap^GAL4/ap^rk568 flies is completely suppressed when wild-type Ap is expressed in dorsal cells using ap-GAL4. Dorsal expression of ChAp produces a comparable rescue. These results show that ChAp behaves like wild-type Ap when ectopically expressed and that ChAp can substitute for Ap in vivo (Milan, 1999).

According to the dimer model, ChAp should be sensitive to the dominant-negative activity of ChipdeltaLID but should not be subject to regulation by dLMO in vivo. dLMO is induced by Ap in the wing disc, and loss-of-function dLMO mutants produce defects that are thought to result from overactivation of Ap. ChAp overexpression in the dorsal compartment of an otherwise wild-type wing disc gives a phenotype that closely resembles the dLMO loss-of-function phenotype. The wings are smaller than wild type and are held in an abnormal position. The dorsal compartment is typically smaller than the ventral compartment, giving the wing a slightly curled appearance. The pattern of veins is also abnormal. Coexpression of the dominant-negative form of the cofactor, ChipdeltaLID, suppresses the ChAp overexpression phenotype. This indicates that dimerization is required for ChAp activity in vivo (Milan, 1999).

To ask whether ChAp is subject to regulation by dLMO, the ability of Ap and ChAp to suppress the effects of dLMO overexpression in vivo were compared. apGAL4/+; UAS-dLMO/+ wings show loss of the normal wing margin and sporadic patches of ectopic wing margin, thus promoting local overgrowth. Overexpression of wild-type Ap does not suppress the dLMO overexpression phenotype. Antibody staining shows that Wg is not expressed at the DV boundary in these discs. The wing pouch is very small, and the normally straight boundary between cells expressing Ap and adjacent nonexpressing cells is uneven. These observations suggest that Ap is nonfunctional in these discs in spite of being overexpressed. In contrast, coexpression of ChAp and dLMO restores Wg expression at the DV boundary even though dLMO is expressed at high levels. The resulting wings have a normal wing margin and resemble the dLMO mutant wing. These results suggest that ChAp overexpression phenocopies the dLMO loss-of-function mutant because ChAp is not sensitive to downregulation by dLMO. Consequently, ChAp remains active in the presence of excess dLMO (Milan, 1999).

The bridged dimer model suggests that removing dLMO activity would result in excess Ap activity. To test this, the properties of Chip and Ap interaction were exploited to regulate Ap activity in a dLMO mutant background. dLMO loss-of-function mutants were generated by excision of a GAL4-P element insertion in the second intron of the dLMO gene. Fortuitously, excision line hdp^R590 strongly reduces dLMO expression but leaves GAL4 and the cis-regulatory region unaffected, so that the mutant expresses GAL4 in the normal pattern of dLMO. hdp^R590 causes aberrant Serrate expression and a reduced dorsal wing pouch. The dLMO loss-of-function phenotype in this mutant can be suppressed by expression of ChipdeltaLID. The small wing size of hdp^R590 is fully rescued, and the abnormal venation is partially suppressed. Likewise, expression of a mutant form of Ap lacking only the homeodomain completely suppresses the hdp^R590 phenotype. Both of these constructs have mild dominant-negative effects that reduce Ap activity in vivo. These results confirm that the dLMO loss-of-function phenotype results principally from excess Ap activity at later stages of wing development. Further, they support the proposal that the normal function of dLMO is to regulate Ap activity levels by interfering with formation of an active complex consisting of two Ap molecules bridged by a dimer of Chip molecules (Milan, 1999).

The finding that ChAp can completely replace Apterous function in vivo suggests that the relevant feature of this tetrameric complex is the formation of a dimer of Ap. This molecule is not subject to negative regulation by dLMO and so remains constitutively active. The consequence is a failure to downregulate Ap activity as development proceeds. The phenotypic consequences of excess Apterous activity are comparable to those of the dLMO lack-of-function mutant. These findings support the view that the tetrameric complex between Ap and its cofactor Chip provides a means to generate the requisite bridged dimer of Ap, while allowing the activity of the complex to be regulated by the competitive inhibitor dLMO. It is suggested that this may provide a general model for regulation of LIM-HD activity. LMO family proteins may be generic antagonists of LIM homeodomain proteins through binding to their common LDB cofactors. The active complexes may be cofactor-bridged homodimers (as is the case for Ap in wing development) or heterodimers with other LIM-HD transcription factors or other types of LDB-binding transcription factors. Combinations of different transcription factors bridged by a cofactor dimer might broaden the range of possible target genes (Milan, 1999).

At the sequence level, Drosophila Lim1 is highly related to its vertebrate homologs. Given their conservation, it was of interest to see if this sequence homology translates into functional similarities at a molecular level. In Xenopus, Xlim-1 and the LIM-domain-binding protein (Xldb-1/NLI/CLIM- 2) interact in vitro, and cooperate in vivo to induce secondary axis structures (Agulnick, 1996). As the name implies, this association takes place through the LIM domains. More recently, the Drosophila homolog of Xldb1/NLI/CLIM-2, Chip has been cloned and shown to interact with the Ap protein (Morcillo, 1997; Fernandez-Funez, 1998). To determine if Lim1 and Chip interact in vitro, co-immunoprecipitation experiments were carried out. Using the Lim1 antibody, the ability of Chip to be immunoprecipitated by full-length and truncated Lim1 proteins was carried out. The results show that Chip can be immunoprecipitated in the presence of full-length Lim1, and a truncated Lim1 protein that contains the LIM domains (LIM-Lim1). HD-Lim1, which lacks the LIM domains and includes the homeodomain fails to coimmunoprecipitate Chip. Additionally, Chip by itself is not immunopreciptated by the Lim1 antibody. These results show that Lim1 has the capacity to interact with the LIM-domain-binding protein, Chip. This interaction requires the LIM domains of Lim1 and is independent of the Lim1 homeodomain. Similar to its vertebrate counterparts, and Ap in Drosophila, Lim1 and Chip may cooperate in vivo to modulate the transcriptional activity of its downstream target genes. Chip is ubiquitously expressed and therefore is present in all Lim1-expressing cells suggesting that an in vivo interaction is possible (Lilly, 1999).

The LIM homeodomain (LIM-HD) protein Apterous (Ap) and its cofactor DLDB/CHIP control dorso-ventral (D/V) patterning and growth of the Drosophila wing. To investigate the molecular mechanisms of Ap/CHIP function, their relative levels of expression were altered and mutants were generated in the LIM1, LIM2 and HD domains of Ap, as well as in the LIM-interacting and self-association domains of CHIP. Using in vitro and in vivo assays it was found that: (1) the levels of CHIP relative to Ap control D/V patterning; (2) the LIM1 and LIM2 domains differ in their contributions to Ap function; (3) Ap HD mutations cause weak dominant negative effects; (4) overexpression of ChipDeltaSAD mutants mimics Ap lack-of-function, and this dominant negative phenotype is caused by titration of Ap because it can be rescued by adding extra Ap; and (5) overexpression of ChipDeltaLID mutants also causes an Ap lack-of-function phenotype, but it cannot be rescued by extra Ap. These results support the model that the Ap-CHIP active complex in vivo is a tetramer (Rincon-Limas, 2000).

The GATA factor Pannier activates the achaete-scute (ASC) proneural complex through enhancer binding and provides positional information for sensory bristle patterning in Drosophila. Chip acts as a cofactor of the dorsal selector Apterous, and both Apterous and Chip also regulate ASC expression. Chip cooperates with Pannier in bridging the GATA factor with the HLH Ac/Sc and Daughterless proteins to allow enhancer-promoter interactions, leading to activation of the proneural genes, whereas Apterous antagonizes Pannier function. Within the Pannier domain of expression, Pannier and Apterous may compete for binding to their common Chip cofactor, and the accurate stoichiometry between these three proteins is essential for both proneural prepattern and compartmentalization of the thorax (Ramain, 2000).

Pnr is a member of the GATA-1 family of transcription factors and activates proneural function by binding to the dorsocentral (DC) enhancer located 4 kb and 30 kb upstream of ac and sc, respectively. Reported in this study is the characterization of Chip^E, a viable allele of Chip, that interacts with pnr genetically. Chip^E mutants show reduced ac-sc expression in the DC, associated with loss of DC bristles, and produce a phenotype similar to that of loss-of-function pnr alleles. This genetic interaction correlates with a physical interaction between Chip, Pnr, and the bHLH heterodimers (Ac/Sc-Da). Pnr interacts with the N terminus of Chip through its COOH terminus encompassing two helices that are conserved between D. melanogaster and D. virilis and that are probably both involved in protein-protein interactions. Chip dimerizes with the bHLH heterodimers through its C-terminal LID, known to mediate heterodimerization with LIM-containing proteins (Ramain, 2000).

In vertebrates, the Ldb1/NLI protein associates with GATA-1, Lmo2, and the bHLH E47, Tal1/SCL in an erythroid complex whose function is poorly understood. A similar Drosophila complex functions in vivo to regulate the ac/sc genes directly during establishment of the proneural prepattern.

Accurate coexpression of ac/sc is mediated by Pnr binding to the DC. The ac and sc promoters include E boxes that are targets for the Ac/Sc-Da heterodimers and support autoregulation during development. In cultured chicken embryonic fibroblast (CEF) cells, Pnr and the Ac/Sc-Da heterodimer activate expression of a CAT reporter linked respectively to the DC enhancer and to the ac promoter. Physical interactions between Pnr and the bHLHs lead to synergistic activation of the reporter when the regulatory sequences are associated. Pnr and the Ac/Sc-Da heterodimers are both required in flies for expression of a LacZ reporter linked to the promoter associated with the enhancer, but the analysis of Chip^E shows that Chip is also required for full activation in vivo. The interactions between Pnr and the bHLH mediated by Chip suggest that Pnr might also be involved in autoregulation. Interestingly, Chip interacts with Ac/Sc through the Ac/Sc bHLH domains, and it has been shown that the overexpression of a homologous bHLH domain is sufficient to mediate the proneural function of Ac/Sc (Ramain, 2000).

Chip has been identified in a genetic screen for mutations that reduce activity of the wing margin enhancer of the cut locus, and it has been proposed that Chip may act as a bridge allowing enhancer-promoter communications. Thus, if the flies have a unique functional Chip allele, they display a cut margin phenotype, and this effect is observed exclusively when they carry a gypsy insertion between the enhancer and the promoter on one chromosome. It has been proposed that binding of the Su(Hw) insulator protein to the gypsy insertion blocks communication on the mutant chromosome, thereby interfering with the functioning of the wild-type homolog. The interchromosomal insulation is detectable when Chip activity is reduced, and Chip and Su(Hw) are antagonistic to each other, suggesting that Chip may be a facilitator target of Su(Hw) (Ramain, 2000).

The Chip^E mutation specifically disrupts interactions with the bHLH and strongly affects the expression of a LacZ reporter linked to the ac promoter/DC enhancer in flies, suggesting that Chip also mediates enhancer-promoter communication in the ASC. Further evidence is provided by the Hw¹ mutant. Hw¹ carries a gypsy insertion within ac such that sc, which is located further downstream from the DC enhancer, is no longer expressed. In addition, the removal of the gypsy insulator largely restores sc expression in the DC proneural cluster (Ramain, 2000).

