Promoter Structure

A candidate teashirt target that is shared with some HOM-C genes has been identified: the modifier of variegation gene modulo (mod). mod is strongly expressed in parasegment 2, where it constitutes an early marker for the salivary placodes, the primordia of the salivary glands. In a Sexcombs reduced mutant, no mod expression is detected in PS2, indicating that the high level of mod expression in the parasegment is under the positive control of the Scr protein. Tsh represses mod in PS3 and this repression is performed independent of Scr. In vitro, Tsh recognizes two specific sites within a 5' control element of the mod gene, which, in vivo, respond to tsh activity. Tsh is therefore a DNA binding protein and might directly control mod expression (Alexandre, 1996).

To determine whether mod expression is transcriptionally controlled by dMyc, mod mRNA level was measured in loss-of function dmyc mutants. No change in mod expression was observed for the viable hypomorphic allele dmycP0. In contrast, both in situ hybridization and quantitative RT-PCR on third instar larval imaginal discs, show that mod transcription is severely impaired in the pupal lethal dmycPL35 mutants. Thus sufficient diminishing of dmyc+ function reveals dMyc requirement for mod expression. The effect of dMyc on mod transcription was also analyzed in gain-of-function experiments, using the UAS/Gal4 system. In third instar wing imaginal discs, dMyc over-expression directed by dpp regulatory sequences leads to a marked increase in mod transcription. Also, in engrailed(en)-Gal4/UASdmyc embryos, mod transcription is strongly induced in the posterior cells of each parasegment where dMyc is over-expressed. Taken together, these results show that dMyc is required for mod expression (Perrin, 2003).

E-boxes constitute functional Myc binding sites that typically reside downstream of the transcriptional start sites of target genes. The action of Myc in regulating transcription has been described as involving binding of Myc homo-or hetero-dimers to an 'E-box' sequence based on a 'CACGTG' motif. A 1 kb DNA fragment located upstream of mod coding sequences has been shown capable of directing reporter gene expression that mimics the mod embryonic expression pattern (Alexandre, 1996). This fragment harbors a canonical CACGTG E-box between the mod initiator ATG and a transcriptional start site assigned by primer extension. A 365 bp fragment (P1) encompassing both the transcription start site and the E-box was fused to a Lac-Z reporter gene. Trangenic flies containing this P1-LacZ chimeric gene express ß-Gal in a pattern similar to endogenous mod. Further, on expressing dMyc in embryos (enGAL4, UASdMyc), mod expression and P1-LacZ expression are augmented in the posterior compartments. To ask whether the responsiveness of P1 to dMyc is E-box dependant, the canonical CACGTG E-box was mutated (CAGGTG) to abolish a potential dMyc-DNA interaction, according to Myc binding specificity. When fused to a Lac-Z reporter gene, this mutated P1 fragment is no longer able to mimic the mod transcription pattern, either in wild type or en-Gal4/UASdmyc embryos. Taken together, these results strongly support the notion that mod transcription is controlled by dMyc, and favor the possibility that dMyc binds directly to the canonical E-box residing in mod regulatory sequences (Perrin, 2003).

It has been shown that diminished mod activity leads to a Minute-like phenotype (Perrin, 1998; Roman 2000), thus suggesting a role for Mod in ribosome biogenesis. A detailed analysis of mod growth-related phenotypes showns that it acts on growth and size of proliferative cells. mod loss-of-function selectively affects imaginal diploid cells but not endoreplicative tissues. In addition, Mod over-expression affects diploid cells but not endoreplicative tissues. For instance, salivary glands cells are bigger in cell and nucleus size upon ectopic expression of dMyc, but look normal following Mod over-expression. Since amino acids directly control growth of endoreplicative tissues, it is unlikely that Mod is related to nutrient availability. Indeed, in agreement with the phenotype specific for proliferative cells, mod transcription is controlled by dMyc. Nevertheless, mod is certainly not involved in all the various cellular processes controlled by dMyc, since in the dmycPL35 mutant, imaginal and endoreplicative tissues are equally affected (Perrin, 2003).

