Protein Interactions

Hairy, Deadpan and E(SPL) proteins have three evolutionarily conserved domains required for their function: the bHLH, Orange, and WRPW domains. However, the suppression of Scute activity by Hairy does not require the WRPW domain. The Orange domain is an important functional domain that confers specificity among members of the Hairy/E(SPL) family. A Xenopus Hairy homology conserves not only Hairy's structure but also its biological activity. Transcriptional repression by the Hairy/E(SPL) family of bHLH proteins involves two separable mechanisms: repression of specific transcriptional activators, such as Scute, through the bHLH and Orange domains and repression of other activators via interaction of the C-terminal WRPW motif with corepressors, such as the Groucho protein (Dawson, 1995).

Hairy-related proteins are site-specific DNA-binding proteins defined by the presence of both a repressor-specific bHLH DNA binding domain and a carboxyl-terminal WRPW (Trp-Arg-Pro-Trp) motif. These proteins act as repressors by binding to DNA sites in target gene promoters and not by interfering with activator proteins, indicating that these proteins are active repressors that should therefore have specific repression domains. The WRPW motif is a functional transcriptional repression domain sufficient to confer active repression to Hairy-related proteins or a heterologous DNA-binding protein, Ga14. The WRPW motif is sufficient to recruit Groucho or the TLE mammalian homologs to target gene promoters. Groucho and TLE proteins actively repress transcription when directly bound to a target gene promoter. Thus Groucho family proteins are active transcriptional corepressors for Hairy-related proteins and are recruited by the 4-amino acid protein-protein interaction domain, WRPW (Fisher, 1996).

The repression by Groucho and Hairy-like proteins requires the S/P (Ser-Pro-rich) domain of Groucho and the conserved carboxl-terminal Trp-Arg-Pro-Trp (WRPW) sequence of the Hairy-like transcription factor for direct interactions between Groucho and Hairy-like proteins. In addition to the S/P domain, a WD40 motif is found in the C-terminal half of Groucho. The WD40 motif is a loosely conserved repeat of 40 amino acids, separated by a Trp-Asp dipeptide sequence. The other sequence recognized in the Drosophila protein is a CcN motif, containing Cdc3 and casein kinase sites and a possible nuclear localization signal. A glutamine-rich region, the Q domain, is conserved as well in mammalian Groucho homologs. Both the Q domain and the WD40 domain are highly conserved, suggesting that they perform a highly conserved function. The S/P domain is less well conserved, implying that each Groucho family member binds preferentially to a particular Hairy-like transcription factor. Groucho-related genes in mammals also include one gene that encodes a truncated homolog of Drosophila, containing only the Q domain and part of the S/P sequence. The short protein may act to regulate the activity of the longer protein. The Q domain is used for dimerization between Groucho family members. Groucho proteins are able to dimerize through the Q domain; that dimerization requires a core of 50 amino acids. Surprising, the dimerization does not require the leucine zipper located within the Q domain. It is suggested that the ability of Groucho proteins to dimerize is conserved, and that the presence of truncated Groucho proteins in mammals provides yet an additional level of transcriptional regulation in mammalian systems (Pinto, 1996).

Runt domain family members are defined based on the presence of the 128-amino-acid Runt domain, which is necessary and sufficient for sequence-specific DNA binding. There exists an evolutionarily conserved protein-protein interaction between Runt domain proteins and the corepressor Groucho. However, the interaction is independent of the Runt domain and can be mapped to a 5-amino-acid sequence, VWRPY, present at the C terminus of all Runt domain proteins. Drosophila melanogaster Runt and Groucho interact genetically; the in vivo repression of a subset of Runt-regulated genes is dependent on the interaction with Groucho and is sensitive to Groucho dosage. Runt's repression of one gene, engrailed, is independent of VWRPY and Groucho, thus demonstrating alternative mechanisms for repression by Runt domain proteins (Aronson, 1997).

Unlike other transcriptional regulatory proteins that interact with Groucho, Runt domain proteins are known to activate transcription. The distinction between the Runt domain consensus (VWRPY) and the Hairy-related/HES consensus (WRPW) raises a question: are the C-termini of these families interchangeable? The ability of Runt domain proteins to activate transcription suggests that the interaction with Groucho is regulated: when Runt domain proteins assemble on a promoter that is to be activated, Groucho must either be absent or in a context where it cannot exert its repressive effects. The difference between the Groucho-recruiting C-termini of the Hairy-related/HES family and the Runt domain family may be the difference between a constitutive Groucho interaction and one that is regulated (Aronson, 1997).

A transgenic embryo assay was employed to discover the mode of repression mediated by Hairy. Hairy can act as a dominant repressor capable of functioning over long distances to block multiple enhancers. Hairy is shown to repress a heterologous enhancer, the Rhomboid NEE, when bound 1 kb from the nearest upstream activator. The binding of Hairy to a modified NEE leads to the repression of both the NEE and a distantly linked mesoderm-specific enhancer with a synthetic modular promoter. Two models are proposed for Hairy's long distance repressive function. (1) Hairy could recruit a cofactor that mediates repression at a distance. This factor would inhibit specific upstream activators bound within the proximal promoter. (2) Hairy could interact directly with one or more components of the basal transcriptional complex. Perhaps Hairy binds TFIIE with a high affinity, thereby resulting in a general silencing of the promoter. Additional evidence that Hairy is distinct from previously characterized embryonic repressors (Krüppel and Knirps function over short distances to quench closely linked upstream activators) stems from the analysis of the gypsy insulator DNA. This insulator selectively blocks the Hairy repressor, but not the linked activators Dorsal and Achaete-Scute, within a modified NEE. The linked activators are unaffected by Hairy and continue to direct expression across the gypsy insulator (Barolo, 1997).

How then does Hairy function? Hairy has been shown to interact with the co-repressor protein Groucho through the C-terminal WRPW motif. Gro is not known to bind DNA, but fusions of GRO with heterologous DNA binding domains have revealed that GRO can act as a transcriptional repressor. The Gro protein contains several repeats of a 40-residue motif, termed the WD40 repeat, that is thought to mediate protein-protein interactions. Tup1, a yeast corepressor protein that also contains WD40 repeats, is recruited to DNA by the alpha2 repressor in alpha-type cells for the silencing of alpha-specific genes. Similarly, Hairy may recruit Gro for silencing specific genes in the Drosophila embryo. The yeast mating-type repressors alpha2 and Tup1 have been reported to interact with histones. This observation raises the possibility that Gro mediates transcriptional silencing by influencing chromatin structure (Barolo, 1997)

hairy is a Drosophila pair-rule segmentation gene that functions genetically as a repressor. To isolate protein components of Hairy-mediated repression, a yeast interaction screen was carried out and a Hairy-interacting protein was identified, the Drosophila homolog of the human C-terminal-binding protein (CtBP). Human CtBP is a cellular phosphoprotein that interacts with the C-terminus of the adenovirus E1a oncoprotein and functions as a tumor suppressor. Drosophila

  • CtBP also interacts with E1a in a directed yeast two-hybrid assay. dCtBP interacts specifically and directly with a small, previously uncharacterized C-terminal region of Hairy. dCtBP activity appears to be specific to Hairy in the Hairy/Enhancer of split [E(spl)]/Dpn basic helix-loop-helix protein class. A P-element insertion was identified within the dCtBP transcription unit that fails to complement alleles of a known locus, l(3)87De (Poortinga, 1998).