Thus, a complex containing Pnr, Chip, and the Ac/Sc-Da heterodimer activates ac-sc expression, and its function is antagonized by Ush, Ap, and Emc. Ush and Emc dimerize respectively with Pnr and the HLH Ac/Sc. The repressing effect of Ap may reflect its ability to interact with Chip, thereby depriving Pnr of its essential cofactor. Alternatively, Ap may weaken the enhancer activity of Pnr. Thus, Ap may compete directly with Chip for binding to Pnr (Ramain, 2000).

Within the domain of Pnr expression, Ap and Pnr compete for binding to their common Chip cofactor. Ap activity is mediated by a Chip dimer, whereas activation of ac-sc by Pnr requires a Chip monomer. Chip acts as a bridge between the Ac-Da heterodimer bound to the E boxes of the ac promoter and Pnr bound to the GATA sites of the DC enhancer. The activity of the resulting complex is antagonized by Ush and Emc, which negatively regulate Pnr and Ac/Sc functions during development. The repressing effect of Ap is mediated either by dimerization of Ap with Chip and/or Pnr or by Chip-assisted binding of Ap to sites located between the DC enhancer and the ac promoter. Additional cofactors, such as dLMO, may participate in this complex (Ramain, 2000).

Chip is required in flies for ASC activation, whereas it appears dispensable in CEF cells. This observation may reflect the nature of the reporter used in the transfection experiments where the DC enhancer is close to the ac promoter and poorly mimics the genomic organization of the ASC, where the DC enhancer has to regulate ac and sc simultaneously. Furthermore, the chromatin structure and its modifications associated with gene expression are probably not reproduced in the transient expression assay. Thus, ASC expression in flies may require additional coactivators recruited by Chip, including chromatin remodeling factors. Moreover, the activation of ac/sc probably requires the assembly of a higher-order nucleoprotein complex containing multiple transcription factors (enhanceosome), and Chip may allow the correct positioning of Pnr and the Ac/Sc-Da heterodimer in this structure (Ramain, 2000).

Chip^E mutants affect the scutellar and dorsocentral bristles in opposite fashions. It will be of interest to compare the regulation of the activity of the corresponding enhancers by Chip and Pnr (Ramain, 2000).

It has been proposed that appropriate combinations of proteins represent the positional cues that activate a given enhancer of the ASC complex. The disc is divided in large territories, but almost nothing is known concerning how these territories are further subdivided or how the positional information revealed by the accurate ac-sc expression is created. The present study provides a link between the spatial regulation of the proneural genes and the compartmentalization of the disc. ac-sc expression is stimulated by a complex containing the prepattern factor Pnr, Chip, and the bHLH proteins Ac/Sc and Da. Chip is an essential cofactor of the dorsal selector Ap, and these interactions coordinate the spatial transcription of the proneural genes. Ap is expressed specifically in the dorsal compartment of the wing pouch, and the juxtaposition of Ap-expressing with Ap-nonexpressing cells defines the dorsal/ventral organizing boundary where wingless (wg) expression is induced. On the thorax, ap and Chip are ubiquitously expressed, whereas wg expression occurs in a stripe straddling the lateral border of the domain of pnr expression. Moreover, Pnr activates wg. Pnr associates with Chip, and the domain of pnr expression appears devoid of Ap activity. As a consequence, this domain may define a boundary between a region devoid of Ap activity and a region where Ap is active. Alternatively, Pnr may associate with Ap, and the resulting heterodimer may regulate wg. Further studies will help to resolve this issue (Ramain, 2000).

The Drosophila mod(mdg4) gene products counteract heterochromatin-mediated silencing of the white gene and help activate genes of the bithorax complex. They also regulate the insulator activity of the gypsy transposon when gypsy inserts between an enhancer and promoter. The Su(Hw) protein is required for gypsy-mediated insulation, and the Mod(mdg4)-67.2 protein binds to Su(Hw). The aim of this study was to determine whether Mod(mdg4)-67.2 is a coinsulator that helps Su(Hw) block enhancers or is a facilitator of activation that is inhibited by Su(Hw). Evidence is provided that Mod(mdg4)-67.2 acts as a coinsulator by showing that some loss-of-function mod(mdg4) mutations decrease enhancer blocking by a gypsy insert in the cut gene. The C terminus of Mod(mdg4)-67.2 binds in vitro to a region of Su(Hw) that is required for insulation, while the N terminus mediates self-association. The N terminus of Mod(mdg4)-67.2 also interacts with the Chip protein, which facilitates activation of cut. Mod(mdg4)-67.2 truncated in the C terminus interferes in a dominant-negative fashion with insulation in cut but does not significantly affect heterochromatin-mediated silencing of white. It is inferred that multiple contacts between Su(Hw) and a Mod(mdg4)-67.2 multimer are required for insulation. It is theorized that Mod(mdg4)-67.2 usually aids gene activation but can also act as a coinsulator by helping Su(Hw) trap facilitators of activation, such as the Chip protein (Gause, 2001).

Apterous is a LIM-homeodomain protein that confers dorsal compartment identity in Drosophila wing development. Apterous activity requires formation of a complex with a co-factor, Chip/dLDB. Apterous activity is regulated during wing development by dLMO, which competes with Apterous for complex formation. Complex formation between Apterous, Chip and DNA stabilizes Apterous protein in vivo. A difference in the ability of Chip to bind the LIM domains of Apterous and dLMO contributes to regulation of activity levels in vivo (Weihe, 2001).

Since dLMO competes with Ap for binding to Chip, the possibility that Ap protein may be protected when it is in a complex with Chip was examined. To test this, Chip^e5.5 mutant clones, which lack Chip protein and therefore lack Ap activity, were created. Ap protein levels were reduced in Chip mutant clones, and increased in the wild-type twin spots which contain a higher level of Chip protein. To verify that reduced Chip activity does not affect ap mRNA levels ap-lacZ reporter gene expression was examined in discs expressing the dominant negative form of Chip, ChipDeltaLID. Ap protein levels were reduced in cells expressing ChipDeltaLID but ap-lacZ levels were unaffected. Thus, loss of Chip leads to reduced levels of Ap protein. It was noted that Chip mutant clones also lack dLMO expression. Thus, loss of Ap protein in Chip mutant clones does not correlate with expression of dLMO, as in wild-type cells. Rather, reduction of Ap levels correlates with the availability of Chip as a binding partner. This suggested that binding to Chip contributes to stabilization of Ap (Weihe, 2001).

ChipDeltaLID is capable of binding to full-length Chip through its dimerization domain, but cannot bind to Ap. Consequently, ChipDeltaLID leads to formation of trimeric complexes and thereby blocks Ap activity in vivo. The observation that ChipDeltaLID leads to reduced Ap stability without affecting ap-lacZ expression suggests that stabilization might require formation of tetrameric complexes between Chip and Ap. The tetrameric form of Chip and Ap is thought to be the active DNA-binding complex. Overexpression of ChipDeltaLID does not decrease the availability of LIM-binding sites in wild-type Chip, but does compete for tetramer formation. This raises the possibility that Ap stability might depend on whether it is able to form a DNA-binding complex with Chip (Weihe, 2001).

To test the effect of Ap-binding sites on Ap protein stability, a constant amount of a plasmid directing expression of a myc-tagged Ap protein was co-transfected with varying amounts of a plasmid carrying nine tandem repeats of a binding-site for the mammalian Ap-homolog Lhx2. This DNA sequence binds Drosophila Ap in S2 cells. The total amount of DNA in the transfection assay was held constant by varying the ratio of plasmid containing the binding sites and empty vector. Increasing the ratio of the plasmid containing the binding sites results in dose-dependent stabilization of Ap-myc protein. This observation supports the idea that availability of binding sites limits the amount of Ap protein that is stable in the cell when mRNA levels are held constant (Weihe, 2001).

Another means to test this possibility is by competition between Ap and a related protein for a fixed number of binding sites. For these experiments use was made of a fusion protein between Chip and Ap (called ChAp). In this protein the dimerization domain of Chip mediates dimerization of the DNA-binding domains of Ap. Thus, ChAp dimers should compete with endogenous Chip:Ap tetramers for DNA-binding sites. Use of the Myc tag versions of both proteins allowed direct comparison of their relative levels in co-transfected cells. Using this assay it was verified that increasing the level of ChAp-myc decreases the level of co-transfected Ap-myc in a dose-dependent manner. Expression of Chip-myc as a control has little effect on Ap-myc levels. Note that the level of Ap-myc construct was held constant in all samples. ChAp-myc and Chip-myc expression levels were controlled by varying the ratio of the expression constructs to the empty expression vector in the transfections (Weihe, 2001).

It was next asked whether competition for DNA-binding sites would affect Ap stability in the wing disc. Fortuitously, the antibody raised against Ap does not recognize ChAp. This allows the level of the endogenous Ap protein to be measured in cells expressing ChAp. ChAp expression under dpp^Gal4 control leads to a decrease in the level of endogenous Ap protein. Together, these observations suggest that Ap protein is unstable in vivo unless bound to DNA as part of an active complex with its co-factor Chip (Weihe, 2001).

This report addresses the problem of asymmetry in the competition between dLMO and Ap. The simplest model for competitive inhibition by dLMO would suggest that Ap should compete effectively with dLMO for binding to Chip when overexpressed. However, overexpression of Ap does not produce an excess of Ap activity. dLMO competes effectively for Ap activity, but the reverse is not true. Swapping the LIM domains of Ap for those of dLMO produces a functional Ap protein that is able to compete effectively with dLMO. This finding may provide an explanation for the non-reciprocal properties of Ap and dLMO. The effectiveness of dLMO as an inhibitor of Ap activity is attributed to an intrinsic difference in the ability of the LIM domains of these two proteins to bind to Chip. It is considered likely that the LIM domains of dLMO bind the LID of Chip with higher affinity than the LIM domains of Ap. However, these proteins have not been produced in soluble form at adequate concentrations and so the affinities of these binding interactions have not been determined directly (Weihe, 2001).

Other proteins might also contribute to stabilization of Chip-dLMO complexes or to destabilization of Chip-Ap complexes in vivo. Interactions involving Ap, Chip and other proteins have been reported. For example, Pannier interacts with Chip and competes with Apterous for patterning of the thorax. In this model, Chip is found in a complex with Pannier and dLMO, which promotes dorsal thorax formation. Chip is also found in a complex with Ap. The level of Chip is not in great excess, so competition occurs between Ap and Pannier for formation of Chip complexes, despite the fact that Pannier and Ap do not bind to Chip in the same way. It was noted that overexpression of dLAp-flag appears to interfere with Pannier complex formation, because it causes the formation of a cleft in the thorax, resembling a pannier loss-of-function mutant phenotype. Comparable overexpression of Ap does not do so. This suggests that dLAp competes more effectively than Ap for binding to Chip and so is more effective at sequestering Chip from Pannier-containing complexes. The relative affinity of these proteins appears to play an important role in maintaining the proper balance of complex formation in vivo. Numerous LIM-HD proteins have been found to play important roles in the development of a number of species. It seems likely that other LIM-homeodomain transcription factors will be regulated in similarly complex ways (Weihe, 2001).