Transcriptional Regulation

Growing evidence involves chromatin structural flexibility in gene regulation during development. modulo is a dominant suppressor of position effect variegation, suggesting the participation of its product in the assembly of higher order chromatin structures. The patterns of modulo expression and regulation during embryogenesis have been examined, analyzed in correlation with phenotypical defects resulting from the amorphic mutation of the gene. Zygotic expression of modulo depends on the activity of genes that pattern the embryo along dorsoventral and anteroposterior axes and specify diversified morphogenesis. Dorsal and the mesoderm-specific genes twist and snail direct modulo expression in the presumptive mesoderm. The homeotic genes Sex combs reduced and Ultrabithorax positively regulate the gene in the ectoderm of parasegment 2 and abdominal mesoderm, respectively, modulo mutants exhibit aberrant morphogenesis of tissues originating from those embryonic primordia which normally express the gene, but do not show defect in cell fate specification. It is proposed that down-stream of pattern-forming genes modulo controls, via chromatin structural changes, genes critical for the process of morphogenesis of several tissue types (Graba, 1994).

Transcriptional Targets

Transcriptional activation in early spermatocytes involves hundreds of genes, many of which are required for meiosis and spermatid differentiation. A number of the meiotic-arrest genes have been identified as general regulators of transcription; however, the gene-specific transcription factors have remained elusive. To identify such factors, the protein that specifically binds to the promoter of spermatid-differentiation gene Sperm-specific dynein intermediate chain (Sdic) was purified and identified as Modulo, the Drosophila homologue of nucleolin. Analysis of gene-expression patterns in the male sterile modulo mutant indicates that Modulo supports high expression of the meiotic-arrest genes and is essential for transcription of spermatid-differentiation genes. Expression of Modulo itself is under the control of meiotic-arrest genes and requires the DAZ/DAZL homologue Boule that is involved in the control of G2/M transition. Thus, regulatory interactions among Modulo, Boule, and the meiotic-arrest genes integrate meiosis and spermatid differentiation in the male germ line (Mikhaylova, 2006).

Although the general regulators of transcription in testes have been extensively characterized, the gene-specific transcription factors have long been elusive. To characterize such factors, attempts were made to identify the protein that binds to the conserved positive regulatory element b2UE1/b2UE2/TSE (see Nurminsky, 1998, for TSE consensus) that is necessary for activity of the β(2)tubulin promoter in Drosophila testes and is present in the promoter of the testes-specific gene Sdic. Modulo is required for transcription of a number of spermatid-differentiation genes, including β(2)tubulin and Sdic. Expression of Modulo itself in testes is positively regulated by the meiotic-arrest genes at the posttranscriptional level and requires the DAZ/DAZLA homologue Boule, the protein that also controls the G2/M meiotic transition through posttranscriptional regulation of Cdc25/Twine (Mikhaylova, 2006).

The promoter of the testes-specific gene Sdic contains the TSE motif that shows similarities to the conserved elements b2UE1 and b2UE2 found in other testes-specific promoters. An abundant TSE-binding protein was detected in protein extracts from Drosophila testes but not from gonadectomized males by using EMSA. Formation of the DNA–protein complex was completely inhibited by a 100-fold molar excess of the unlabeled TSE probe. At the same time, the presence of a 104-fold molar excess of the heterologous double-stranded oligonucleotide competitor 1 in all EMSA reactions did not inhibit formation of the DNA–protein complex, indicating that binding of the protein is sequence-specific. Further addition of the different oligonucleotide competitor 2 in 100-fold excess to the probe did not interfere with the complex formation. To corroborate this finding, five more different heterologous oligonucleotides were tested using the same conditions, and none of them impeded complex formation. Thus, a protein that specifically binds to the conserved TSE promoter motif is up-regulated in testes and may be involved in transcriptional regulation of Sdic (Mikhaylova, 2006).

To identify the TSE-binding factor, a multistep procedure was developed for its biochemical purification from whole adult flies Modulo is a broadly expressed protein that has been detected in ovaries, embryonic epidermis and mesoderm, larval imaginal discs, salivary glands, and brain and in cultured cells. Western blot analysis showed that the size of Modulo differs between testes and somatic tissues represented by the gonadectomized males. In testes, mobility of the protein is more consistent with the predicted 60.3-kDa size of the Modulo polypeptide; no signal was detected in testes of the mod07570 male sterile mutant, thus confirming the specificity of the assay. However, the apparent molecular mass of the Modulo variant expressed in somatic tissues is ~50 kDa, which corresponds with the size of the TSE-binding protein that was purified from the whole-fly extracts and identified as Modulo (Mikhaylova, 2006).