    To target protein interactions with specific conserved regions of the Hairy protein, a two-hybrid screen was carried out using a LexA-tagged Hairy partial protein bait. This strategy also allowed the circumvention of the reporter system repression that was encountered when using full-length Hairy protein as a bait. A VP16-tagged Drosophila library constructed from 0-4 h embryonic mRNAs was screened with a bait that encodes the Hairy Orange domain through to the C-terminus (h-C, amino acids 93-343. In directed yeast two-hybrid assays, this fragment, h-C28, interacts weakly with full-length Hairy, but more strongly with Hairy partial proteins and with E(spl)m, another member of Hairy-class bHLH proteins. It does not interact with Dpn and interacts poorly, if at all, with E(spl)m3, -m5, -m8, -mbeta -mgamma. h-C28 does not show interaction with proteins from other HLH classes (i.e. Scute, Emc). The region of Hairy required for interaction with h-C28 was mapped using a series of Hairy deletions and partial proteins fused to LexA. h-C28 interacts strongly with a 25 amino sequence immediately upstream of, but not including, the C-terminal WRPW motif. This identifies a previously undefined protein interaction domain within Hairy. dCtBP also interacts with itself (Poortinga, 1998).

    Yeast SIR2 (Silent Information Regulator 2) is a nicotinamide adenine dinucleotide (NAD)+-dependent histone deacetylase required for heterochromatic silencing at telomeres, rDNA, and mating-type loci. The Drosophila Sir2 also encodes deacetylase activity and is required for heterochromatic silencing, but unlike ySir2, is not required for silencing at telomeres. Drosophila HREF="../polycomb/sir2-1.htm">Sir2 interacts genetically and physically with members of the Hairy/Deadpan/E(Spl) family of bHLH euchromatic repressors, key regulators of Drosophila development. Drosophila Sir2 is an essential gene whose loss of function results in both segmentation defects and skewed sex ratios, associated with reduced activities of the Hairy and Deadpan bHLH repressors. These results indicate that Sir2 in higher organisms plays an essential role in both euchromatic repression and heterochromatic silencing (Rosenberg, 2002).

    To identify the earliest stage at which segmentation is affected, the expression of genes at different tiers of the segmentation gene hierarchy were examined. Protein expression patterns of the gap genes Krüppel and knirps are unaffected in progeny from females with reduced maternal Sir2 (Sir205327/+;wimp/+ or Sir2ex10/+;wimp/+ transheterozygous mothers). Sir2 is first required for regulation of segmentation at the level of pair rule gene expression. Pair rule genes can be separated into two classes: primary pair rule genes establish double segment periodicity, whereas secondary pair rule genes respond to this pattern. Pair rule gene products are expressed as a series of seven transverse stripes in wild-type or wimp/+ embryos. Stripes of the secondary pair rule gene fushi tarazu (ftz) are severely derepressed (stripes are broadened) in embryos from mothers with reduced Sir2 expression. Aberrant regulation of Ftz stripe expression in Sir2 mutant embryos is consistent with reduced function of the primary pair rule gene, Hairy, which behaves genetically as a repressor of ftz. Hairy expression was examined in Sir2 mutant embryos: in contrast to Ftz expression, which is significantly altered, Hairy is largely unaffected in these embryos (Rosenberg, 2002).

    The Ftz derepression phenotype in Sir2 embryos is reminiscent of the Ftz expression pattern seen in hairy mutants. Sir2 was examined for genetic interaction with hairy; these mutations exhibit a dominant genetic interaction. Progeny from either hairy heterozygous mothers or Sir2 heterozygous mothers mated to wild-type males are viable and exhibit wild-type cuticle phenotypes. In contrast, embryos derived from mothers heterozygous for both Sir2 and hairy (Sir2/+; hairy/+ trans-heterozygous mothers) mated to wild-type males exhibit moderate to severe cuticle abnormalities. Consistent with this segmentation cuticle phenotype, Ftz is derepressed in these embryos, with a reduction in expression of stripes 4, 6, and 7, suggesting that these segmentation defects are largely mediated by interaction of Sir2 with Hairy. Interestingly, Hairy stripes 3 and 4 are also affected in progeny from mothers trans-heterozygous for Sir2 and hairy (Sir2/+; hairy/+ females), suggesting interactions between Sir2 and other developmental regulators. Sir2 was tested for interaction with repression cofactors groucho (gro) and dCtBP, as well as the other primary pair rule genes, even skipped (eve), and runt (run). No dominant synthetic lethal interactions were detected between Sir2 and any of these mutations. Hairy was tested for genetic interaction with the class I HDAC, Rpd3, which has been proposed to be recruited to Hairy via the corepressor Groucho. However, no genetic interaction was detected between Rpd3 and hairy (Rosenberg, 2002).

    To test whether Sir2 interacts with Hairy directly, GST pull-down assays were performed using in vitro translated (IVT-) Sir2. Consistent with the genetic results, IVT-Sir2 does not bind to Groucho or dCtBP or to GST alone. However, it does bind specifically to a full-length GST-Hairy fusion protein. To map the region of Hairy required for this interaction, a series of Hairy protein fragments fused in frame to GST were generated. Sir2 binds to all fragments containing the Hairy basic domain, indicating that this domain is sufficient for binding. A series of small basic domain deletions were generated within the context of full-length Hairy protein to identify the smallest region required for Sir2 binding. One of these deletions, DeltaRRAR, disrupts Sir2 binding, while adjacent four amino acid deletions have no effect. The Hairy DeltaRRAR mutation does not affect Hairy homodimerization or binding to other Hairy-interacting proteins, including dCtBP (Rosenberg, 2002).

    The basic domain is highly conserved among HES family proteins, including the invariant RRAR residues, so Sir2 was assayed for binding to other bHLH proteins within this family by GST pull-down. IVT-Sir2 binds to GST-Deadpan (Dpn), but, surprisingly, not efficiently to GST-fusions to the E(Spl)m3 and E(Spl)m8 members of the HES family, suggesting that Sir2 may recognize additional features within the basic domain or in distal regions of HES proteins to permit interaction with a specific subset of these similar proteins in a variety of developmental processes (Rosenberg, 2002).

    One possible consequence of cofactor binding to the basic domain of HES proteins could be interference with their DNA binding abilities. Since Sir2 is required for Hairy function but binds to the basic domain, a gel electrophoretic mobility shift assay (EMSA) was used to test whether Sir2 and Hairy could be detected in a stable complex on DNA. Both full-length-Hairy and a fragment containing the bHLH domain of Hairy are able to efficiently shift 32P-labeled N-box probe and are competed by cold wild-type competitor. No complex with altered mobility was detected upon addition of Sir2, although these proteins are able to interact in vitro. Addition of Sir2 to Hairy either before or after incubation of Hairy with DNA does not prevent Hairy from binding to target DNA, suggesting that Sir2 does not interfere with Hairy binding to DNA and that there may be other consequences of Sir2 binding to Hairy within this region (Rosenberg, 2002).

    Post-transcriptional Regulation

    In common with several transcription units of the E(spl)-C, including E(spl)m4, Bearded contains two novel heptanucleotide sequence motifs in its 3' untranslated region (UTR), suggesting that all these genes are subject to a previously un-recognized mode of post-transcriptional regulation. These sequence motifs are called the Brd box (AGCTTTA) and the GY box (GTCTTCC). Like known sequence elements that function in post-transcriptional regulation, both of these motifs are found in a single orientation and specifically in the UTRs of the genes that include them. Many mRNAs are translationally inactive until they undergo additional cytoplasmic polyadenylation, a process controlled by cytoplasmic polyadenylation elements (CPEs). Polyadenylation is implicated in Brd box function. Negative regulation by the Brd box motif affects steady-state levels of both RNA and protein. This result indicates that Brd boxes have an additional role in regulating translation, beyond the effect attributable to transcript level differences. Thus, the Brd 3' UTR confers negative regulatory activity in vivo. This activity is spatially and temporally general, in that most or all cells are able to respond to Brd boxes. This suggests that some genes expressed outside of proneural clusters may be regulated by these motifs as well. Three other genes that encode negative regulators of PNS development also contain these sequences in their 3' UTRs. In particular, kuzbanian (kuz) and extramacrochaetae (emc) each include single Brd boxes, while hairy contains a GY box. emc also includes four copies of a GY box-related sequence (GTTTTCC) in its 3' UTR, which may be relevant for its regulation. kuz has functions in SOP selection and lateral inhibition, so its expression certainly includes proneural clusters. However, emc and h are expressed in spatial patterns that are largely complementary to proneural clusters in the leg and wing imaginal discs, and are thus possible examples of genes regulated by the Brd box (and possibly the GY box) in territories outside the clusters. Interestingly, the Emc and H proteins, as members of the HLH family, are structurally related to the E(spl)-C bHLH proteins. In contrast, kuz encodes a metalloprotease/disintegrin protein of the ADAM family (Lai, 1997 and references).