Lim1 expression becomes discernible slightly later than expression al, clawless (cll/C15) and Bar, and maximal al expression in late third instar depends on Lim1 function. Cll expression is also significantly reduced in Lim1^7B2 (a null allele) clones in late third instar discs, indicating that not only al but also cll is positively regulated by Lim1 in the late third instar. Chi encodes a LIM domain binding protein and has been suggested to act as a co-factor for Lim1. Cll and Al signals are significantly reduced in clones of Chi^e5.5, a null allele of Chi. The concerted action of Lim1 and Chi is thus shown to be required for the maximal expression of cll and al in the late-third-instar pretarsus region (Kojima, 2005).

When Lim1 is misexpressed using blk-GAL4, the expression of al but not cll is induced. Thus, unlike al, cll may require an additional component for its maximal expression. Alternatively, cll may be less sensitive to activation by Lim1 than al (Kojima, 2005).

Ssdp proteins interact with the LIM-domain-binding protein Ldb1 to regulate development

The LIM-domain-binding protein Ldb1 is a key factor in the assembly of transcriptional complexes involving LIM-homeodomain proteins and other transcription factors that regulate animal development. Ssdp proteins (previously described as sequence-specific, single-stranded-DNA-binding proteins) have been identified as components of Ldb1-associated nuclear complexes in HeLa cells. Ssdp proteins are associated with Ldb1 in a variety of additional mammalian cell types. This association is specific, does not depend on the presence of nucleic acids, and is functionally significant. Genes encoding Ssdp proteins are well conserved in evolution from Drosophila to humans. Whereas the vertebrate Ssdp gene family has several closely related members, the Drosophila Ssdp gene is unique. In Xenopus, Ssdp encoded by Drosophila Ssdp or mouse Ssdp1 mRNA enhances axis induction by Ldb1 in conjunction with the LIM-homeobox gene Xlim1. Furthermore, an interaction between Ssdp and Chip (the fly homolog of Ldb1) in Drosophila wing development was identified. These findings indicate functional conservation of Ssdp as a cofactor of Ldb1 during invertebrate and vertebrate development (Chen, 2002).

To search for new interaction partners of Ldb1, a HeLa cell line was generated that expresses Ldb1 proteins carrying an N-terminal FLAG/HA epitope tag. Nuclear extracts prepared from these cells (and from nontransduced control cells) were incubated with immobilized anti-FLAG antibodies and the specifically bound materials were eluted by competition with excess amounts of FLAG peptide. Thereafter, immobilized anti-HA antibodies were used in a second round of purification. SDS/PAGE separation and silver staining of the final eluate revealed at least six polypeptides that were specific for the epitope-tagged Ldb1 sample and were not observed in the mock control (Chen, 2002).

All specific bands were analyzed by mass spectrometry. As expected, the 56-kDa band corresponded to epitope-tagged Ldb1. The 50-kDa doublet contained closely related proteins. Two peptides, SAQTFLSEIR and NSPNNISGISNPPGTPR, present in tryptic digests of the lower band of this doublet, correspond to human Ssdp1 (gi:13449489). The upper band contained several tryptic peptides (LALYVYEYLLHIGAQK, SAQTFLSEIR, and SSPGAVAGLSNAPGTPR) that correspond to Ssdp3 (gi:13400104), a structural relative of Ssdp1. Database searches revealed four human and three mouse Ssdp sequences that are closely related. In addition, a unique Drosophila Ssdp sequence was identified. Mouse Ssdp1 and Drosophila Ssdp were selected for functional studies reported below (Chen, 2002).

The function of Lim1 in axis formation in Xenopus depends on cooperation with Ldb cofactors. It was asked whether Ssdp might synergize with Xenopus Lim1 and Ldb1 in this system. Xenopus embryos were injected in the prospective ventral marginal zone with different combinations of synthetic mRNAs encoding Xlim1, Ldb1, and mouse or Drosophila Ssdp. Neither of the Ssdp proteins alone had any axis-inducing activity at the levels tested . High levels of Xlim1 and Ldb1 mRNAs induced incomplete secondary axes, whereas lower levels were ineffective. The lower levels of Xlim1 and Ldb1 mRNAs became highly effective, however, when coinjected with either mouse Ssdp1or Drosophila Ssdp mRNAs. Injection of low levels of Ldb1 plus Ssdp, or of Xlim1 plus Ssdp RNAs did not induce secondary axes. All secondary axes generated in this manner were incomplete. It seems that injection of high levels of Xlim1 plus Ldb1 mRNAs or of the triple combination of mRNAs caused both secondary axis induction and an inhibition of gastrulation, because those embryos that did not display a secondary axis were abnormal, mostly because of an open blastopore. This finding may explain the fact that injection of 80 pg of Xlim1, 80 pg of Ldb1, and 100 pg of mouse Ssdp1 mRNAs led to a lower proportion of secondary axis induction than 40 pg of Xlim1, 40 pg of Ldb1, and 100 pg of mouse Ssdp1 mRNAs. It is believed that the higher Xlim1/Ldb1 levels more effectively interfered with gastrulation, leading to a higher proportion of abnormal embryos rather than axis-duplicated embryos. Drosophila Ssdp mRNA was almost as effective as mouse Ssdp1 mRNA in inducing secondary axes when coinjected with Xlim1 and Ldb1 mRNAs. Thus, mouse Ssdp1 and Drosophila Ssdp proteins are sufficiently similar in their functions as to be interchangeable in ectopic expression experiments. A deletion of amino acid residues 1 to 121 from mouse Ssdp1 yielded a protein that did not bind Ldb1 after cotransfection into cultured cells and was also unable to synergize with Xlim1 and Ldb1 in axis induction. It is concluded from these results that Ssdp synergizes with Xlim1 and Ldb1 in vivo during gastrulation in Xenopus, and that this synergy is likely to require interaction between Ssdp and Ldb1 (Chen, 2002).

The single D. melanogaster Ssdp protein is encoded by CG7187 in polytene chromosome bands 90F1–2 in the right arm of the third chromosome. Searches of the EST databases identified 29 Ssdp ESTs, as well as ESTs for genes that flank Ssdp both proximally and distally on the chromosome. One Ssdp EST (GM14473) was completely sequenced by the Drosophila Genome Project and corresponds to a transcription unit with a single intron of 1,581 bp. Alignments of the remaining ESTs show that Ssdp encodes at least two transcripts that derive by alternative splicing at the 3' end of the first exon (both transcripts splice to the same second exon). Twenty-two ESTs match the GM14473 sequence; in four ESTs (GH23938, RE28366, GH18277, and RE64068) the first exon is 505 bp shorter than the first exon of GM14473. The predicted ORFs for both transcripts are entirely within the common second exons, suggesting that both transcripts encode identical proteins (Chen, 2002).

Five alleles of Ssdp were available for this study. All of these are lethal when homozygous or hemizygous (heterozygous to a chromosomal deletion that includes Ssdp). Most of the homozygotes and hemizygotes die during the pupal stages. Likewise, most transheterozygotes of various combinations of Ssdp alleles die as pupae. A few Ssdp¹¹/Ssdp^BG01663 and Ssdp¹¹/Ssdp^KG03600 flies survive to eclose as adults with mild cuticular defects, including a slight distortion of the posterior scutellar bristles, often accompanied by duplication of the anterior scutellar bristles. The survival of many homozygous Ssdp mutants to late pupal stages could be due to the maternal expression of Ssdp. Mitotic recombination in the germ line was used to create oocytes that lack maternal contributions of either Ssdp³¹ or Ssdp^neo48. When fertilized by a sperm that lacks the Ssdp gene, the zygotes that lack both maternal and zygotic Ssdp die at the beginning of the second larval instar. Survival of these animals through embryogenesis may conceivably be sustained by residual activity of the Ssdp alleles that were used in this study. Paternal rescue of Ssdp exists; when oocytes that lack Ssdp are fertilized by a wild-type sperm, the Ssdp heterozygotes often survive to eclose as normal adults. These observations show that Ssdp is an essential gene (Chen, 2002).

Given the interactions between the vertebrate Ssdp and Ldb1 proteins, whether their Drosophila homologs might also interact in the context of the whole organism was tested. Previous work has shown that Chip (the Drosophila homolog of Ldb1) forms a dimer capable of binding two molecules of the LIM-homeodomain transcription factor Apterous. The Chip-Apterous tetramer activates transcription of a reporter gene in cultured cells and regulates the transcription of target genes involved in morphogenesis of the wing. Dlmo, a LIM-only protein, competes with Apterous for binding to Chip, and elevated levels of Dlmo lead to the displacement of Apterous from the complex, which renders the complex transcriptionally inactive, causing scalloped wings. Similar wing defects are displayed by homozygous apterous (ap) mutants and by double heterozygotes for mutations in ap and Chip (Chen, 2002).

Double heterozygotes for a Chip mutation and any of the five Ssdp alleles have scalloped wings; all of the single heterozygous mutants have normal wings. This genetic interaction is highly reminiscent of the genetic interaction displayed by double heterozygotes for Chip and ap, and suggests that Chip and Ssdp interact in vivo and that Ssdp is a positive cofactor required for normal function of the Chip-Apterous complex. Bx mutations, hypermorphic alleles of dlmo, also cause scalloped wings. Double heterozygotes for Bx mutations and any of the five Ssdp alleles display marked enhancement of the wing scalloping characteristic of Bx flies. Similar results were observed for several different Bx mutations. A similar enhancement of the wing scalloping of Bx/+ was reported in double heterozygotes for Bx and either Chip or ap mutations. No wing scalloping was observed, however, in double heterozygotes for any of the five Ssdp alleles and either ap⁴ or ap^56f. These observations support the model that Ssdp interacts in vivo with the Chip complex to regulate normal wing development (Chen, 2002).

It is not clear whether the in vivo functions of Ssdp are related to its ability to bind single-stranded DNA in vitro. The results imply that protein-protein interactions are essential for Ssdp function, but this implication does not preclude the possible importance of single-stranded nucleic acid interactions (Chen, 2002). Drosophila embryos defective in both maternal and zygotic Ssdp can develop into larvae. However, the alleles that were studied do not support subsequent development. Ssdp function in the fly may not be restricted to interactions with the Chip–Apterous complex, because the phenotype of the existing Ssdp mutants differs from that of known Chip and ap mutants. Furthermore, the facts that Ssdp genes are so well conserved throughout evolution, that they are expressed in a wide variety of cell types, and that they are indispensable for development seem to suggest that the encoded proteins function in many different transcriptional contexts (Chen, 2002).

Ssdp binds to Chip and regulates the activity of Apterous complexes in vivo

LIM-homeodomain transcription factors control a variety of developmental processes, and are assembled into functional complexes with the LIM-binding co-factor Ldb1 (in mouse) or Chip (in Drosophila). The identification and characterization is described of members of the Ssdp family of proteins, that are shown to interact with Ldb1 and Chip. The N terminus of Ssdp is highly conserved among species and binds a highly conserved domain within Ldb1/Chip that is distinct from the domains required for LIM binding and self-dimerization. In Drosophila, Ssdp is expressed in the developing nervous system and imaginal tissues, and it is capable of modifying the in vivo activity of complexes comprised of Chip and the LIM-homeodomain protein Apterous. Null mutations of the ssdp gene are cell-lethal in clones of cells within the developing wing disc. However, clones mutant for a hypomorphic allele give rise to ectopic margins, wing outgrowth and cell identity defects similar to those produced by mutant clones of Chip or apterous. Ssdp and Ldb/Chip each show structural similarity to two Arabidopsis proteins that cooperate with one another to regulate gene expression during flower development, suggesting that the molecular interactions between Ssdp and Ldb/Chip proteins are evolutionarily ancient and supply a fundamental function in the regulated control of transcription (van Meyel, 2003).