A analysis indicated that the 50-kDa protein is a truncated variant missing the N terminus of the full-size Modulo. All of the identified trypsin-generated peptides were located in the C-terminal portion of the molecule, whereas not a single peptide was identified near the N terminus. In addition, an unusual peptide flanked with the trypsin cleavage site at only one (the C-terminal) end was repeatedly identified with high confidence during the analysis. The N-terminal end of the peptide thus possibly represents the N terminus of the truncated 50-kDa Modulo variant, and its position is consistent with the size of the protein (predicted molecular mass, 46.2 kDa). In this case, the 50-kDa variant is missing the highly acidic N-terminal domain that is present in the full-size protein (Mikhaylova, 2006).

The 50-kDa somatic Modulo variant is capable of specific binding to the TSE motif. Thus, a specific variant of Modulo is expressed in testes, where it is capable of specific binding to the TSE-containing promoters. This full-size Modulo variant contains the N-terminal acidic domain, the structure of which is characteristic for the acidic activators that facilitate assembly of the core transcription machinery on the promoter and recruitment of chromatin-remodeling factors. Known acidic activators interact with the TFIID complex and facilitate interaction of TFIID with TFIIA and TFIIB. In testes, TFIID is represented by a specific variant encoded by the meiotic-arrest genes of the sa group. Modulo coimmunoprecipitates with the testes-specific TFIID subunit Sa and, thus, probably interacts with the testes-specific TFIID during transcriptional activation of testes-specific genes (Mikhaylova, 2006).

The suggested activity of Modulo as transcriptional activator in testes is not consistent with its role in somatic tissues, where it is involved in multiple vital activities that probably include chromatin-mediated transcriptional repression [based on the demonstrated Su(var) phenotype of the modulo mutants]. Structural differences between the Modulo variants may underlie this apparent discrepancy. The 50-kDa somatic Modulo variant is able to bind to DNA but is missing the N-terminal acidic domain and, thus, cannot establish the activating interactions observed in testes. Instead, the somatic Modulo variant may contribute to repression of testes-specific genes in somatic tissues (Mikhaylova, 2006).

The Modulo-binding element TSE is present in the promoter of the testes-specific gene Sdic that is up-regulated in primary spermatocytes. To regulate Sdic expression, Modulo has to be present at the same or earlier stage of spermatogenesis. Localization of the zone of up-regulation of Modulo in adult whole-mount testes using immunofluorescence showed that this zone, indeed, precedes and overlaps the zone of Sdic expression. To visualize Sdic expression, advantage was taken of the Sdic::GFP fusion transgene. Modulo is weakly expressed in early spermatogonia and stem cells located at the tip of the testis, but is up-regulated in late spermatogonia/early spermatocytes. Within the cells, Modulo is localized in both the nucleus and the cytoplasm. Specificity of the immunofluorescence staining was confirmed by the absence of appreciable signal in testes of the mod07570 and achi1 mutants that show severe down-regulation of Modulo; also, no staining was observed in wild-type testes when the primary antibody was omitted (Mikhaylova, 2006).

Thus, in spermatogenesis, up-regulation of Modulo precedes activation of its putative regulation target, Sdic. Attempts were made to see whether this pattern is also present for the general transcriptional regulators (the meiotic-arrest genes) and their regulation targets (the spermatid-differentiation genes). To dissect the temporal order of expression of these genes, transcript levels were quantitated in testes dissected from developing larvae using real-time RT-PCR. In Drosophila, spermatogenesis begins early in larval development, and the first wave of meiosis commences at the time of pupation. All of the five tested spermatid-differentiation genes, including Sdic, show drastic up-regulation in late third-instar larvae, i.e., during spermatocyte maturation before meiotic divisions. This finding is consistent with the observed pattern of up-regulation of the Sdic::GFP transgene in adult testes. However, up-regulation of the three studied meiotic-arrest genes precedes up-regulation of the spermatid-differentiation genes. Hence, Modulo is up-regulated in the male germ line in a manner similar to known general transcriptional regulators, before the major wave of transcriptional activation that includes spermatid-differentiation genes (Mikhaylova, 2006).

Although knockout modulo mutations result in lethality, a male sterile hypomorphic mutation mod07570 has been described; thus, mutation, caused by a transposon insertion, results in testes-specific modulo knockdown. To analyze the role of Modulo in transcriptional regulation, transcripts of a number of spermatogenesis-related genes were quantitated in testes of the mod07570 mutant and of the wild type, using real-time RT-PCR. The constitutive transcript of the ribosomal protein gene Rp49 was used as the cDNA template-loading reference. The ubiquitous transcripts RpL9 and Act5C and the broadly expressed spermatogenesis-related genes des and twe were not significantly affected by the mod07570 mutation (Mikhaylova, 2006).