    The 3' untranslated regions (3' UTRs) of Bearded, hairy, and many genes of the E(spl)-C contain a novel class of sequence motif, the GY box (GYB, GUCUUCC); extra macrochaetae contains the variant sequence GUUUUCC. The 3' UTRs of three proneural genes include a second type of sequence element, the proneural box (PB, AAUGGAAGACAAU). The full 13 nt PB is found once each in ac, l'sc, and ato, along with a second, variant version in both l'sc and ato. The presence of these motifs in such distantly related paralogs as hairy and certain bHLH genes of the E(spl)-C (for the GYB), and ato and two genes of the AS-C (for the PB), indicates that both classes of sequence element are subject to strong selection. Furthermore, both the PB and the GYB are conserved in the orthologs of ac and E(spl)m4 from the distantly related Drosophilids D. virilis and D. hydei, respectively, though these 3' UTRs are otherwise quite divergent from their D. melanogaster counterparts. These findings strongly suggest functional roles for both of these sequence elements (Lai, 1998).

    Intriguingly, the central 7 nt of the PB and the GYB are exactly complementary, and are often located within extensive regions of RNA:RNA duplex predicted to form between PB- and GYB-containing 3' UTRs. Indeed, using in vitro assays, RNA duplex formation has been observed between the ato/Brd and ato/m4 3' UTR pairs that is PB- and GYB-dependent. It is noteworthy that the predicted duplex interactions involving the GYB of Brd are significantly stronger than those involving the GYBs of the other transcripts. For example, Brd and ato are perfectly complementary over 18 contiguous nucleotides. This difference in the degree of PB:GYB-associated complementarity is likely to have functional consequences (Lai, 1998).

    In C. elegans, small antisense RNAs encoded by lin-4 mediate translational repression of lin-14 and lin-28 transcripts by binding to complementary sequences in their 3' UTRs. In Drosophila, PB- and GYB-bearing transcripts may likewise participate in a regulatory mechanism mediated by RNA:RNA duplexes, but with the feature that both partners are mRNAs that also direct the synthesis of functionally interacting proteins. The opportunity to form such duplexes clearly exists, since transcripts from proneural genes and their regulators very frequently accumulate in coincident or overlapping patterns. Moreover, while 7 nt is the minimum length of complementarity between any PB and any GYB, the longest possible uninterrupted duplex between a given GYB-bearing transcript and a given proneural partner is almost always considerably longer (8-12 nt). It is worth noting that in a lin-4/lin-14 duplex that has been shown to be sufficient for proper regulation in vivo, the longest region of uninterrupted complementarity is only 7 nt (Lai, 1998 and references therein).

    The formation of the postulated RNA duplexes may serve to regulate proneural gene function, consistent with the known roles of hairy, emc, and the bHLH genes of the E(spl)-C. This might explain occasional C-to-U transitions in the GYB sequence (in emc and D. hydei m4); these variants retain complementarity with the PB due to G:U base-pairing. It is equally plausible that GYB-containing transcripts are regulated by duplex formation. A third very interesting possibility is that RNA:RNA duplexes formed between PB- and GYB-containing transcripts function to initiate a downstream regulatory activity affecting as-yet-unknown targets. Ample precedent exists establishing the trans-regulatory potency of double-stranded RNA. In any case, the apparent capacity of transcripts from the proneural genes and their regulators to form duplexes in their 3' UTRs suggests further complexity in the already complex regulatory interactions that control Drosophila neurogenesis (Lai, 1998).

    C-Terminal binding protein (CtBP) interacts with a highly conserved amino acid motif (PXDLS) at the C terminus of adenovirus early region 1A (AdE1A) protein. This amino acid sequence has recently been demonstrated in the mammalian protein C-terminal interacting protein (CtIP) and a number of Drosophila repressors including Snail, Knirps and Hairy. The structures of synthetic peptides identical to the CtBP binding sites on these proteins have been investigated using NMR spectroscopy. Peptides identical to the CtBP binding site in CtIP and at the N terminus of Snail form a series of beta-turns similar to those seen in AdE1A. The PXDLS motif towards the C terminus of Snail forms an alpha-helix. However, the motifs in Knirps and Hairy did not adopt well-defined structures in TFE/water mixtures as shown by the absence of medium range NOEs and a high proportion of signal overlap. The affinities of peptides for Drosophila and mammalian CtBP were compared using enzyme-linked immunosorbent assay. CtIP, Snail (N-terminal peptide) and Knirps peptides all bind to mammalian CtBP with high affinity [K(i) of 1.04, 1.34 and 0.52 microM, respectively]. However, different effects were observed with dCtBP, most notably the affinity for the Snail (N-terminal peptide) and Knirps peptides are markedly reduced [K(i) of 332 and 56 microM, respectively] whilst the Hairy peptide binds much more strongly [K(i) for dCtBP of 6.22 compared to 133 microM for hCtBP]. In addition peptides containing identical PXDLS motifs but with different N and C terminal sequences have appreciably different affinities for mammalian CtBP and different structures in solution. It is concluded that the factors governing the interactions of CtBPs with partner proteins are more complex than simple possession of the PXDLS motif. In particular the overall secondary structures and amino acid side chains in the binding sites of partner proteins are of importance as well as possible global structural effects in both members of the complex. These data constitute evidence for a multiplicity of CtBPs and partner proteins (Molloy, 2001).

    Conserved signals and machinery for RNA transport in Drosophila oogenesis and embryogenesis

    Localization of cytoplasmic messenger RNA transcripts is widely used to target proteins within cells. For many transcripts, localization depends on cis-acting elements within the transcripts and on microtubule-based motors; however, little is known about other components of the transport machinery or how these components recognize specific RNA cargoes. In Drosophila the same machinery and RNA signals drive specific accumulation of maternal RNAs in the early oocyte and apical transcript localization in blastoderm embryos. It has been demonstrated in vivo that Egalitarian (Egl) and Bicaudal D (BicD), maternal proteins required for oocyte determination, are selectively recruited by, and co-transported with, localizing transcripts in blastoderm embryos; interfering with the activities of Egl and BicD blocks apical localization. It is proposed that Egl and BicD are core components of a selective dynein motor complex that drives transcript localization in a variety of tissues (Bullock, 2001).

    Asymmetric RNA localization is evident during zygotic development, especially in the unicellular syncytial blastoderm embryo. At this stage, several transcripts including those of the pair-rule and wingless (wg) segmentation genes lie exclusively apically of the layer of several thousand peripheral nuclei. Localization of these transcripts seems to be mediated by signals within their 3' untranslated regions (UTRs), and to be driven on microtubules by the minus-end-directed molecular motor, dynein. The linkers and other factors that provide the cargo specificity are unknown. Nor is it clear if transcript localization in blastoderm embryos relates to that in other types of cells (Bullock, 2001).