From a yeast two-hybrid screen to identify binding partners for mouse Ldb/NLI proteins, a murine homolog was isolated of avian sequence-specific single-stranded DNA-binding protein (SSDP). First identified in an experimental paradigm for induced differentiation of avian chondrocytes in culture, SSDP has been shown to selectively bind the promoter of the alpha2(I) collagen gene. Two mouse genes encoding highly similar proteins, Ssdp1 and Ssdp2 have been identified (van Meyel, 2003).

Both Ssdp2 and Ldb1 have orthologous counterparts in Drosophila, called Ssdp and Chip. Fly Ssdp residues 1-98 can bind strongly to Chip, and this interaction is dependent upon amino acids 387-426 of Chip. Chip residues 387-435 are 94% identical to Ldb1 amino acids 201-249, and this region has been named the Ldb1/Chip conserved domain (LCCD). Taken together, the results indicate that the N terminus of Ssdp proteins bind Ldb/Chip proteins in a region that is distinct from the two domains needed to form the tetrameric complex, namely the dimerization domain (DD) and the LIM interaction domain (LID) (van Meyel, 2003).

Searches of the NCBI databases indicate that Ssdp proteins comprise a family of highly related proteins in which there are four members in humans, three in mice and only one in Drosophila. Comparisons among primary sequences from Ssdp proteins from these and other species reveal a high degree of amino acid identity, particularly within the first 100 amino acids. Between flies and mice there is 90% identity over the N-terminal region. As is the case for all family members, the remainders of these proteins are characterized by an unusually high proportion of proline, glycine and methionine residues. For example, of the 352 amino acids of Drosophila Ssdp from amino acids 93-445, 21% are proline, 27% are glycine and 9% are methionine, for a total of 57% of all residues. Within this overall architecture, there are three small regions that are highly conserved across species (van Meyel, 2003).

There is also significant similarity in the N terminus of Ssdp family members to LEUNIG, a protein first identified in Arabidopsis and for which the N-terminal domain has been termed a LUFS domain, based on its similarity to other proteins in plants, Flo8 in yeast and to Ssdp. Although the LUFS domain remains functionally uncharacterized to date, it contains a Lissencephaly type 1-like homology motif (LisH) with a curious additional motif comprised at its core of the following sequence P-X-GFL-XX-WW-X-VFWD (van Meyel, 2003).

Like the LUFS domain of Ssdp proteins, the LCCD of Ldb1/Chip has been highly conserved through evolution, with 94% identity between mice and flies over a stretch of 49 amino acids (van Meyel, 2003).

The single ssdp gene in flies consists of two exons, the second of which contains the single open reading frame that encodes a 445 residue polypeptide. Epitope-tagged versions of Drosophila Ssdp were prepared in which five copies of the Myc epitope were fused to the C terminus of full-length Ssdp (SsdpFL) or a mutant lacking amino acids 2-92 (SsdpDelta2-92). These constructs were used to generate transgenic lines in which transgene expression is under the control of UAS sequences. Different GAL4 driver lines were used to express these recombinant proteins in a variety of cell types, including neurons and muscles. SsdpFL localizes to nuclei, with no staining in the cytoplasm. By contrast, SsdpDelta2-92 was found throughout the cytoplasm. Therefore nuclear localization of Ssdp is dependent upon the Chip-interacting LUFS domain, despite the fact that this region does not appear to encode a nuclear localization sequence (NLS). To address whether Chip, which does have an NLS, is required for translocation of Ssdp into the nucleus, tests were perfomed to see whether SsdpFL is properly localized to the nucleus in Chip^e5.5 null mutants. In contrast to wild-type, SsdpFL is distributed throughout the cytoplasm of cells lacking zygotic Chip. Occasionally, staining was detected in nuclei in addition to cytoplasmic staining. This may reflect residual activity in these embryos of maternally provided Chip. These results argue that nuclear targeting of Ssdp occurs through a Chip-dependent mechanism (van Meyel, 2003).

The pattern of ssdp expression was determined using in situ hybridization of digoxigenin-labeled antisense cRNA probes to embryos and third instar larvae. In embryos of syncytial blastoderm stage, ssdp transcript is ubiquitous, suggesting that there is maternal contribution. By the time of germband extension, although still widespread, expression appears to be enriched in the developing central nervous system (CNS). During germband retraction this enrichment of transcript in the embryonic CNS is more apparent, such that by stage 13-14 ssdp expression is largely restricted to the brain and ventral nerve cord. Closer examination of the pattern of expression in the ventral nerve cord suggests that expression occurs in all neurons of the CNS, with no major subclasses excluded. This pattern of expression is maintained through later stages of embryogenesis. In third instar larvae, ssdp is no longer detected in the ventral nerve cord, but moderate ssdp expression is observed in the optic lobes of the brain hemispheres. High levels of ssdp expression are observed in imaginal discs, including the anterior region of the antennal-eye disc, the wing and haltere discs and all leg discs, as well as in the salivary gland. With the exception of the eye-antennal disc, expression in imaginal discs is largely uniform (van Meyel, 2003).

To test the role of Ssdp in vivo, null mutations in the Drosophila ssdp gene were generated. P-element transposition was used to imprecisely excise the P-element EP(3)3097 and chromosomes were generated carrying deletions that were completely lethal in complementation tests with l(3)neo48. Several deletion lines were thus generated, including ssdp^L7 and ssdp^L5. DNA sequencing of the breakpoints of the ssdp^L7 deletion reveals that it is a complete null allele of ssdp and in all analyses where it has been examined, ssdp^L5 has had effects identical to those of ssdp^L7, arguing that it too is a null allele (van Meyel, 2003).

Each of the P-element and deletion alleles was tested in complementation analyses with the others and viability of the progeny was assessed at first and third instar larval stages using marked balancer chromosomes to distinguish homozygotes. The results indicate that the following allelic series exists with respect to increasing severity of the lethal phenotype: ssdp^l(3)neo48<ssdp^EP(3)3097 <ssdp^EP(3)3004<ssdp^L5 and ssdp^L7. In fact, the combination of EP(3)3097 and l(3)neo48 was not lethal in all cases, with 35% of EP(3)3097/1(3)neo48 individuals surviving through eclosion. Interestingly, most of these viable flies displayed mutant phenotypes, including wing blisters, a mild cleft in the notum along the AP axis, and thin, gnarled macrochaetae on the notum (van Meyel, 2003).

Ap is expressed in the dorsal compartment of the wing disc and is required to establish the DV affinity boundary, the wing margin, wing outgrowth and dorsal-specific wing structures such as sensory bristles. In the absence of Ap, the wing fails to develop. Ap functions through a tetrameric complex in which two molecules of Ap are bridged by a homodimer of Chip. Chip mutants interact genetically with ap to cause disruptions of the wing margin, and clones of Chip mutant cells in the wing disc behave like ap mutant clones, causing ectopic wing margins and outgrowths (van Meyel, 2003).

In contrast to a previous study, no phenotypes were detected in simple trans-heterozygous combinations of a null allele of Chip with any ssdp alleles used here, including ssdp^KG03600 and the two null alleles ssdp^L5 and ssdp^L7. Nor were any phenotypes detected in transheterozygous combinations of ap and ssdp. Thus, to address the role of Ssdp in the function of Chip/LIM-HD complexes in vivo, the GAL4-UAS system was used to reduce Ap/Chip complex activity to levels that would be sensitive to the effects of reducing ssdp gene dosage. ap^GAL4, a GAL4 P-element insertion in the ap gene, which faithfully expresses GAL4 in Ap-expressing cells, was used to drive expression of UAS transgenes in the dorsal compartment of the wing disc (van Meyel, 2003).

Over-expression of UAS-Chip disrupts wing patterning by titrating endogenous Ap into incomplete complexes in which LID domains of Chip molecules remain vacant. Relative to controls, such wings are small and lack regular structure, and the wing margin is poorly demarcated. These phenotypes resemble hypomorphic ap mutants, and can be completely suppressed by simultaneous overexpression of UAS-ap. This indicates that the stoichiometry between Ap and Chip is an important factor in the formation of functional complexes. The effect of removing one copy of the ssdp gene was examined; the resulting flies have little or no residual wing tissue, consistent with a further reduction of the activity of the complex (van Meyel, 2003).

Fusion of Chip and Ap into one chimeric molecule, called ChipDeltaLID:ApDeltaLIM, results in a hyperactive complex, since it is not susceptible to downregulation of activity imposed by LMO, a LIM-only factor that competes efficiently with Ap for binding with Chip. Flies that overexpress ChipDeltaLID:ApDeltaLIM have blistered wings in which the dorsal and ventral surfaces fail to fuse, and which are held upward and away from the thorax in a fashion resembling LMO loss-of-function mutants. Removal of one copy of ssdp suppresses the blistered wing phenotype and the surfaces fuse properly, although the wings remain held up. Thus, Ssdp can modify the activity of Chip/Ap tetrameric complexes of both reduced and hyperactive function (van Meyel, 2003).

Finally, the effects of Chip overexpression were compared with those produced by expression of a Chip variant lacking the LCCD (ChipDeltaLCCD). ChipDeltaLCCD is capable of self-dimerization and binding to Ap, but it cannot bind Ssdp. If Ssdp were required for function of the complex, ChipDeltaLCCD would be predicted to have a more potent dominant-negative effect on the function of the complex than would Chip itself, since the latter can still recruit Ssdp. Expression of ChipDeltaLCCD with ap^GAL4 consistently produced more extreme wing defects than Chip. ChipDeltaLCCD sequesters Ap into nonfunctional complexes, but it cannot bind Ssdp. Therefore, removal of one copy of ssdp would not be expected to suppress the phenotype caused by ChipDeltaLCCD, and indeed it does not. Collectively, these results argue that in addition to forming the dimeric bridge for two molecules of Ap, Chip also recruits Ssdp to the complex (van Meyel, 2003).

Clones of ap mutant cells in the dorsal compartment of the wing disc induce an ectopic wing margin and therefore ectopic wing outgrowth. These ap mutant cells differentiate ventral wing margin structures, despite the fact that they remain in the dorsal compartment. Chip mutant clones induced in the dorsal compartment give rise to strikingly similar phenotypes. The effects of Chip clones are influenced both by the timing of their induction as well as their position within the disc. For example, clones induced later (third instar) result in ectopic margin tissue, but do not lead to outgrowth (van Meyel, 2003).

If Ssdp were an additional member of the Ap/Chip complex, then mutations of ssdp would be predicted to give rise to mutant phenotypes similar to those of ap and Chip. To test this, the FRT/FLP recombinase system was used to induce clones of cells mutant for ssdp in an otherwise heterozygous animal. Clones were generated in larvae at second and third instar by heat-shock induction at 36 hours, 48 hours, 72 hours or 96 hours after egg laying (AEL). The effects of clone induction were observed in newly eclosed adults. Clones of mutant cells were identified by the presence of the cell-autonomous marker pawn (pwn). Each of the mutant alleles ssdp^L7, ssdp^L5 and ssdp^l(3)neo48 were tested, as was a control chromosome with no mutation, and the experiment was repeated on four separate occasions, each time observing many individuals of each genotype (van Meyel, 2003).