Conversely, a number of genes with testes-biased expression were down-regulated to various extents in the mod07570 mutant testes. In particular, several meiotic-arrest genes (including aly, can, nht, and rye) were down-regulated 5- to 7-fold. Among the 13 other testes-biased genes examined, 7 showed moderate 2- to 5-fold down-regulation. However, four testes-biased genes showed >10-fold down-regulation, and two more genes were down-regulated 7- to 9-fold. Thus, there is a subset of testes-biased genes [including Sdic, Ssl, fzo, dhod, β, and dj] that are specifically affected by the Modulo deficiency (Mikhaylova, 2006).

Because the meiotic-arrest genes themselves are involved in transcriptional regulation in testes, some effects of the Modulo deficiency may be mediated by their down-regulation. Such effects should be similar to the effects caused by mutations in the meiotic-arrest genes themselves. This possibility was addressed by analyzing gene expression patterns in testes of the meiotic-arrest mutants achi1, sa1, and Taf12LKG00946 (rye), and by comparing them to the pattern of gene expression in the mod07570 mutant testes. Among the 13 testes-biased genes analyzed, four genes were affected differently by the meiotic arrest and the modulo mutations. The genes Mst98Ca, Pros28.1B, and CG10934 were very sensitive to mutations in meiotic-arrest genes but not in modulo and, conversely, the gene Ssl did not show striking sensitivity to the mutations in the meiotic-arrest genes sa and rye but was severely affected in the modulo mutant. Therefore, the effect of Modulo deficiency on transcription in testes cannot be reduced to down-regulation of the meiotic-arrest genes. It is possible that such down-regulation leads to the subpar performance of the meiotic-arrest genes that is still sufficient to carry the germ line of the mod07570 mutant through the meiotic divisions but results in moderate (e.g., 3- to 4-fold) down-regulation of testes-biased genes, such as Mst98Ca, Pros28.1B, and CG10934. However, a more severe effect of the Modulo deficiency on a subset of spermatid-differentiation genes probably reflects disruption of gene-specific transcriptional regulation and provides a molecular basis for the spermatid-differentiation failure observed in the mod07570 mutant (Mikhaylova, 2006).

The subset of genes strongly affected in the mod07570 mutant includes Sdic and β(2)Tubulin. These genes possess the TSE-like Modulo-binding motifs and, thus, probably represent the direct regulatory targets of Modulo. Interestingly, it was not possible to detect specific binding of Modulo to the promoters of fzo and dj that are also strongly affected by the modulo mutation. At the same time, the studied dj promoter fragment contained all sequences necessary for efficient testes-specific transcription. This finding implies that Modulo has indirect target genes such as dj and, probably, fzo that may be regulated by transcription factors that are, in turn, under the control of Modulo. A broad survey of 96 transcriptional regulators expressed in testes identified nine putative transcription factors that are down-regulated >10-fold in the mod07570 mutant testes. Thus, mutation in modulo can lead to disruption of the downstream cascade of transcriptional regulation that includes Modulo-dependent transcription factors and their regulation targets (Mikhaylova, 2006).

To determine whether Modulo is sufficient to induce ectopic transcription of spermatid-differentiation genes, recombinant full-size Modulo was expressed in the Schneider-2 cultured cells under the control of metallothionein promoter. Stable transfected clones were selected, and expression of the transgene was induced by various concentrations of Cu2+ in the culture media. Unexpectedly, it was observed that the Schneider-2 cells naturally express the full-size Modulo variant. Nevertheless, these cells do not show significant expression of the Modulo-dependent testes-specific genes Sdic, Ssl, and dhod, and increase of the Modulo dose by induced expression of the transgene did not affect the levels of these transcripts. Therefore, other, presumably testes-specific factors, (such as the testes-specific TFIID) have to cooperate with Modulo to induce expression of spermatid-differentiation genes, thus defining tissue specificity of the Modulo-mediated transcriptional regulation (Mikhaylova, 2006).