    There is a rapid apical localization of fluorescently labelled fushi tarazu ( ftz) pair-rule transcripts injected into the basal cytoplasm of the cycle 14 blastoderm embryo. Although these experiments indicated a requirement for nuclear proteins fluorescein, labelling compromizes the structure of the transcripts, and pair-rule [even-skipped, hairy (h), ftz, paired and runt] and wg transcripts labelled with several other fluorochromes localize apically within 5-8 min without the need for exogenous protein. Indeed, injected unlabelled transcripts also localize apically. The protein-free assay retains specificity for apical transport, since transcripts that are normally unlocalized [Krüppel (Kr), huckebein] or enriched in the basal cytoplasm (string) are not transported apically and instead diffuse away from the site of injection (Bullock, 2001).

    Blastoderm localization signals can drive transcript transport during oogenesis. This view is supported by more detailed analysis of maternally expressed pair-rule transcripts. The injection assay reveals a minimum region between positions 1,374 and 1,579 in ftz that is necessary and sufficient for localization in blastoderm embryos. A similar region of ftz seems to be required for localization of transcripts into the oocyte. Furthermore, h and runt transcripts, driven maternally by the Hsp70 promoter, also accumulate specifically in the oocyte and later reside at its anterior cortex, whereas Kr or truncated h transcripts lacking most of the 3' UTR fail to localize either in blastoderm embryos or during oogenesis (Bullock, 2001).

    Whether Egl and BicD are present in early embryos was examined. Both proteins are supplied maternally to the embryo. They are noticeably enriched apical to the nuclei at blastoderm stages where they colocalize with dynein heavy chain (Dhc) -- a component of the motor associated with apical transcript transport. Nevertheless, a large proportion of both of the proteins is present in the basal cytoplasm (Bullock, 2001).

    Whether endogenous Egl and BicD can associate with injected localizing transcripts, as might be expected if they are components of the RNA localization machinery, was tested. Injection of h transcripts leads to marked enrichment of Egl and BicD protein levels at the sites of RNA localization. Similar results are found on injection of the other tested maternal and zygotic localizing transcripts ( ftz, bcd, grk, K10, nos, osk and w). Both proteins accumulate basally at the site of injection within 1-2 min. Protein recruitment is not inhibited in embryos preincubated with colcemid, showing that it is not dependent on intact microtubules. Thus, the proteins are recruited locally before transport and are transported together apically with transcripts (Bullock, 2001).

    Whether BicD and Egl are required for apical localization in blastoderm embryos was examined. Strong BicD alleles block oogenesis early, and weaker mutant mothers that lay fertilized eggs (BicDHA40/BicDR26 and BicDH3/BicDR26) retain sufficient BicD activity for a normal apical distribution of endogenous pair-rule transcripts. However, the reduced BicD activity in these embryos no longer supports efficient transport of injected transcripts: 62% of BicDHA40 /BicDR26 and 73% of BicDH3/BicDR26 embryos show no or weak localization 5-8 min after injection, compared with 10% of wild-type embryos. Moreover, an antibody against BicD blocks RNA transport. Preinjection into the basal cytoplasm of anti-BicD antibody 4C2 strongly inhibits the localization of injected h, ftz, grk and stg-K10TLS transcripts in 70%-75% of embryos. The microtubule cytoskeleton is not obviously affected by the brief (~20 min) antibody treatment, indicating that the effects on RNA transport are probably direct. Injection of anti-BicD antibody prevents apical localization of endogenous pair-rule transcripts, also leading to anteroposterior smearing of their distribution. Thus, apical transcript localization seems to be important in restricting the range of activity of pair-rule genes, and allowing their combinatorial control of Drosophila segmentation (Bullock, 2001).

    Injecting blastoderm embryos with anti-Egl also inhibits apical localization of both exogenous and endogenous pair-rule transcripts, without overtly disrupting the microtubule network. Moreover, its effect is more potent in embryos from mothers containing only a single copy of the egl gene, indicating that the antibody disrupts RNA localization by inhibiting the activity of Egl. Egl and BicD are probably also involved in transporting other cargoes. The arrangement of peripheral nuclei is disrupted after injection of antibodies to either of the two proteins, consistent with data showing a requirement for BicD in nuclear migration in eye imaginal disc cells. Embryos injected with either antibody undergo abnormal morphogenesis, which is also indicative of Egl and BicD transporting additional cargoes (Bullock, 2001).

    These results indicate that Egl and BicD are principal elements of a complex that transports RNA in blastoderm embryos. Egl and BicD appear to be present as pools of excess cytoplasmic protein that associate selectively with localizing transcripts and are transported together apically. Protein recruitment occurs before transport and does not require microtubule integrity; rather, transport depends on Egl and BicD activity. Egl and BicD probably act directly to mediate RNA transport associated with establishment and maintenance of the oocyte. Thus, mutant transcripts that are defective in export from nurse cells into the oocyte fail to recruit Egl or BicD in blastoderm embryos. grk transcripts are also recognized by the Egl-BicD-microtubule transport pathway, which is consistent with the hypothesis that nurse cells are a source of these transcripts for the early oocyte and that they do not derive exclusively from the oocyte nucleus (Bullock, 2001).

    Egl/BicD is enriched at sites of RNA localization in both blastoderm embryos and oocytes, presumably as the consequence of protein/RNA co-transport. The complex may have an additional role in anchoring transcripts at their destination. Alternatively, maintenance of localized transcripts might not depend on an independent anchorage step, but result from sustained minus-end-directed transport (Bullock, 2001).

    Dhc, Egl and BicD have markedly similar distributions during oogenesis and in blastoderm embryos, and seem to function together in specifying oocyte identity. It is proposed that an Egl/BicD complex links specific RNAs to dynein and the microtubules. The same machinery may operate elsewhere in Drosophila. For example, inscuteable transcripts, which localize asymmetrically in neuroblasts, also localize apically when injected into blastoderm embryos. Indeed, germline transcripts localize apically when expressed in follicle cells. Egl and BicD homologs have been identified in Caenorhabditis elegans and mammals, and might comprise part of an evolutionarily conserved cytoskeletal system for transporting transcripts and other cargoes (Bullock, 2001).

    A hairy RNA localization signal modulates the kinetics of cytoplasmic mRNA transport

    In several Drosophila cell types, mRNA transport depends on microtubules, the molecular motor dynein and trans-acting factors including Egalitarian and Bicaudal-D. However, the molecular basis of transcript recognition by the localization machinery is poorly understood. The features of hairy pair-rule RNA transcripts that mediate their apical localization have been characterized using in vivo injection of fluorescently labelled mRNAs into syncytial blastoderm embryos. A 121-nucleotide element within the 3'-untranslated region (HLE) is necessary and sufficient to mediate apical transport. The signal comprises two essential stem-loop structures, in which double-stranded stems are crucial for localization. Base-pair identities within the stems are not essential, but can contribute to the efficiency of localization, suggesting that specificity is mediated by higher-order structure. Using time-lapse microscopy, the kinetics of localization has been measured; impaired localization of mutant signals is due to delayed formation of active motor complexes and, unexpectedly, to slower movement. These findings, and those from co-injecting wild-type and mutant RNAs, suggest that the efficiency of molecular motors is modulated by the character of their cargoes (Bullock, 2003).

    Efficient recognition of the h transcript depends on two stem-loops, SL1 and SL2a; each is necessary for robust localization, but neither is sufficient alone. The mutagenesis data demonstrate that evolutionarily conserved double-stranded stems of SL1 and SL2a are indispensable for proper transcript transport. However, secondary structure of the stems is not the sole determinant of signal activity because transversions that alter base-pair identities lead to inefficient localization. In addition, although many predicted single-stranded regions are inessential for signal activity, a mutant in which all of these bases are removed or altered localizes only weakly. Thus, higher-order RNA structure is also likely to be important for specific recognition by the localization machinery (Bullock, 2003).