In controls, many clones of various sizes were induced, as evidenced by the presence of pwn mutant cells. These clones occurred on both the ventral and dorsal surfaces of the wing blade, but no mutant phenotypes were ever observed. By contrast, clones of cells mutant for either ssdp^L7 or ssdp^L5 (as marked by pwn) were never observed on either surface of the wing blade, indicating that both alleles have cell-lethal effects in the wing disc. In addition, there were fewer than the expected number of adults eclosing of the appropriate genotype for clone induction, suggesting that the cell-lethal effects, presumably in tissues other than the wing, led to decreased viability (van Meyel, 2003).

In contrast to the cell lethality associated with ssdp null alleles, there were striking phenotypes observed in clones of cells mutant for the hypomorphic ssdp^l(3)neo48 allele. Many pwn mutant clones located both ventrally and dorsally were observed. However, as for ap and Chip clones, associated phenotypes were found only when the clone arose on the dorsal surface of the wing. ssdp^l(3)neo48 clones induced earlier (at 36 hours and 48 hours AEL) give rise to ectopic margins and occasional wing outgrowth. The outgrowths were associated with ssdp mutant cells but were not entirely made up of them, indicating that, as for ap and Chip, outgrowth results from the induction of wild-type tissues in proximity to the clone. Clones induced at 72 hours and 96 hours AEL give rise to margin defects but not outgrowth, indicating that there is a temporal restriction on the extent to which ssdp mutation is capable of inducing outgrowth, similar to what has been shown for Chip (van Meyel, 2003).

Induction of ectopic margin bristles was the most commonly observed effect of dorsal ssdp mutant clones. They were primarily observed in proximity to a clone near the native anterior wing margin and comprised at least one row of extra sensory bristles. Most ectopic bristles were not marked by pwn, indicating they were induced by the neighboring mutant (pwn) cells. ssdp^l(3)neo48 mutant clones that occurred within the margin, rather than near it, resulted in the loss of dorsal-specific sensory bristles. Occasionally a large clone was observed to straddle the dorsoventral boundary, and in these instances, the entire margin, including some nearby non-margin tissue, was lost (van Meyel, 2003).

In general, there is a striking resemblance between the phenotypes resulting from ssdp^l(3)neo48 mutant clones and those reported for clones of Chip or ap. This provides strong evidence that Ssdp is an important additional component of Chip/Ap transcriptional complexes in vivo (van Meyel, 2003).

Enhancer-promoter communication mediated by Chip during Pannier-driven proneural patterning is regulated by Osa

The GATA factor Pannier activates proneural achaete/scute (ac/sc) expression during development of the sensory organs of Drosophila through enhancer binding. Chip bridges Pannier with the (Ac/Sc)-Daughterless heterodimers bound to the promoter and facilitates the enhancer-promoter communication required for proneural development. This communication is regulated by Osa, which is recruited by Pannier and Chip. Osa belongs to Brahma chromatin remodeling complexes, and this study shows that Osa negatively regulates ac/sc. Consequently, Pannier and Chip also play an essential role during repression of proneural gene expression. This study suggests that altering chromatin structure is essential for regulation of enhancer-promoter communication (Heitzler, 2003).

Chip^E is a viable allele of Chip that is associated with a point mutation in the LIM-interacting domain (LID), which specifically reduces interaction with the bHLH proteins Ac, Sc, and Da. As a consequence, the Chip^E mutation disrupts the functioning of the proneural complex encompassing Chip, Pnr, Ac/Sc, and Da. A homozygous Chip^E mutant shows thoracic cleft and loss of the DC bristles, similar to loss of function pnr alleles (Heitzler, 2003).

To identify new factors that regulate this proneural complex, a screen was performed for second-site modifiers of the Chip^E phenotypes. One allele of osa (osa^E17) was found among the putative mutants. Osa^E17 corresponds to a loss-of-function allele, and homozygous embryos die with normal cuticle patterning. Both osa^E17 and null alleles of osa (osa⁶¹⁶ or osa¹⁴⁰⁶⁰) enhance the cleft but suppress the loss of DC bristle phenotypes of Chip^E flies. Indeed, Chip^E flies with only one copy of osa⁺ (Chip^E;osa⁶¹⁶/+) are weak and sterile but show wild-type DC bristle pattern (Heitzler, 2003).

These genetic interactions suggest that Osa can antagonize the function of Pnr. Moreover, overexpressed Osa (+/UAS-osa;Gal4-pnr^MD237/+) induces a thoracic cleft and the loss of DC bristles similar to the loss-of-function pnr alleles. In contrast, loss-of-function osa alleles display an excess of DC bristles similar to overexpressed Pnr. For example, (osa¹⁴⁰⁶⁰/+), (osa⁶¹⁶/+), and (osa^E17/+) flies exhibit respectively 2.35 ± 0.12, 2.38 ± 0.12, and 2.43 ± 0.17 DC bristles per heminotum (Oregon wild-type flies have 2.00 DC bristles/heminotum). Furthermore, transallelic combination of osa¹⁴⁰⁶⁰ with the hypomorphic osa^4H (osa^4H/osa¹⁴⁰⁶⁰) accentuates the excess of DC bristles compared with (osa¹⁴⁰⁶⁰/+). (osa^4H/osa¹⁴⁰⁶⁰) flies display 4.17 ± 0.19 DC bristles per heminotum. In contrast, (osa^4H/osa^4H) flies display 2.50 ± 0.11 DC bristles per hemithorax. The development of the extra DC bristles revealed by phenotypic analysis was compared with the positions of the DC bristle precursors detected with a LacZ insert, A101, in the neuralized gene that exhibits staining in all sensory organs. In (osa¹⁴⁰⁶⁰/osa^4H) discs, additional DC precursors are observed that lead to the excess of DC bristles. The pnr^D alleles encode Pnr proteins carrying a single amino acid substitution in the DNA binding domain that disrupts interaction with the U-shaped (Ush) antagonist. Consequently, Pnr^D constitutively activates ac/sc, leading to an excess of DC bristles. This excess is accentuated when osa function is simultaneously reduced (pnr^D1/osa⁶¹⁶) (Heitzler, 2003).

Since osa shows genetic interactions with trithorax group genes encoding components of the Brm complex like moira (mor) and brm, whether mutations in mor and brm suppress the Chip^E phenotype was investigated. Loss of one copy of brm⁺ in (Chip^E; brm²/+) flies suppresses the lack of DC bristles observed in Chip^E flies, similar to loss of one copy of osa⁺. This shows that brm and osa both act during Pnr-dependent patterning, in agreement with the fact that they have been shown to be associated in the Brm complex. In contrast, reducing the amount of Mor by half [(Chip^E;mor¹/+) flies] is not sufficient to modify the Chip^E phenotype. This does not definitely exclude the possibility that mor is directly or indirectly involved, via the Brm complex, in Pnr-dependent patterning (Heitzler, 2003).

The complete osa open reading frame of 2715 amino acids and the intronic splicing signals were PCR amplified from genomic DNA prepared from homozygous embryos (osa^E17 and osa¹⁴⁰⁶⁰) and homozygous first instar larvae (osa^4H). For osa¹⁴⁰⁶⁰ and osa^4H, the sequence analysis revealed a single mutation in the N terminus that causes a glutamine to stop codon substitution. The conceptual translation of osa¹⁴⁰⁶⁰ leads to a truncated Osa protein lacking both functional domains, whereas Osa^4H retains the ARID domain but lacks the C-terminal EHD. Wild-type osa function is necessary for patterning of the DC bristles. Although osa^E17 behaves as a stronger allele than osa¹⁴⁰⁶⁰ and osa^4H, molecular identity of the mutation is unknown. Hence, the osa^E17 phenotype may result from a mutation in regulatory sequences that affects osa expression (Heitzler, 2003).

It has been shown that a complex containing Pnr, Chip, and the (Ac/Sc)-Da heterodimer activates proneural expression in the DC proneural cluster and promotes development of the DC macrochaetae. Osa and Pnr/Chip have antagonistic activities during development because loss of osa function (osa^4H and osa¹⁴⁰⁶⁰) displays additional DC bristles. However, since the current study reveals that osa genetically interacts with pnr and Chip, it was asked whether Osa physically interacts with the Pnr and Chip proteins. Immunoprecipitations of protein extracts made from Cos cells cotransfected with expression vectors for tagged Osa and either Pnr or tagged Chip were immunoprecipitated. Because Osa is a large protein, several expression vectors encoding contiguous domains of Osa were used. Osa coimmunoprecipitates with Pnr and Chip and can be detected on Western blots with appropriate antibodies. The interactions appear to require the overlapping domains Osa E (His1733/Glu2550) and Osa F (Ala2339/Ala2715) corresponding to the EHD. Enhancer-promoter communication during proneural activation and development of the DC bristles requires regulatory sequences scattered over large distances and appears to be negatively regulated by interaction of Pnr and Chip with Osa through the EHD. Interestingly, the EHD is not conserved in yeast. In yeast, the UAS sequences are generally close to the promoter and there is no requirement for long-distance interactions. This observation could support the idea that the EHD is essential for long-distance enhancer-promoter communication. Alternatively, yeast may just lack proteins like Chip or Pnr (Heitzler, 2003).

The DNA-binding domain and the C-terminal region are essential for the function of Pnr during development of the DC sensory organs. The pnr^VX1 and pnr^VX4 alleles (collectively pnr^VX1/4) are characterized by frameshift deletions that remove two C-terminal alpha-helices and result in reduced proneural expression and loss of DC bristles (Heitzler, 2003).

The molecular interactions between Osa and Pnr^D1 and between Osa and Pnr^VX1 were investigated. Pnr^D1 protein interacts with the EHD as efficiently as wild-type Pnr. In contrast, the physical interaction is disrupted when the C terminus of Pnr encompassing the alpha-helices is removed. Because the C terminus of Pnr is required for the Pnr-Osa interaction in transfected cells extracts, the abilities of in vitro translated 35S-labeled Osa domains to bind to GST-C^TPnr attached to glutathione-bearing beads were investigated. Only Osa E and Osa F interact with the C terminus of Pnr. The interaction between Chip and Osa, and it was found that Osa associates with the N-terminal homodimerization domain of Chip, also required for the interaction between Chip and Pnr, was investigated. Furthermore, Osa E and Osa F also bind to immobilized GST-Chip. Deletion of the alpha helix H1 disrupts the interactions between Pnr and Osa. Interestingly, the same deletion also disrupts the interaction with Chip. Therefore, the functional antagonism between Chip and Osa during neural development may result from a competition between these proteins for association with Pnr. Alternatively, the deletion of H1 may affect the overall structure of the C terminus of Pnr and disrupt the physical interactions with Chip and Osa. To discriminate between these hypotheses, immunoprecipitations of protein extracts containing a constant amount of Pnr, a constant amount of the tagged Osa E domain, and increasing concentrations of Chip were performed. Pnr immunoprecipitates with immunoprecipitated tagged Osa E and the amount of Pnr immunoprecipitated increases in the presence of increasing concentrations of Chip. The presence of increasing amounts of Chip does not inhibit the Osa-Pnr interaction as would be expected if Osa and Chip were to compete for binding to Pnr. In contrast, it suggests that Chip and Pnr act together to recruit Osa and to target its activity and possibly the activity of the Brm complex to the ac/sc promoter sequences (Heitzler, 2003).