To analyze the regulation of Modulo expression in testes, a number of mutants that control different stages of spermatogenesis were analyzed. In testes of the bam mutant, both the Modulo protein and modulo transcript are severely down-regulated. Thus, high levels of Modulo expression in the testes require the onset of the meiosis/differentiation program. Furthermore, mutations in the meiotic-arrest genes achi/vis, sa, and rye result in severe down-regulation of Modulo protein in testes; however, modulo transcription is not affected. Therefore, Modulo expression in the testes is regulated by the meiotic-arrest genes at posttranscriptonal levels, similar to the regulation of the meiotic entry control protein Cdc25/Twine. Translation of Twine in the testes requires the RNA-binding protein Boule. To investigate whether a similar mechanism is involved in the regulation of Modulo, testes of the boule mutants were examined, and it was found that Modulo expression in testes is severely affected by the Boule deficiency (Mikhaylova, 2006).

Regulation of Modulo expression in testes by Boule provides a mechanistic link between meiosis and spermatid differentiation in the male germ line. The meiotic-arrest genes are required for expression of a number of spermatogenesis-related genes, including boule. Boule is required for expression of Modulo, which, in turn, is necessary to maintain expression of several meiotic-arrest genes. These events establish a positive regulatory loop that sustains high levels of expression of Boule, Modulo, and the meiotic-arrest genes after the onset of the meiosis/differentiation program in spermatocytes. Boule further regulates the G2/M transition in meiosis by positive translational regulation of Cdc25/Twine, and Modulo and the products of the meiotic-arrest genes are required for expression of a number of spermatid-differentiation genes. Thus, the pathways that lead to meiosis and to expression of the spermatid-differentiation genes in the male germ line are integrated in a single mechanism to ensure coordinated execution of meiotic divisions and spermatid differentiation (Mikhaylova, 2006).

Dynamics of the sub-nuclear distribution of Modulo and the regulation of position-effect variegation by nucleolus in Drosophila

modulo belongs to the class of Drosophila genes named 'suppressor of position-effect variegation', suggesting the involvement of the encoded protein in chromatin compaction/relaxation processes. The sub-nuclear distribution of Modulo was examined using complementary procedures of cell fractionation, immunolocalization on mitotic and polytene chromosomes and crosslinking immunoprecipitation of genomic DNA targets. While actually associated to condensed chromatin and heterochromatin sites, the protein is also abundantly found at nucleolus. From a comparison of Modulo pattern on chromosomes of different cell types and mutant lines, a model is proposed in which the nucleolus balances the Modulo protein available for chromatin compaction and PEV modification. At a molecular level, repetitive elements instead of rDNA constitute Modulo DNA targets, indicating that the protein directly contacts DNA in heterochromatin but not at the nucleolus. Consistent with a role for Modulo in nucleolus activity and protein synthesis capacity, somatic clones homozygous for a null mutation express a cell-autonomous phenotype consisting of growth alteration and short slender bristles, traits characteristic of Minute mutations, which are known to affect ribosome biogenesis. The results provide evidence suggesting that Modulo participates in distinct molecular networks in the nucleolus and heterochromatin and has distinct functions in the two compartments (Perrin, 1998).

Based on the suppression of PEV caused by mod and on the DNA-binding activity of its product, it has been proposed that Mod serves to anchor multimeric complexes promoting chromatin compaction and silencing (Garzino, 1992). The combined use of polytene and mitotic chromosomes has revealed distinct but consistent features, and allows a clear picture of Mod at chromatin. The prominent conclusion from immunolocalization on polytene chromosomes is the association to the chromocenter and almost all dense bands on chromosome arms. Binding to heterochromatin is best seen in mitotic cells, at the pericentric heterochromatin of all chromosomes (except the fourth) and specific sites on chromosome Y. The data thus provide direct evidence of Mod binding to condensed chromosomal regions. Consistent with a role in chromatin structure, mutation of one gene copy results in PEV suppression and a strong decrease of Mod at pericentric heterochromatin and chromosome bands (Perrin, 1998).

Cross-linking experiments demonstrate that Mod interacts with several repetitive elements, which is in general agreement with its association to heterochromatin and the PEV suppression phenotype of mod mutant animals. Several points, however, indicate that these interactions do not direct the protein to specific heterochromatic sites. (1) It was never found associated to chromosome four, which is essentially heterochromatic, on mitotic cell preparations. Thus, Mod is not a general component of heterochromatin. (2) The Mod pattern does not strictly overlap the distribution of the cognate repetitive elements that show a wider distribution spectrum on mitotic chromosomes. (3) While directly contacting DNA (Garzino, 1992), Mod has no canonical DNA binding domain and does not recognise a specific DNA motif in vitro. This suggests that Mod might associate with unknown factors that provide DNA binding specificity and direct the complex towards particular sites, at which interaction of Mod with repetitive elements could favor and stabilize the formation of condensed heterochromatin (Perrin, 1998).