    Specificity could reside in the tertiary conformation of the RNA, as has been demonstrated for various well-characterized RNA-protein interactions. Such three-dimensional structures would be difficult to infer solely from mutagenesis data, especially because of the large assortment of potential non-canonical interactions between bases (Bullock, 2003).

    Intermolecular RNA interactions could also be involved in h mRNA recognition. Transcript oligomerization appears to be important for localization of bcd transcripts to the anterior of the late oocyte/early embryo, although the signals and trans-acting factors driving this transport process seem to be distinct from those acting in Egl/BicD-mediated early export into the oocyte. Studies of mutant h RNA transcripts have not yet revealed evidence that oligomerization is necessary for transport. Mixtures of up to three different transcripts were injected; rescue of a non-localizing mutant RNA signal by a co-injected localizing transcript has never been detected. Nonetheless, time-lapse studies show transport of injected h RNA in particles that contain numerous transcript molecules, although the imaging is not sensitive enough to detect cargoes of individual molecules. Indeed, particle formation is not sufficient to direct formation of an active transport complex: non-localizing mRNAs are also found in particles (Bullock, 2003).

    Although K10 and bcd localization signals share no obvious primary sequence similarities with the HLE, they share structural features, suggesting that they are recognized similarly. The K10 localization signal is only 44 nt long and, unlike the HLE, comprises only a single stem-loop region; nonetheless, it recruits Egl and BicD. bcd transcripts also harbor a stem-loop (the 57-nt stem-loop V) that is required for early transport from nurse cells into the oocyte and for apical localization of injected bcd transcripts in the embryo and their association with Egl and BicD. Like h, the activities of both the bcd and K10 stem-loops rely heavily on double-stranded stems in which exact base-pair identities contribute to, but do not determine, efficient localization; base-pair transversions in all the stems can compromise the efficiency of localization. In common with the h SL1, the bcd stem-loop V is not sufficient for localization, but is fully active when dimerized (Bullock, 2003 and references therein).

    The apparent complexity and redundancy of the HLE supports a model for signal recognition in which multiple protein-RNA contacts are needed for the formation of a specific, stable complex. In the HLE, weak binding sites for the machinery may be distributed in SL1 and SL2a. Thus, transcripts with two h SL1 domains are at least as active as those with a wild-type HLE. SL2a may provide quantitatively weaker binding signals; it is unable to support any localization either alone or when multimerized. One possibility is that SL1 alone establishes low affinity interactions with the localization machinery, and binds with high affinity together with SL2. The same mode of recognition could also apply for K10, if, unlike h and bcd, the requisite sites are located within a single stem-loop (Bullock, 2003).

    Despite the overall similarities of the structural requirements for localization of bcd, K10 and h, no significant shared base-pair identities were identified within essential regions of the signals. The possibility that different transcripts are recognized by distinct RNA-binding factor(s) and recruited to shared components of the machinery cannot be excluded. However, the same localization signals are active in a variety of cell types. Also, stem-loops from different transcripts, each of which is relatively inactive in isolation, can complement to mediate completely efficient localization when combined in the same transcript. Thus, the view is favored that different transcripts share similar higher-order features, such as tertiary RNA conformations of the stems or RNA oligomers, which are recognized by the same factor(s). Multiple RNA motifs per signal and/or RNA or protein oligomerization would lead to the formation of the multiple protein-RNA contacts that confer specificity (Bullock, 2003).

    RNA/BicD/Egl association appears to be a prerequisite for transport. BicD is unlikely to bind RNA directly because it lacks a known RNA-interaction domain, but BicD could hetero- or homo-oligomerize via its heptad repeat domains and thereby increase the numbers of protein-RNA contacts. Egl includes a domain with homology to certain 3'-5' exonucleases and a variety of other nucleic acid-interacting proteins, and thus might recognize RNA directly. However, its ability to recognize specific RNA sequences or structures has yet to be demonstrated (Bullock, 2003).

    Differential cytoplasmic mRNA localisation adjusts pair-rule transcription factor activity to cytoarchitecture in dipteran evolution

    Establishment of segmental pattern in the Drosophila syncytial blastoderm embryo depends on pair-rule transcriptional regulators. mRNA transcripts of pair-rule genes localise to the apical cytoplasm of the blastoderm via a selective dynein-based transport system and signals within their 3'-untranslated regions. However, the functional and evolutionary significance of this process remains unknown. Subcellular localisation of mRNAs from multiple dipteran species has been analyzed both in situ and by injection into Drosophila embryos. Transcript localisation was assayed in four species that can be cultured in the laboratory. Two of them, Episyrphus (Syrphidae) and Megaselia (Phoridae), are cyclorrhaphan flies (i.e. higher dipterans) but, unlike Drosophila, belong to basal branches of this taxon; the other two, Coboldia (Scatopsidae) and Clogmia (Psychodidae), belong to different branches of lower Diptera. Although localisation of wingless transcripts is conserved in Diptera, localisation of even-skipped and hairy pair-rule transcripts is evolutionarily labile and correlates with taxon-specific changes in positioning of nuclei. In Drosophila localised pair-rule transcripts target their proteins in close proximity to the nuclei and increase the reliability of the segmentation process by augmenting gene activity. These data suggest that mRNA localisation signals in pair-rule transcripts affect nuclear protein uptake and thereby adjust gene activity to a variety of dipteran blastoderm cytoarchitectures (Bullock, 2004).

    Apical localisation of pair-rule mRNAs in Drosophila syncytial blastoderm embryos was first noted 20 years ago, but the developmental and evolutionary significance of this process has remained unclear. Apical pair-rule mRNA localisation is conserved in cyclorrhaphan species that diverged over 145 million years ago, indicating that this process has a significant developmental role under natural conditions. Likewise, the widespread maintenance of wg transcript localisation in Diptera supports the importance of this process on a phylogenetic scale, even though, in Drosophila, wg appears to be less sensitive than pair-rule genes to a reduction in endogenous transcript localisation (Bullock, 2004).

    Unlike wg transcripts, pair-rule mRNAs do not localise in some branches of lower Diptera, and the phylogenetic occurrence of this process provides interesting insights into its functional significance. Enrichment of pair-rule transcripts in the apical cytoplasm correlates with the position of blastoderm nuclei: efficient apical localisation of pair-rule gene transcripts is found in species which retain an asymmetric apical position of nuclei throughout the blastoderm stage (Drosophila, Megaselia); less efficient localisation is seen when the nuclei move from an apical to a more central position during blastoderm stages (Episyrphus), and no apical enrichment of transcripts is seen in species where blastoderm nuclei are surrounded uniformly by a thin layer of cytoplasm (Coboldia, Clogmia). Localisation signals are also found in several pair-rule transcripts of the lower dipteran Anopheles. Like Cyclorrhapha, but unlike many other lower Diptera and most other insects, this culicid species has evolved a thickened blastoderm with apically positioned nuclei, probably to allow rapid development as an adaptation to ephemeral larval habitats: columnar cells that emerge from thickened blastoderms can enter gastrulation directly, whereas cuboidal cells that emerge from thin blastoderms still have to elongate prior to undergoing the requisite cell shape changes (Bullock, 2004).

    In Drosophila, pair-rule proteins are enriched in the apical cytoplasm prior to import into the nuclei in wild-type blastoderms, whereas they are detected basally in egl mutant embryos, in which transcript localisation is inefficient. The apical accumulation of pair-rule proteins under normal circumstances is consistent with the observation that apical RNA targeting restricts diffusion of cytoplasmic ß-galactosidase. Apically targeted protein is most likely confined by the cellularisation process, in which the plasma membrane invaginates between the nuclei and encloses the apical compartment first (Bullock, 2004).