Using expression vectors encoding contiguous domains of Osa, it was shown that the EHD of Osa mediates interactions with Pnr and Chip. Because the EHD is lacking in the truncated Osa¹⁴⁰⁶⁰ and Osa^4H, it is hypothesized that the loss of interaction with Pnr and Chip are responsible for the excess of DC bristles observed in osa^4H and osa¹⁴⁰⁶⁰ (Heitzler, 2003).

To investigate whether these interactions between Osa, Pnr, and Chip function in vivo during DC bristle development, the effects of both loss of function and overexpression of osa were examined on the activity of a LacZ reporter whose expression is driven by a minimal promoter sequence of ac fused to the DC enhancer (transgenic line DC:ac-LacZ). It was found that expression of the LacZ transgene is increased in osa¹⁴⁰⁶⁰/osa^4H wing discs when compared with the wild-type control. For overexpression experiments, the UAS/GAL4 system was used, using as a driver the pnr^MD237 strain that carries a GAL4-containing transposon inserted in the pnr locus (driver: pnr-Gal4). This insert gives an expression pattern of Gal4 indistinguishable from that of pnr. It was found that overexpressed Osa leads to a strong reduction of LacZ staining in the DC area, consistent with the lack of DC bristles. Thus, overexpressed Osa represses activity of the ac promoter sequences required for DC ac/sc expression and development of the DC bristles. It has been previously reported that wingless expression is also required for patterning of the DC bristles. However, the repressing effect of Osa on development of the DC bristles is unlikely to be the result of an effect of Osa on wingless expression because overexpressed Osa driven by pnr^MD237 has no effect on the expression of a LacZ reporter inserted into the wingless locus. Thus, Osa acts through the DC enhancer of the ac/sc promoter sequences to repress ac/sc and neural development (Heitzler, 2003).

Chip^E disrupts the enhancer-promoter communication and strongly affects expression of the LacZ reporter driven by the ac promoter linked to the DC enhancer. Because null alleles of osa suppress the loss of DC bristles displayed by Chip^E, the consequences of reducing the dosage of osa was examined in Chip^E flies. The expression of the LacZ reporter is not affected in Chip^E flies when Osa concentration is simultaneously reduced (Heitzler, 2003).

In conclusion, Pnr function during proneural patterning is regulated by interaction with several transcription factors. Pnr function is negatively regulated by Ush, which interacts with its DNA-binding domain. Chip associates with the C terminus of Pnr, bridging Pnr at the DC enhancer with the AC/Sc-Da heterodimers bound at the proneural promoters, thus activating proneural gene expression. The current study reveals that Pnr function can also be regulated by interaction with Osa. Thus, Osa activity is specifically targeted to ac/sc promoter sequences and the binding of Osa therefore has a negative effect on Pnr function, leading to reduced expression of the proneural ac/sc genes. Osa belongs to Brm complexes, which are believed to play an essential role during chromatin remodeling necessary for gene expression. For example, in vitro transcription experiments with nucleosome assembled human beta-globin promoters have shown that the BRG1 and BAF155 subunits of the mammalian SWI/SNF homolog are essential to target chromatin remodeling and promote transcription initiation mediated by GATA-1. In contrast to what was observed in vitro, the current results suggest that in vivo the SWI/SNF complexes can also act to remodel chromatin in a way that represses transcription. Alternatively, the observed repression of proneural genes may simply define a novel function of Osa, independent of chromatin remodeling (Heitzler, 2003).

Osa modulates the expression of Apterous target genes in the Drosophila wing

The establishment of the dorsal-ventral axis of the Drosophila wing depends on the activity of the LIM-homeodomain protein Apterous. Apterous activity depends on the formation of a higher order complex with its cofactor Chip to induce the expression of its target genes. Apterous activity levels are modulated during development by dLMO (Beadex). Expression of dLMO in the Drosophila wing is regulated by two distinct Chip dependent mechanisms. Early in development, Chip bridges two molecules of Apterous to induce expression of dLMO in the dorsal compartment. Later in development, Chip, independently of Apterous, is required for expression of dLMO in the wing pouch. A modular P-element based EP (enhancer/promoter) misexpression screen was conducted to look for genes involved in Apterous activity. Osa, a member of the Brahma chromatin-remodeling complex, was found to be a positive modulator of Apterous activity in the Drosophila wing. Osa mediates activation of some Apterous target genes and repression of others, including dLMO. Osa has been shown to bind Chip. It is proposed that Chip recruits Osa to the Apterous target genes, thus mediating activation or repression of their expression (Milan, 2004).

This study presents evidence that Osa, a member of a subset of Brahma chromatin remodeling complexes, behaves overall as a general activator of Apterous activity in the Drosophila wing. Overexpression of Osa rescues and loss of Osa enhances the Beadex¹ phenotype. It does so by modulating the expression levels of Apterous target genes, some of them being activated (e.g. Serrate and probably other unknown target genes) and some repressed (e.g. Delta, fringe). Chip has been shown to bind Osa. The fact that Osa has different effects on the transcription of Apterous target genes suggests that Chip recruits Osa to the promoters and in combination with other unknown factors mediates either transcriptional repression or activation. Osa mediates repression of both Apterous dependent and independent expression of fringe, suggesting a direct and probably Chip independent effect of Osa on fringe transcription (Milan, 2004).

Apterous activity is regulated during development by dLMO. Osa is required to mediate repression of dLMO expression. Since both early and late expression of dLMO depend on Chip, it is postulated that Chip forms a transcriptional complex with Apterous in the D compartment and an unknown transcription factor expressed in the wing pouch. Osa may interact with Chip thus recruiting the Brahma complex to the dLMO locus and remodeling chromatin in a way that limits dLMO transcriptional activation. High levels of dLMO protein reduce Apterous activity and the Notch dependent organizer is not properly induced along the DV boundary. Osa mediated repression of dLMO expression may ensure moderate levels of expression of dLMO in the wing, thus allowing proper wing development. Gain of function mutations that cause misexpression of vertebrate LMO proteins have been implicated in cancers of the lymphoid system. Truncating mutations in the human SWI-SNF complex, the human homologues of the Brahma complex, cause various types of human cancers. The SWI-SNF complex may be required to mediate repression of LMO expression in lymphoid tissues. Thus, it would be very interesting to analyze if truncating mutations in members of the human SWI-SNF complex cause higher levels of LMO expression and are associated with lymphoid malignancies (Milan, 2004).

It has been shown that the Brahma complex plays a general role in transcription by RNA Polymerase II. Then, is Osa having a general effect on the expression levels of every gene involved in wing patterning? Several observations indicate this is not the case. (1) Osa is a component of a subset of Brahma (Brm) chromatin complexes. (2) Brahma and Polycomb were shown to have non-overlapping binding patterns in polytenic chromosomes. Those genes involved in wing patterning and regulated by Polycomb (i.e. Hedgehog) may not be affected by overexpression of Osa. (3) Overexpression of Osa has different effects on the expression levels of Serrate, Delta and fringe. (4) Osa has been shown to specifically regulate the expression of Wingless target genes and the Achaete-scute complex genes, interestingly by restricting their expression levels (Milan, 2004).

Chip-mediated partnerships of the homeodomain proteins Bar and Aristaless with the LIM-HOM proteins Apterous and Lim1 regulate distal leg development

Proximodistal patterning in Drosophila requires division of the developing leg into increasingly smaller, discrete domains of gene function. The LIM-HOM transcription factors apterous (ap) and Lim1 (also known as dlim1), and the homeobox genes Bar and aristaless (al) are part of the gene battery required for the development of specific leg segments. Genetic results show that there are posttranslational interactions between Ap, Bar and the LIM-domain binding protein Chip in tarsus four, and between Al, Lim1 and Chip in the pretarsus, and that these interactions depend on the presence of balanced amounts of such proteins. In vitro protein binding is observed between Bar and Chip, Bar and Ap, Lim1 and Chip, and Al and Chip. Together with evidence for interactions between Ap and Chip, these results suggest that these transcription factors form protein complexes during leg development. It is proposed that the different developmental outcomes of LIM-HOM function are due to the precise identity and dosage of the interacting partners present in a given cell (Pueyo, 2004).

Biochemical studies in vitro have shown that LIM-HOM transcription factors confer little transcriptional activation of target genes on their own. LIM-HOM proteins interact with a variety of proteins, including members of the bHLH family, the POU family and also other LIM family members, to control specific developmental processes. It has been suggested that these protein interactions confer specificity and modulate LIM-HOM activity. For example, Dlmo proteins reduce LIM-HOM activity, and Lbd proteins such as Chip modulate LIM-HOM activity by acting as a bridge between LIM-HOM proteins and Chip-binding cofactors, thus leading to the formation of heteromeric complexes. LIM-HOM protein activity functions in different contexts is the development of Drosophila (Pueyo, 2004).

Bar and ap genes are expressed in the fourth tarsal segment and are required for its proper development, whereas the al and Lim1 genes are expressed and required in the pretarsus. All of these genes encode putative transcription factors and display canonical regulatory relationships. Thus, al activates lim1 expression and then both genes cooperate to repress Bar expression in the pretarsus. Reciprocally, Bar represses al and lim1 expression while activating the expression of ap in tarsus four. After the refinement of their gene expression domains by these regulatory interactions, Bar directs tarsus five development, whereas cooperation between al and lim1 directs pretarsus development, and cooperation between Bar and ap directs tarsus four. The results of this study offer more evidence for the existence of this regulatory network, but also suggest an interesting role for direct protein interactions in its mechanism (Pueyo, 2004).

The cooperation between Bar and Ap on the one hand, and Al and Lim1 on the other, is likely to be carried out by transcriptional complexes involving Chip. The Chip protein is required for development of the tarsus four, five and pretarsus, and Gst (Glutathione-S-transferase-Chip fusion construct) experiments reveal Chip's ability to bind Ap, Bar, Lim1 and Al. However, the results also show that modulation of LIM-HOM protein activity by Chip alone does not explain distal leg development. For example, Ap function is not modulated primarily by Chip and Dlmo. The relative amount of Chip and Ap has to be grossly unbalanced before a phenotype is obtained in the leg, and dlmo is not expressed or required in leg development. Furthermore, the interaction between Ap and Chip does not confer the developmental specificity that allows LIM-HOM proteins to produce different outcomes in different parts of the leg. (1) Ap and Chip also interact in the wing and the CNS. (2) A chimaeric Lim3-Ap protein containing the LIM domains of Lim3 and the HOM domain of Ap does not behave as a dominant negative when expressed in tarsus four, and is even able to fulfil Ap function and rescue ap mutants. In the distal leg, developmental specificity seems to be achieved at the level of DNA binding and the transcriptional control of targets genes, mediated by partnerships between LIM-HOM and HOM proteins (Pueyo, 2004).