Only a small fraction of Mod (about 10%) remains bound to chromatin during step-wise salt extraction of embryonic nuclei, while 90% is recovered with the nucleoplasmic fraction. Mod thus appears either to be essentially free in the nucleoplasm or loosely bound to a nuclear organelle and released during the experiment. Double staining experiments for Mod and the nucleolar antigen AJ1 on a variety of embryonic or larval tissues unambiguously demonstrate that the most intensive staining is always associated with nucleolus. It is therefore concluded that in vivo the major part of Mod is nucleolar, and it becomes released in the nucleoplasm during cell fractionation (Perrin, 1998).

It is proposed that the nucleolus titrates the bulk of Mod available for chromatin compaction and PEV modification. Competition for Mod between chromatin and nucleolus is supported by three lines of evidence. (1) In salivary glands cells, the nucleolar staining is far less sensitive to a reduction of mod gene dosage than is chromatin staining. (2) Mod does associate to the ectopic mininucleolus that results from rDNA insertion in euchromatin. No significant reduction of chromosomal association of Mod has been observed in this experiment, which is actually not surprising since the rib7(1A) line contains a single extra rDNA unit in addition to about 200 units normally found in the Drosophila genome. (3) In PNCs Mod is mainly found at the nucleolus and barely detectable on chromosomes. The difference in pattern between PNCs and salivary glands is best understood by considering the relative amount of rDNA in the two cell types. While rDNA is heavily under-replicated during salivary gland development, it endoreplicates at almost the same rate as the rest of the genome in nurse cells. The nucleolus is therefore large, a process related to the production of protein and rRNA for the oocyte. Competition for Mod between this large and active nucleolus and chromatin sites explains the poor chromosomal staining of PNCs. In comparison the reduced size and activity of salivary gland nucleoli improves the amount of protein available for chromatin and accordingly allows immunodetection at pericentric heterochromatin and bands (Perrin, 1998).

The nucleolus organizer (NO), which is composed of tandem repeats of rDNA cistrons, has long been implicated in modification of variegation, since a decrease in its dosage leads to PEV enhancement and an increase in PEV suppression. In a similar way, the reduced potential of polytene chromosome-containing pseudonurse cells to modulate PEV is presumably due to the unusually large nucleolus and rDNA over-representation. The simplest explanation for this rDNA effect is that the nucleolus and pericentric heterochromatin compete for common protein factors. Mod is the first candidate playing a role in such a mechanism to be identified. In animals heterozygous for mod, displacement of the protein from dense DNA regions is amplified by the competitive effect of the nucleolus, thus leading to the expression of the PEV suppression phenotype (Perrin, 1998).

Recent developments in silencing effects in yeast established that the nucleolus is a compartment for Silent Information Regulator proteins. These proteins are involved in silencing at telomeres and mating-type loci. At the nucleolus they are associated with rDNA and are thought to function in the extension of life span. Since PEV in Drosophila and silencing in yeast are likely to be related phenomena, one can hypothesize that Mod also directly binds rDNA. Several lines of evidence actually argue against this hypothesis. The low abundance of rDNA sequences in the library of Mod DNA targets makes it unlikely that the preferential accumulation of the protein in the nucleolus results from an interaction with rDNA. However, nucleolar Mod is loosely bound to the organelle and is released into nucleoplasm at low ionic strength. Finally, unpublished work has shown (1) that the protein purified from nucleoplasm is highly phosphorylated and unable to bind DNA in vitro, and (2) that it is associated with an RNA molecule and migrates as a riboprotein complex on native gel electrophoresis. It is therefore concluded that Mod does not contact rDNA at the nucleolus. The strong ectopic signal of Mod labelling on polytene chromosomes of the rib7(1A) transgenic line does not imply that Mod binds the extra rDNA unit. Since the transgene is active in transcription and nucleolus formation, this result simply indicates that Mod is associated with the mini-nucleolus that develops from the rDNA insertion site. It is therefore tempting to assume that nucleolar localization of Mod depends on interaction with a RNA molecule involved in nucleolus biology, either rRNA or small nucleolar RNA (Perrin, 1998 and references therein).