    It has been speculated that mRNA localisation prevents pair-rule proteins from moving into inter-stripe regions, where they would cause dominant patterning defects. However, when pair-rule mRNA localisation is compromised, either by interfering with the localisation machinery or the RNA signals, no expansion of RNA or protein stripes or ectopic phenotypic effects are found. Rather, a reduction of pair-rule activity is seen in their domains of expression in these experiments, indicating that transcript localisation augments gene function. Pair-rule mRNA localisation does not appear to be obligatory for protein activity in Drosophila but makes the segmentation process more reliable: egl mutants, in which transcripts localise very inefficiently, have a mild increase in segmentation defects and are acutely sensitive to the reduction of pair-rule gene dose (Bullock, 2004).

    By what mechanism does pair-rule mRNA localisation augment the activity of their transcription factor products? For h it is demonstrated that suppression of transcript localisation reduces nuclear levels of its protein. Pair-rule proteins could be specifically modified in the apical cytoplasm, or localising transcripts could be translated more efficiently. However, given the diffuse distribution of pair-rule proteins in the basal cytoplasm when RNA localisation is disrupted in egl mutants and the correlation between cytoarchitecture and pair-rule transcript localisation in Diptera, a third possibility is favored, namely that apical mRNA localisation increases nuclear uptake of their proteins by targeting translation in close proximity to the nuclei. Proteins from non-localising mRNAs would not be available at high levels in the immediate vicinity of the nuclei, which would result in a decreased nuclear uptake. Such a role for apical pair-rule mRNA localisation would be redundant in lower Diptera with only a thin layer of cytoplasm surrounding the nuclei, which provides little room for diffusion of pair-rule proteins prior to nuclear import. A mechanism for perinuclear protein targeting might be particularly significant for nuclear proteins with short half-lives, such as those encoded by pair-rule genes. Interestingly, localisation of mRNA in the vicinity of the nucleus to aid import of nuclear proteins has also been reported in cultured mammalian cells and may be a widespread mechanism to efficiently exploit a limited pool of transcripts in cells that are polarised or have a high cytoplasmic:nuclear ratio (Bullock, 2004).

    The relationship between cytoarchitecture and apical pair-rule transcript localisation does not appear to be absolute because a signal is detected in eve, but not h, from Haematopota, which has retained the ancestral, cuboidal blastoderm morphology and because no localisation signal was detected in Anopheles-eve. Although the developmental context in which these signals are used cannot yet be discerned (in situ hybridisation is currently not possible in these species because of egg shells that are difficult to remove and because of difficulties in obtaining embryos) these data raise the possibility that, within a single species, the differential ability of transcripts to be recognised by the localisation machinery is used to fine-tune transcriptional control of target genes in the blastoderm by modulating the nuclear concentration of pair-rule proteins (Bullock, 2004).

    The ability of eve and h pair-rule transcripts to use the localisation machinery varies in Diptera. A range of localisation efficiencies is observed in situ that is mirrored in all of 11 cases, upon injection into Drosophila embryos. Thus, differences in localisation efficiency appear to reflect changes in the respective localisation signals, rather than alterations in the specificity of the protein machinery. These findings are consistent with previous studies with artificial variants of the Drosophila hairy localisation signal, which suggest that the character of localisation signals modulates the efficiency of localisation by determining the kinetics of both the initiation of transport and the transport process itself. Localisation efficiency appears to be determined by multiple RNA:protein interactions, the sum of which affects the stability and/or activity of the RNA:motor complex. Therefore, the efficiency of the localisation process can be modified gradually during evolution by the addition, loss or modification of individual recognition sites within mRNAs (Bullock, 2004).

    It seems that localisation signals in pair-rule genes have emerged multiple times within Diptera. For example, although the possibility that localisation signals in h have been lost in multiple different lineages of lower Diptera cannot be ruled out, the most parsimonious explanation for the phylogenetic distribution of signals in this transcript is that they evolved independently in response to changes in cytoarchitecture in the lineages leading to Cyclorrhapha and Culicomorpha. Injection of transcripts from additional species into Drosophila will determine whether eve localisation signals emerged independently in the lineages leading to Haematopota and Cyclorrhapha, or were lost in the lineage leading to Empis (Bullock, 2004).

    Work in mammalian cells has provided insights into how localisation signals might initially appear. These studies suggest that non-localising mRNAs can also interact with a motor complex, albeit with a comparatively small probability, and undergo short movements on microtubules. Localisation signals appear to augment these interactions and lead to the net translocation of an RNA population along a polarised cytoskeleton by increasing the frequency and duration of directed transport. The localisation machinery in Diptera may also have a general, weak affinity for mRNAs because a small proportion of particles of injected non-localising transcripts are transported over short distances in Drosophila embryos. Asymmetric accumulation of a population of transcripts may therefore evolve gradually as a result of selection for increased interaction between a specific transcript and the localisation machinery (Bullock, 2004).

    Egalitarian is a selective RNA-binding protein linking mRNA localization signals to the dynein motor

    Cytoplasmic sorting of mRNAs by microtubule-based transport is widespread, yet very little is known at the molecular level about how specific transcripts are linked to motor complexes. In Drosophila, minus-end-directed transport of developmentally important transcripts by the dynein motor is mediated by seemingly divergent mRNA elements. Evidence is provided that direct recognition of these mRNA localization signals is mediated by the Egalitarian (Egl) protein. Egl and the dynein cofactor Bicaudal-D (BicD) are the only proteins from embryonic extracts that are abundantly and specifically enriched on RNA localization signals from transcripts of gurken, hairy, K10, and the I factor retrotransposon. In vitro assays show that, despite lacking a canonical RNA-binding motif, Egl directly recognizes active localization elements. A physical interaction was revealed between Egl and a conserved domain for cargo recruitment in BicD and data is presented suggesting that Egl participates selectively in BicD-mediated transport of mRNA in vivo. This work leads to the first working model for a complete connection between minus-end-directed mRNA localization signals and microtubules and reveals molecular strategies that are likely to be of general relevance for cargo transport by dynein (Dienstbier, 2009).

    Many proteins achieve an asymmetric localization within the cytoplasm through the transport of their mRNAs along the cytoskeleton by molecular motors. Despite the widespread occurrence of mRNA transport, the detailed mechanisms by which specific transcripts are recognized and recruited to motor complexes are poorly understood. One exception is during bud-specific enrichment of mRNAs along actin filaments in the yeast Saccharomyces cerevisiae, where proteins have been identified that can account for a complete link between localizing mRNAs and the cytoskeleton. However, many metazoans rely on microtubules to deliver mRNAs over the requisite longer distances, and mechanistic insights into how these transcripts are linked to motors are relatively sparse (Dienstbier, 2009).

    One of the best prospects for elucidating microtubule-based mRNA transport is in the Drosophila syncytial blastoderm embryo, where a pathway for apical localization of a subset of endogenous mRNAs can be accessed by microinjection of in vitro synthesized, fluorescently labeled transcripts. Consistent with the nucleation of the minus ends of the microtubules in the apical cytoplasm, localization of these transcripts is driven by cytoplasmic dynein together with its accessory complex dynactin. Related machinery delivers mRNAs to the minus ends of microtubules in other Drosophila cell types, including oocytes and neuroblasts (Dienstbier, 2009).

    The cis-acting RNA elements mediating asymmetric localization by dynein have been studied in detail for seven transcripts (the developmentally important mRNAs bicoid [bcd], fushi tarazu [ftz], gurken [grk], hairy [h], fs(1)K10 [K10], and wingless [wg], and the I Factor retrotransposon RNA) and contain one or more stem-loop structures. These 'localization signals' are necessary for minus-end-directed localization and also sufficient when inserted into heterologous transcripts (Dienstbier, 2009).