The evidence for this is presented first by dosage interactions between LIM-HOM and HOM proteins. Whereas there seems to be a relative abundance of endogenous Ap in tarsus four, an excess of Bar or Chip leads to a mutant phenotype, which is rescued by restoring the normal balance between Ap, Bar and Chip proteins in co-expression experiments. The effects observed could be explained simply by independent competition and the binding of Bar and Ap to Chip, leading, for example, to an excess of Bar-Chip complexes and a reduction of the pool of Chip available for Ap-Chip complexes. However, this hypothesis alone does not explain the additional dominant-negative effects of ectopic LIM-HOM and HOM proteins in tarsus four (Lim3, Islet and Al), which are also not mediated by transcriptional regulation but are nonetheless rescued by co-expression of appropriate endogenous proteins. For example, ectopic expression of UAS-islet or UAS-Lim3 in the ap domain produces loss of tarsus four without affecting Ap or Bar expression, and simultaneous co-expression of UAS-Bar partially suppresses this phenotype. If the sole effect of both UAS-Bar and UAS-Lim3 or UAS-islet were to quench Chip away from Ap, then simultaneous co-expression of Bar and Lim3 or Islet should worsen the phenotype, not correct it as observed. Moreover, ectopic expression of Islet or Lim3 proteins is not corrected by simultaneous co-expression of either UAS-Chip or UAS-ap. Altogether these results show instead that UAS-islet and UAS-Lim3 must interfere posttranslationally with Bar. The most direct explanation is that Islet and Lim3 have the ability to quench Bar protein into a non-functional state. Interestingly, the hybrid UAS-Lim3:ap does not behave as dominant negative but as an endogenous Ap protein in these experiments, since it does not produce a mutant phenotype on its own and it rescues UAS-Bar overexpression. This suggests that the LIM domains are not very specific when it comes to interaction with Bar, and points to the involvement of a common LIM-binding intermediary such as Chip. These results suggest that a protein complex involving Ap, Chip and Bar is the correct functional state of these proteins in tarsus four, and deviations from this situation into separate Bar-Chip, Ap-Chip, or Bar-Chip-Lim3 or Bar-Chip-Islet complexes leads to a mutant phenotype (Pueyo, 2004).

The notion of a protein complex involving Ap, Chip and Bar together is also supported by the Gst pull-down assays. The domain of Chip involved in Ap binding, the LIM interaction domain (LID), is not involved in Bar binding. However, the LID and the dimerisation domains of Chip are necessary to rescue the dominant-negative effect of UAS-Bar on tarsus four, suggesting a requirement for the formation of a complex with a LIM-HOM protein such as Ap. In agreement with this view, the Ap protein, and the LIM domains of Ap alone, are able to retain Bar protein in a Gst assay (Pueyo, 2004).

In the pretarsus, Al and Lim1 are possibly engaged in a partnership with Chip similar to that suggested for Ap, Chip and Bar. Synergistic cooperation between Al and Lim1 is required to direct pretarsus development and to repress Bar expression and function. Their cooperation entails a close functional relationship because a proper balance of Al, Lim1 and Chip is required, as is shown by the loss of pretarsal structures in UAS-Chip or UAS-Lim1 flies. Ectopic expression of LIM-HOM proteins in the pretarsus also disrupts pretarsal development without affecting Lim1 and Al expression. The possibility of direct protein interactions between Al, Lim1 and Chip is also suggested by the reciprocal ability of Al to interfere posttranscriptionally with Bar and Ap in tarsus four, and by the binding of Chip to Lim1 and to Al in in vitro experiments (Pueyo, 2004).

Comparison of tarsal development with other developmental processes illustrates how LIM-HOM proteins are versatile factors to regulate developmental processes. It had been observed that the outcome of LIM-HOM activity depends on their developmental context. This context can now be analysed as being composed of the presence, concentration and relative affinities of other LIM-HOM proteins, Ldb adaptors, and other cofactors such as LMO proteins and HOM proteins. It is proposed that the different developmental outcomes of LIM-HOM protein function could be due to the precise identity and dosage of cofactors available locally (Pueyo, 2004).

Ectopic expression experiments distort these contexts and lead to non-functional or misplaced LIM-HOM activities. In the wing, a finely balanced amount of functional Ap protein is modulated by Dlmo and Chip. Over-abundance of Chip stops the formation of functional tetramers in the wing but not in the CNS, where the relative amount of Ap, which is not modulated by Dlmo, is limiting for the formation of Ap-Chip functional complexes. In tarsus four, the Ap-Chip-Bar partnership is affected by experimentally induced over-abundance of Chip, presumably also because ectopic Ap-Chip tetramers typical of the CNS and the wing, and Bar-Chip complexes typical of tarsus five, are produced. Similarly, an excess of Bar might be interpreted by the cells as being a wrong developmental outcome, since high levels of Bar in the absence of Ap direct tarsus five development. Overexpression of Ap rescues this Bar dominant-negative effect, by restoring the relative amounts of Bar and Ap, which are determinant and limiting for tarsus four development. Finally, the dominant-negative effects produced by overexpression of either Chip or Lim1 in the pretarsus could either prevent the formation of Al-Chip-Lim1 complexes, or could favor the existence of Lim1-Chip complexes typical of the CNS (Pueyo, 2004).

The wing and the CNS models have postulated that Ap function is carried out by an Ap-Chip tetramer; however, the molecular scenario might be more complex. A new component of Ap-Chip complexes, named Ssdp, has been identified and is required for the nuclear localisation of the complex. Thus it is possible that an Ap-Chip tetramer also contains two molecules of Ssdp. In addition, different types of Chip-mediated transcriptional complexes and different regulators have been identified in other developmental contexts, such as in sensory organ development and thorax closure, in which the GATA factor Pannier forms a complex with Chip and with the bHLH protein Daughterless. Heterodimers of this complex are negatively regulated by a protein interaction with Osa. Thus, although the current results indicate that in different segments of the leg there exist specific interactions between LIM-HOM, Chip and HOM proteins, the involvement of further elements in these multiprotein complexes is not excluded (Pueyo, 2004).

The results support a partnership between HOM and LIM-HOM proteins in the specification of distinct segments of the leg, and the results are compatible with Ap-Chip-Bar, Bar-Chip and Lim1-Chip-Al forming transcriptional complexes. Although the characterisation of the target sequences, followed by further biochemical and molecular assays, is necessary to study the transcriptional mechanism of these interactions, it has been shown that LIM-HOM proteins can interact specifically and directly with other transcription factors to regulate particular genes. For instance, mouse Lim1 (Lhx1) interacts directly with the HOM protein Otx2. In addition, the bHLH E47 transcription factor interacts with Lmx1, and both synergistically activate the insulin gene. This interaction is specific to Lmx1, since E47 is unable to interact with other LIM-HOM proteins such as Islet. Moreover, Chip is able to bind to other Prd-HOM proteins, such as Otd, Bcd and Fz, to activate downstream genes. Chip also complexes with Lhx3 and the HOM protein P-Otx, increasing their transcriptional activity. The current results reinforce the notion of Chip as a multifunctional transcriptional adaptor that has specific domains involved in each interaction (Pueyo, 2004).

Experiments in Drosophila have demonstrated a conservation of LIM-HOM activity at the functional and developmental level in the CNS between Drosophila and vertebrates. In addition, xenorescue experiments have shown that the mechanism of action of Ap and its vertebrate homolog Lhx2 is very conserved in Drosophila wings, whereas ectopic expression of dominant-negative forms of chick Lim1, Chip, Ap and Lhx2 mimic both Ap and Lhx2 loss-of-function phenotypes. The developmental role of Ap, Bar and Al in the fly leg, and their putative molecular interactions may also have been conserved because their vertebrate homologs Lhx2, Barx and Al4 are also co-expressed in the limb bud. It is expected that the interactions between the LIM-HOM and Prd-HOM proteins shown here represent a conserved mechanism to specify different cellular fates during animal development (Pueyo, 2004).

The Drosophila LIM-homeodomain protein Islet antagonizes proneural cell specification in the peripheral nervous system: Isl antagonizes Pnr activity both by dimerization with the DNA-binding domain of Pnr and via competitive inhibition of the Chip-bHLH interaction

The pattern of the external sensory organs (SO) in Drosophila depends on the activity of the basic helix-loop-helix (bHLH) transcriptional activators Achaete/Scute (Ac/Sc) that are expressed in clusters of cells (proneural clusters) and provide the cells with the potential to develop a neural fate. In the mesothorax, the GATA1 transcription factor Pannier (Pnr), together with its cofactor Chip, activates ac/sc genes directly through binding to the dorsocentral enhancer (DC) of ac/sc. The LIM-homeodomain (LIM-HD) transcription factor Islet (Isl) was identified by genetic screening and its role in the thoracic prepatterning was investigated. isl loss-of-function mutations result in expanded Ac expression in DC and scutellar (SC) proneural clusters and formation of ectopic sensory organs. Overexpression of Isl decreases proneural expression and suppresses bristle development. Moreover, Isl is coexpressed with Pnr in the posterior region of the mesothorax. In the DC proneural cluster, Isl antagonizes Pnr activity both by dimerization with the DNA-binding domain of Pnr and via competitive inhibition of the Chip-bHLH interaction. It is proposed that sensory organ prepatterning relies on the antagonistic activity of individual Chip-binding factors. The differential affinities of these binding-factors and their precise stoichiometry are crucial in specifying prepatterns within the different proneural clusters (Biryukova, 2005).

During Drosophila development, the expression of transcription factors divides the dorsal thorax into three domains -- one median and two lateral domains. The lateral domains are specified by the homeobox-containing proteins of the iroquois-complex (iro), whereas the GATA factor Pnr is required to establish the median domain. Within the mesothorax, Pnr together with U-shaped (Ush) and Chip plays a key role in dorsal closure. This report presents evidence that Isl is an essential regulator of the dorso-median patterning of the thorax. isl⁻ clones generated adjacent to the thoracic midline, induce a strong cleft, suggesting that Isl is required for proper dorsal closure during metamorphosis. Ectopic expression of Pnr leads to wing-to-thorax transformations, consistent with its role as medio-dorsal patterning factor. Ectopic Isl expression does not exhibit this phenotype, excluding the LIM-HD factor from a direct function as a prothoracic selector. Pnr is also known to activate wingless (wg) in dorsal thorax. isl loss-of-function has no significant effect on wg expression. However, overexpressed Isl strongly reduces the size of the wg thoracic stripe. This result is consistent with a repressive activity of Isl on Pnr (Biryukova, 2005).

Iro proteins and Pnr are direct activators of the proneural genes in their respective domains. Pnr binds directly to the DC enhancer of ac/sc, providing therefore region-specific control of the proneural prepatterning. Flies with reduced or lack of Pnr function fail to activate ac/sc and to develop DC and SC sensory organs. The proneural activity of Pnr is antagonized by Ush, the vertebrate homologue of the FOG (friend of GATA). Ush is expressed only in the dorsal-most cells of the medial region. As a consequence, the segregation of the sensory organ precursors occurs along two stripes at the border of the medial domain of Pnr expression, where Ush is absent or insufficient to repress Pnr (Biryukova, 2005).

Several lines of evidence indicate that Isl interferes with the proneural activity of Pnr as a repressor. (1) isl loss-of-function mutants show an opposite phenotype with regard to Pnr or Chip loss-of-function mutants: an excess of DC and SC sensory organs. (2) A genetic synergism exists between Pnr^D and isl⁻ alleles. This genetic interaction is less sensitive than that between Pnr^D and ush⁻, implying an alternative route for Isl to modulate the Pnr proneural activity. (3) Isl is coexpressed with Pnr within the posterior mesothorax. (4) Isl modulates the activity of a DC:ac-lacZ reporter. Loss-of-function isl mutants expand the DC:ac-lacZ expression as in ush⁻ or Pnr^D constitutive mutants, whereas overexpressed Isl reduces the DC:ac-lacZ expression (Biryukova, 2005).