The involvement of Mod in different molecular networks in heterochromatin and the nucleolus suggests that it has distinct functions in the two compartments. It is noteworthy that the intranucleolar distribution of the protein changes during development, from a homogeneous distribution in the entire nucleolus of actively dividing cells (young embryos, imaginal discs) to a restricted accumulation at the organelle periphery in post-mitotic cells (late embryos, salivary glands). This change in pattern in mitotic versus quiescent cells could tentatively be related to a function for Mod at the nucleolus. In support of this, somatic clonal analysis has revealed that mod loss of function is cell-autonomous and results in growth alteration and in short slender bristle morphology. These phenotypes are characteristic traits of a large class of dominant mutations, the Minute class, widely believed to affect ribosomal protein genes. Two additional classes of mutations, mini and bobbed, which affect rRNA genes, also alter protein synthesis and actually present similar phenotypes. The Minute-like phenotypes of mod somatic clones point to a lesion in the protein synthesis capacity of progenitor cells from which the clones derive. When considered together, these data strongly suggest that Mod, in addition to its function in chromatin compaction processes, has a role in the regulation of nucleolus activity (Perrin, 1998).

Protein Interactions

A yeast two-hybrid screen of a Drosophila cDNA library was performed to assess binding of the well-conserved dCBP C-terminal C/H3 domain to nuclear factors. Three positive clones encoded Modulo, a DNA-binding protein that appears to be involved in the compaction of chromatin in Drosophila. Both the C/H3 domain and the N-terminal region of the dCBP activation domain are necessary for Modulo binding. Amino acids 219-493 of Modulo are sufficient to bind dCBP. Binding between dCBP and Modulo was confirmed in vitro by using a bacterially expressed dCBP fusion protein (GST-dCBP-2278-2678) that contains the C/H3 domain and the first 200 aa of the glutamine-rich activation domain. An in vitro translated Modulo polypeptide (amino acids 219-544) corresponding to the fragment isolated with the two-hybrid screen, as well as the full-length Modulo, binds strongly to GST-dCBP-2278-2678. However, no binding of Modulo is observed to GST fusion proteins containing the dCBP CREB-binding domain or the bromo-zinc finger domain. Whole-cell extracts from Drosophila Kc cells were used to examine Modulo;dMyc co-immunoprecipitation -- an antibody directed against dCBP specifically immunoprecipitates Modulo, indicating that both proteins are part of the same complex in vivo (Bantignies, 2002).

The in vivo functional significance of this interaction was examined. The phenotype was determined of embryos hemizygous for the dCBP allele nej1 and either homozygous or heterozygous for the mod allele L8. These phenotypes were then compared with those of nej1/+; L8 and nej1/+; L8/+ control embryos. The cuticles were examined of 300-400 unhatched embryos per genotype. Most of the nej1 hemizygotes develop to the end of embryogenesis and are phenotypically wild type. Approximately 13% of the embryos had head defects and 4.5% had fusions of the segmental denticle belts, primarily between the denticle belts of the fourth and the fifth abdominal segments. Most, over 90%, of the L8 homozygous embryos survive to hatching. Approximately 27% of the nonhatching embryos showed head defects, tail defects, or both. The denticle belts of L8 homozygous embryos were virtually wild type, and only 3% had segmental fusions. To analyze nej1 hemizygous embryos in an L8 background, nej1/FM7-GFP; L8/TM3-GFP female flies were crossed to FM7-GFP; L8/TM3-GFP males and cuticles were prepared. Fifteen percent of the nej1; L8 embryos had fusions of the segmental denticle belts. These fusions occurred between the second and the third, the third and the fourth, or the fourth and the fifth abdominal segments. In contrast, only 3% of the nej1;L8/TM3-GFP embryos had denticle belt fusions. Thus, nej and mod mutations appear to have a synergistic effect on denticle belt fusions. 24.2% of the nej1; L8 embryos had head defects, whereas 12% of the nej1; L8/TM3-GFP embryos had head defects. The percentage of head defects observed in the nej1; L8 double mutants might be additive if the two genes affect head development through two independent pathways. The percentage of head defects observed in nej1; L8 double mutants is neither enhanced nor suppressed, and is the same as that seen in the L8 homozygotes, suggesting that mod might be epistatic to dCBP in head development. Similar results were obtained with the nej3 allele. Taken together, these results indicate that a synergistic interaction between dCBP and Mod is required during embryogenesis for proper segmentation. Furthermore, because the nej1; L8 double mutants have an enhanced phenotype, rather than a nej1 or L8 phenotype, it has been suggested that Modulo and dCBP may be members of a larger regulatory complex that is sensitive to changes in their relative dosages (Bantignies, 2002).