    The localization signals in the different transcripts do not share significant primary sequence similarity and often have different lengths. This has led to two competing models: the first in which the RNA elements contain cryptic features that associate with a common recognition machinery, and the second in which they are recognized by different proteins, each able to independently provide a link to the dynein complex. It has not been possible to discriminate between these scenarios, because proteins that specifically bind any of these elements and are required for transport have not been identified (Dienstbier, 2009).

    In addition to dynein/dynactin, the Egalitarian (Egl) and Bicaudal-D (BicD) proteins are also essential for targeting of mRNAs to the minus ends of microtubules. Egl and BicD are found in a complex with each other in vivo (together with other copies of themselves), although it is not known whether they interact directly. Egl and BicD also associate with dynein light chain (Dlc) and the dynein/dynactin complex, respectively, and are recruited to injected localizing mRNAs in embryos to bias the net movements of a bidirectional mRNA transport complex apically. Together, these observations have led to a model in which Egl and BicD associate with localization signals and increase the frequency of minus-end-directed dynein/dynactin movements. Because neither Egl nor BicD has a known RNA-binding motif, it has been reasoned that they are recruited to localization signals by intermediary factors that directly contact the message (Dienstbier, 2009 and references therein).

    Whether Egl has roles outside of mRNA transport has not been reported, but BicD functions in the transport of a subset of other cargoes for dynein. It has been proposed that the N-terminal two-thirds of mammalian BicD are sufficient for stimulating dynein transport and that the remaining C-terminal sequences (hereafter referred to as the CTD [C-terminal domain]) mediate a link between cargoes and the motor. This is based on the findings that the CTD can be functionally substituted by heterologous motifs for organelle recognition and can bind Rab6, a membrane-linked GTPase that recruits dynein to Golgi vesicles (Dienstbier, 2009).

    This study attempts to elucidate the mechanism of linkage of different mRNA localization signals to dynein. The surprising finding is reported that Egl is a selective RNA-binding protein that directly contacts active localization signals. Thus, seemingly divergent mRNA signals are recognized by the same factor. Egl associates with a conserved domain for cargo recruitment in BicD and is selectively required for mRNA transport in vivo. This work provides unique insights into the molecular links between localizing mRNAs and microtubule-based motors, and also sheds light on general mechanisms of cargo transport by dynein (Dienstbier, 2009).

    Because of difficulties in finding shared features between dynein-dependent localization signals in different transcripts, it was not known whether dedicated factors are responsible for recognizing each of these elements. This uncertainty has severely restricted the ability to generalize conclusions from studies of localization mechanisms of individual transcripts. This work demonstrates that the same protein, Egl, is capable of specifically contacting minus-end-directed localization signals from multiple different transcripts. This conclusion is supported by the findings that (1) Egl and BicD are the only factors visibly enriched from embryonic extracts on all four localizing elements tested relative to a number of nonlocalizing controls, (2) Egl function in Drosophila is required for BicD-mediated transport of mRNAs and not other cargoes tested, (3) the majority of Egl, but not BicD, in cell extracts is found in a complex whose size is sensitive to Rnase treatment, and (4) recombinant Egl, but not BicD, binds RNA in vitro and is capable of discriminating between active apical localization signals and those containing subtle inactivating mutations (Dienstbier, 2009).

    In addition to the four elements tested in this study, Egl is also likely to associate directly with other mRNA localization signals because bcd, ftz, and wg recruit Egl in vivo and depend on its function for minus-end-directed transport. Indeed, Egl binding may be the major, and perhaps only, specific determinant of the activity of an apical localization signal, as all three subtle inactivating mutations that were tested inhibit association of Egl from embryonic extracts (TLSδbub, TLSU6C, and hSL1C15G), and a fourth inactive point mutant (bcdSLV4496G-U) prevents recruitment of Egl to bcd injected into embryos. Presumably, despite differences in primary sequence composition, all of the characterized localization elements contain cryptic structural features that are recognized by Egl. Elucidating the structural basis of this recognition event will be the goal of future long-term studies (Dienstbier, 2009).

    Interestingly, Egl exhibits some affinity for inactive localization elements when expressed recombinantly, as well as in embryonic extracts. Egl may well exhibit greater selectivity for active signals in the appropriate in vivo context. This could be because the composition of in vitro binding buffers is suboptimal. Alternatively, the incorporation of mRNAs into oligomeric particles within the cell may give rise to cooperative interactions between individual Egl and BicD complexes, thereby increasing cargo specificity. Nonetheless, an inherent degree of promiscuity by Egl in vivo would fit with a previous finding that its overexpression in embryos is sufficient to target a small amount of an endogenous nonlocalizing transcript population to the apical cytoplasm (Bullock, 2006) and could also be the basis of repeated emergence of apical localization signals during dipteran evolution (Dienstbier, 2009).

    The mRNA elements that direct apical transport in the blastoderm embryo are also capable of mediating localization of transcripts toward the minus ends of microtubules during oogenesis. It is therefore very likely that direct binding of Egl to these stem-loops is also functionally significant during these stages. Indeed, Egl and BicD have been shown to be components of motor complexes that transport grk from the nurse cells into the oocyte. Interestingly, within the oocyte the h and K10 elements are involved in localization to the anterior cortex, whereas those in grk and the I factor are also sufficient for translocation from the anterior to the dorso-anterior corner. Dorsalward movement is presumably due to the binding of the ILS and GLS to oocyte-specific factors in addition to Egl, either sequentially or simultaneously, or by modulating the mode of Egl binding (Dienstbier, 2009).

    It has been shown that Egl and BicD are in a complex together in vivo. The current data shows for the first time that Egl, through its N-terminal 79 amino acids, directly interacts with BicD. In addition, Egl also binds Dlc through a consensus light chain-binding site between amino acids 963 and 969 (Navarro, 2004). BicD is able to recruit the dynein/dynactin complex (Hoogenraad, 2003) and Dlc associates with other dynein subunits. Thus, together with evidence for Egl RNA binding through amino acids 1-814, it is now possible to build a working model of a complete link between minus-end-directed mRNA signals and microtubules for the first time (Dienstbier, 2009).

    Egl, BicD, and mRNA elements do not appear to be obligatory for particle assembly or bidirectional mRNA motility (Bullock, 2006). Instead, they are likely to be essential parts of a cassette that up-regulates minus-end-directed movement of a generic bidirectional mRNA transport complex. Other RNA-binding factors presumably package both localizing and nonlocalizing RNAs and provide additional links to motors (Dienstbier, 2009).

    Within the minus-end regulatory cassette, the role of Egl is probably to recruit both BicD and Dlc to the mRNA to ensure efficient targeting of transcripts to the minus ends of microtubules. The presence of both Egl-interacting partners might be required for the stability of the motor complex. Alternatively, previous observations of the effects of altering protein concentrations on mRNA transport are consistent with Egl-Dlc and Egl-BicD interactions regulating different aspects of motility of the bidirectional motor complex: processivity and switching behavior, respectively (Bullock, 2006). Like Egl, Rab6 is able to associate with both BicD and a Dlc. Association with both BicD and Dlc may therefore be a common strategy used by cargo adaptors to ensure efficient minus-end-directed transport (Dienstbier, 2009).

    Binding of both Egl and Rab6 to BicD is sensitive to the same amino acid substitution in the CTD. Egl and Rab6 recognize localizing mRNAs and Golgi vesicles, respectively, raising the possibility that BicD functions in the transport of different cargoes through mutually exclusive association of the CTD with cargo-specific adaptors. It was found that relatively subtle overexpression of Egl not only augments BicD-dependent apical mRNA transport (Bullock, 2006), but also antagonizes BicD function in lipid droplet motility. This implies that, through competition for the BicD CTD, the pathways for microtubule-based transport of different cargoes can be finely balanced. Alteration of the availability of adaptors for BicD is therefore a potentially effective strategy for regulating net sorting of cargoes (Dienstbier, 2009).