In the DC region, the regulation of Pnr concentration is critical for the proper position and shape of the DC proneural cluster. Isl expression overlaps with the dorsal-most domain of Pnr and DC proneural activity coincides with the posterior border of Isl expression. Therefore, it proposed that both Isl and Ush restrict Pnr activity in the mesothorax. Interestingly, the regulation of the concentration of the mammalian Pnr ortholog, GATA-1, is similarly critical for proper erythroid, megakaryocytic, eosinophilic and mast cell lineages (Biryukova, 2005).

Ush behaves as either an activator or a repressor of Pnr, depending on developmental context. No evidence was found for a direct Isl-Ush interaction by GST pull down assay: Ush, Pnr and Isl could be co-immunoprecipitated from transient transfected S2 cells. Both Ush and Isl may behave as positive cofactors of Pnr for nonneural activities, such as cardiac development, embryonic dorsal closure and metamorphosis. Several reports emphasize the role of the Pnr homolog, GATA-1 and Isl1 in human blood disorders. It seems likely that GATA:Islet interactions represent a conserved mechanism to specify different cell fates in humans and other organisms (Biryukova, 2005).

Isl proteins are known as positive regulators of transcription in vertebrates. In flies, Isl mediates repression of Pnr-driven proneural activity via binding to the DNA-binding domain of Pnr. Interestingly, these interactions are less specific than for the Pnr-Ush interaction, where the amino-terminal zinc finger of Pnr is specifically involved (Biryukova, 2005).

Genetic analyses of mutants reveal that the DC and the SC proneural clusters show differential sensitivities during neurogenesis. Ush mutants display ectopic DC bristles and a few additional SC bristles. This phenotype is similar to Pnr^D constitutive mutants, in which Pnr-Ush interactions are greatly reduced. In contrast, isl mutants show the opposite phenotype, with a large excess of SC bristles and a few additional DC bristles. The Chip^E mutant exhibits antagonistic phenotypes: lack of DC bristles, reflecting Pnr loss-of-function and an excess of SC bristles, reflecting Isl loss-of-function. The differential topography of DC and SC enhancer binding sites presumably underlies differential transcription-complex binding affinities (Biryukova, 2005).

Chip is the ortholog of Ldb factors that are ubiquitous multiadaptor proteins in vertebrates. Each Ldb-dependent developmental event is specified by modification of the transcriptional complex and is dependent on the stoichiometry of the region-specific Ldb partners. During normal development of the thorax, different partners of Chip (i.e., Isl, Ap and Pnr) are expressed in the same region. The Chip^E mutant is highly sensitive to the dosage of these factors. In Chip^E flies, removing one copy of either Pnr or Isl causes pupal lethality associated with extreme morphogenetic phenotypes. Removing one copy of Ap, however, rescues the Pnr-dependent phenotypes of Chip^E flies. Taken together, these results indicate selective competition between the different partners of Chip, suggesting that hierarchical protein interactions depending on differential affinities and the strict stoichiometry of Chip and its partners, are critical to establish proper transcriptional codes within different proneural fields (Biryukova, 2005).

isl mutants were isolated in genetic screens for dominant enhancers of the Chip^E phenotype. This study demonstrates that the LIM-HD transcription factor Isl can bind to the LID of Chip. The binding of the LID domain of Chip with LIM domains has been conserved throughout evolution as has Chip binding with bHLHs proteins. LID contains two subdomains: a small N-terminal hydrophobic β patch (VMVV) followed by a large α helix. Chip^E mutation has a single substitution that changes an Arg to Trp (R504W) in the middle of the α helix. This residue is highly conserved among species and mediates high-affinity contact with the LIM domains. Interestingly, the R504W substitution in Chip abolishes, or strongly reduces, both interactions with the bHLHs and also interactions with Isl. This result implies that bHLHs and Isl recognize the same site within the LID domain of Chip. The data argue that competition between bHLHs and Isl for the LID domain of Chip may be critical for modulating the activity of transcription complexes during development. In vertebrates, the NLI homolog of Chip mediates direct coupling of the proneural bHLH factors Ngn2, NeuroM and the LIM-HD transcription factors (Isl1 and Lhx3). This interaction leads to transcriptional synergism and the synchronization of motor neuron subtype specification with neurogenesis in the embryonic spinal cord of chicken. This work demonstrates that Isl is able to interfere with proneural activity of Chip-Pnr-bHLH transcription complex and therefore, Isl is thought to be able to antagonize proneural specification (Biryukova, 2005).

Interestingly, the Chip^E mutation has little or no effect on interactions with other LIM-containing factors, such as Ap and dLMO, suggesting that different factors have different affinities with the Chip LID domain. Therefore, the Chip^E mutation changes the hierarchy of the affinities among the different partners of Chip in the mesothorax (Biryukova, 2005).

A transcription-complex 'cassette' model is proposed for the specification of region-specific patterns of specialized cell types. In this model, the presence of one of a number of alternative binding factors modifies the specificity of a core transcription complex. This model makes the prediction that, while the core components of the transcription complex will be strongly conserved in evolution, the specificity cassette components will vary significantly between species showing divergent morphogenetic patterns. Comparison of these variable components in related species should provide insights into the fundamental mechanisms of encoding the pattern of differentiated cell types within morphogenetic fields (Biryukova, 2005).

Drosophila LIM-only is a positive regulator of transcription during thoracic bristle development

The Drosophila LIM-only (Lmo) protein DLMO functions as a negative regulator of transcription during development of the fly wing. This study reports a novel role of Dlmo as a positive regulator of transcription during the development of thoracic sensory bristles. New dlmo mutants, which lack some thoracic dorsocentral (DC) bristles, were isolated. This phenotype is typical of malfunction of a thoracic multiprotein transcription complex, composed of Chip, Pannier (Pnr), Achaete (Ac), and Daughterless (Da). Genetic interactions reveal that dlmo synergizes with pnr and ac to promote the development of thoracic DC bristles. Moreover, loss-of-function of dlmo reduces the expression of a reporter target gene of this complex in vivo. Using the GAL4-UAS system it was also shown that dlmo is spatially expressed where this complex is known to be active. Glutathione-S-transferase (GST)-pulldown assays showed that Dlmo can physically bind Chip and Pnr through either of the two LIM domains of Dlmo, suggesting that Dlmo might function as part of this transcription complex in vivo. It is proposed that Dlmo exerts its positive effect on DC bristle development by serving as a bridging molecule between components of the thoracic transcription complex (Zenvirt, 2008).

The results presented in this study uncover a novel role of Dlmo in regulation of the development of the thoracic DC bristles. Homozygous, or hemizygous, loss-of-function (dlmo^hdp) mutants lack the anterior pair of the DC bristles. Moreover, these dlmo mutants displayed genetic interactions with mutants in genes known to regulate DC bristle development, such as pnr and ac, to reduce the number of DC bristles. Consistently, overexpression alleles of dlmo (dlmo^Bx) also exhibited genetic interactions with these pnr and ac mutants, resulting in an increased number of bristles. In addition, the finding that overexpression of pnr under the regulation of dlmo-GAL4 affects DC bristle development suggests that dlmo is expressed in the region of the wing disc that gives rise to these bristles (Zenvirt, 2008).

These results suggest a role of Dlmo in positive regulation of transcription. The negative role of Dlmo in modulation of transcription during Drosophila wing development has been well documented. The findings indicate that in another context, namely in regulation of DC bristle development by the Chip, Pnr, Ac and Da (CPAD) complex, Dlmo has another role, as a positive regulator of transcription. Lowering the level of Dlmo (in dlmo^hdp mutants) results in a reduction in the expression of a reporter driven by regulatory sequences of a bona fide target gene of the CPAD transcription complex, suggesting that Dlmo is a positive regulator of CPAD-dependent transcription. While the mechanism by which Dlmo positively regulates transcription in the context of the CPAD complex remains to be elucidated, a first clue to this mechanism may lie in the finding that Dlmo can bind constituent proteins of this complex, including Pnr and Chip, in vitro. Should these interactions also take place in vivo, Dlmo may exert its positive role in transcriptional regulation as a component of the CPAD complex (Zenvirt, 2008).

Insights into the mechanism of positive transcriptional regulation by Dlmo can be gleaned from LMO2, one of the mammalian homologs of Dlmo. LMO2 was demonstrated to participate in a multiprotein transcription complex that contains Ldb1, a GATA factor (GATA-1 or GATA-2), and the bHLH transcription factors TAL1 and E2A, which are homologous, respectively, to the fly components of the CPAD complex, Chip, Pannier, Achaete, and Daughterless. Various lines of evidence indicate that in mammals LMO2 serves as a bridge between components of the complex, and silencing of LMO2 causes disruption of the complex and decreases in the activation of transcription of its target genes, just as does silencing of Ldb1 or Tal1. Similarly to LMO2, Dlmo might serve as a bridge between components of the CPAD complex. LIM domains are protein-interaction modules and could serve Dlmo to bind components of the CPAD complex. This suggestion is supported by the finding that each single LIM domain of Dlmo is capable of binding components of the CPAD complex in vitro, and it agrees with similar reports on other LIM-containing proteins. Notably, a single LIM domain from LMO2 and LMO4 is sufficient to interact with Ldb1 or the related protein CLIM-1a. However, both LIM domains are required for the highest-affinity interactions (Zenvirt, 2008).

This proposed mode of action of Dlmo, as a bridging molecule, which binds a different protein through each one of its LIM domains, predicts that a Dlmo molecule with one defective LIM domain and one intact LIM domain would bind only one protein at a time and not be able to bridge between molecules. Indeed, in our new dlmo mutants we found that deletions that span the second zinc finger of the second LIM domain of Dlmo, namely dlmo^hdp48-1 and dlmo^hdp185-1, resulted in dlmo loss-of-function mutations. These mutants display partial loss of thoracic DC bristles along with reduced expression of a target gene of the thoracic transcription complex. Interestingly, the wing size of these mutants is normal, unlike the small wings of mutants with lesions in the 5'-UTR of Dlmo, such as dlmo^hdp58-1, dlmo^hdp67-2, and dlmo^hdpR590. This may suggest that the defective Dlmo protein, which has only a single intact LIM domain, is sufficient for its function in the context of the wing, where Dlmo acts as a negative regulator that binds only one protein (CHIP), but is not sufficient when Dlmo acts as a bridging molecule in the thoracic CPAD transcription complex. Finally, our finding that Dlmo can bind other Dlmo molecules to generate homodimers or multimers might provide Dlmo with a greater flexibility of bridging between distant components of the complex. This possibility remains to be examined (Zenvirt, 2008).

In conclusion, Dlmo appears to have a dual role in regulation of transcription, depending on the context. Such a phenomenon has been documented for other transcription cofactors, whose dual function in transcription regulation varies according to their binding partners, the specific tissue, or the developmental stage. Likewise, these results indicate Dlmo has such a dual role, being a negative regulator with respect to the Ap-Chip complex and a positive regulator in the context of the CPAD complex (Zenvirt, 2008).

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

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