Because segmental fusion phenotypes were obtained in the nej1; L8 double mutants, whether dCBP and Modulo are co-expressed during embryonic development was determined. Confocal microscopy was used to show that both proteins are ubiquitously expressed during embryogenesis. Furthermore, Modulo is not detected in the epidermis of the L8 homozygous mutants. Modulo and dCBP are expressed in the nuclei of all segments; however, the subnuclear colocalization is difficult to determine because dCBP is distributed ubiquitously within the nucleus. dCBP is not localized to the nucleolus, where Mod is predominantly located. Although a band of colocalization can sometimes be seen at the periphery of the nucleolus, it is not possible to assess whether this is significant. To determine whether dCBP and Mod colocalize to specific loci in chromatin, an immunohistochemical analysis of polytene chromosomes was performed with antibodies directed against dCBP and Mod. There are some loci that bind only Mod, others that bind only dCBP and a large subset of bands that bind both dCBP and Mod. Both dCBP and Mod bind to heterochromatin however the Mod signal is considerably more intense. The fact that dCBP and Mod coimmunoprecipitate from cell extracts and colocalize on polytene chromosomes reinforces the idea of a functional interaction between dCBP and Modulo during embryonic development (Bantignies, 2002).

As third-instar larvae, L8 homozygous mutants develop melanotic tumors, which form as a result of an immune response to the presence of abnormal target tissues. The melanotic tumor phenotype of L8 is characterized by incomplete penetrance and variable expressivity of the tumor phenotype. The nej1 and nej3 heterozygous larvae never develop melanotic tumors. Whether the dosage of dCBP affects the melanotic tumor phenotype of mod-deficient animals was determined. In this experiment, y1w1; L8/TM6B,Tb males were crossed to females from a nej1/FM7c; L8/TM6B,Tb stock. This cross allowed a comparison of nej1/y1w1; L8/L8 (black mouth hooks) to FM7c/y1w1; L8/L8 (yellow mouth hooks) third-instar females. For each population, a large number of third-instar larvae were analyzed and divided, by tumor number, into four different groups. The nej1/y1w1; L8/L8 population has a greater number of tumors than the FM7c/y1w1; L8/L8 population. The L8/L8 population containing the nej allele are significantly different form the L8/L8 population. The same results were observed with the nej3 allele. This result suggests that when associated with the chromatin-binding factor Modulo, dCBP can affect the immune response needed to suppress the formation of melanotic tumors in fly larvae (Bantignies, 2002).

While the genetic interactions suggest that Modulo and dCBP act in the same or parallel developmental pathways, the biochemical and colocalization data provide evidence supporting the idea that the two proteins interact directly. Interestingly, like mutations in mod, mutations in dCBP act as dominant suppressors of PEV. Thus, although the molecular mechanism of the interaction between dCBP and Modulo is not known, it probably involves the direct or indirect modification of chromatin structure (Bantignies, 2002).

Previous studies have shown that CBP interacts with components of the preinitiation complex and histone acetyltransferases. The results of this study indicate that CBP interacts with an additional class of chromatin-associated proteins as well and suggests that CBP effects on chromatin may not be limited to the level of the nucleosome and may also involve the modulation of higher-order chromatin structure (Bantignies, 2002).

Localization of mRNAs, a process essential for embryonic body patterning in Drosophila, requires recognition of cis-acting signals by cellular components responsible for movement and anchoring. A large multiprotein complex has been isolated that binds a minimal form of the bicoid mRNA localization signal in a manner both specific and sensitive to inactivating mutations. Identified complex components include the RNA binding proteins Modulo, PABP, and Smooth (Sm), the known localization factor Swallow, and the kinesin family member Nod. Localization of bcd mRNA is defective in modulo mutants. The presence of three required localization components (Swallow, Modulo, and specific RNA binding activity) within the recognition complex strongly implicates the complex in mRNA localization (Arn, 2003).

modulo: Biological Overview | Evolutionary Homologs | Developmental Biology | Effects of Mutation | References

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

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