    Experiments involving the tethering of cargoes to BicD domains also shed light on potential general mechanisms of dynein-based cargo transport. As is the case for mammalian BicD, removal of the CTD of the Drosophila protein stimulates transport by dynein. This situation presumably mimics a version of the full-length protein bound to a cargo adaptor in which an autoinhibitory effect of the C terminus is negated. The N terminus of BicD can efficiently capture dynein/dynactin components from cell extracts, suggesting that this interaction could be entirely sufficient for productive transport. However, the results indicate that, at least in Drosophila, the capacity of BicDδC to mediate net movement of tethered cargoes is dependent on its association with endogenous BicD transport complexes. Such a scenario was not directly tested in the previous mammalian cell assays (Dienstbier, 2009).

    In the case of minus-end-directed mRNA transport in flies, the CTD appears to provide an essential link, through Egl, to Dlc. In addition, the CTD can associate with the dynamitin subunit of dynactin. The significance of this interaction was not clear in light of a model in which only the N-terminal sequences of BicD are important for transport by dynein. The finding that the CTD is needed in trans for the activity of BicDδC revives the possibility that the dynamitin interaction is functionally important (Dienstbier, 2009).

    The ability of BicDδC::CP, but not BicD::CP, to target heterologous cargoes apically is likely to reflect a role for the CTD in inhibiting association with other copies of BicD. Consistent with this notion, BicDδC::CP accumulates in large, apically enriched puncta, whereas the full-length protein fused to the coat protein fails to form discrete particles and has a uniform distribution. Together with the observation that BicD is able to associate with other copies of itself in vivo, these results imply that dimerization or oligomerization of BicD could be an important step in the activation of transport by cargo binding. Future experiments will be aimed at determining the copy number of components of the transport complex in the presence and absence of a bound consignment (Dienstbier, 2009).

    The influence of dynein processivity control, MAPs, and microtubule ends on directional movement of a localising mRNA

    Many cellular constituents travel along microtubules in association with multiple copies of motor proteins. How the activity of these motors is regulated during cargo sorting is poorly understood. This study addressed this issue using a novel in vitro assay for the motility of localising Drosophila mRNAs bound to native dynein-dynactin complexes. High precision tracking reveals that individual RNPs within a population undergo either diffusive, or highly processive, minus end-directed movements along microtubules. RNA localisation signals stimulate the processive movements, with regulation of dynein-dynactin's activity rather than its total copy number per RNP, responsible for this effect. These data support a novel mechanism for multi-motor translocation based on the regulation of dynein processivity by discrete cargo-associated features. Studying the in vitro responses of RNPs to extrinsic microtubule-associated proteins (MAPs) and microtubule ends provides insights into how an RNA population could navigate the cytoskeletal network and become anchored at its destination in cells (Soundararajan, 2014).

    Several previous studies have manipulated the copy numbers of cargo-associated motors in order to elucidate how multiple motors orchestrate sorting. One experimental approach has been to alter the numbers of isolated motors or motor domains attached to artificial cargos such as beads or DNA origami. Although very informative, these studies did not include physiological cargo, cargo adaptors, and motor co-factors that could potentially modulate motor behaviour. Other studies have used genetic manipulations to alter the numbers of motor complexes available to cargos in vivo. These approaches are physiologically relevant but cannot rule out influences from the cellular environment, including possible indirect effects of altered motor concentration on other processes that impinge on cargo motility. This study has manipulated the copy number of native motor complexes on a physiological type of cargo, by incubating cellular extracts with RNA variants, and studied the consequences on motility in a defined in vitro setting using high spatiotemporal resolution imaging (Soundararajan, 2014).

    hairy (h) wild-type ribonucleoprotein complexes (RNPs) associate with dynein-dynactin and can undergo either unidirectional motion in the minus end direction that is highly processive or bidirectional motion that has characteristics of a diffusive process. The experiments indicate that unidirectional RNP movement is driven by active, ATP hydrolysis-mediated translocation of dynein along the microtubule. The most parsimonious explanation for the bidirectional motion is that it is also due to dynein undergoing back-and-forth movements along the microtubule, a behaviour that has been observed in several studies of the purified motor in vitro and also appears to occur in vivo. Indeed, bidirectional movement was observed of a significant subset of RNPs containing only the h localisation element HLE (~30% in the absence of MAPs and ~60% in their presence), on which binding of a kinesin family member was not observed under conditions in which dynein and dynactin were readily detected. Interestingly, bidirectional motion of RNPs is not overtly sensitive to inhibition of dynein's ability to hydrolyse ATP. Thus, these findings are compatible with those of of a previous study that reported passive diffusion along microtubules of dynein in complex with a dynactin component in the presence of ATP or ATP-vanadate. No evidence was fpimd that bidirectional RNPs can undergo long, ATP hydrolysis-dependent runs in both directions akin to those documented for individual, GFP-labelled dynein-dynactin complexes purified from mouse brain (Soundararajan, 2014).

    Varying the number of SL1 elements within the context of the h 3'UTR revealed a correlation between the total copy number of dyneins per RNP and the probability of entering into the unidirectional, minus end-directed state. Previous studies using purified dynein bound to artificial cargos have demonstrated that increasing motor copy number is sufficient to stimulate processive movement towards minus ends. However, experiments with the isolated HLE signal indicated that total motor number is not the key determinant of the unidirectional mode of RNP movement. The HLE alone has a statistically indistinguishable copy number of dynein components to the hΔLE RNA, in which the localisation signal has been replaced by a heterologous sequence in the context of the h 3'UTR, yet only the former RNA is capable of unidirectional motion. These data suggest that features associated with the RNA signal are sufficient to increase the probability of processive movement of the associated dynein. It has recently been shown in Schizosaccharomyces pombe that microtubule-associated dynein can switch from diffusive to processive behaviour upon contacting cortical anchors, an event that regulates the generation of pulling forces on the microtubul. The current data suggest that regulation of dynein processivity by associated factors may be a widespread phenomenon (Soundararajan, 2014).

    Collectively, these results support a novel model in which the same cargo species can interact with processive or non-processive dynein, with discrete cargo-associated features regulating the probability of switching between the two behaviours. Thus, the regulatory mechanisms underpinning sorting of RNPs in this system appear distinct from those of other well-studied bidirectional cargos, which involve the interplay of opposite polarity force-generating motors, such as dyneins and kinesins. How might localisation signals regulate dynein processivity? One possibility is that proteins recruited by the localisation signal directly regulate the activity of the motor. Two candidates to serve such a role are the adaptor proteins Egl and BicD, which are associated only with the dynein-dynactin bound to the localisation signals. It is also conceivable that the structure or rigidity of the RNA signal plays an architectural role in presenting dynein-dynactin to the microtubule in a manner that favours processive movement (Soundararajan, 2014).

    This study also sheds light on how physiological cargo-motor complexes respond to extrinsic factors within the cytoskeletal environment. RNPs frequently reverse when encountering the regions of microtubules bound by puncta of both MAPs studied, rigor kinesin and tau23. The behaviour of RNPs at MAPs is therefore highly reminiscent of that seen for individual, purified dynein-dynactin complexes in vitro. Interestingly, comparison of the behaviour of the h RNA variants reveals that the number of dynein-dynactins associated with an RNP does not increase the probability of passing microtubule-associated obstacles (Soundararajan, 2014).

    The ability of RNPs to reverse at MAPs may help them navigate to their destination in vivo. For example, reversals upon meeting an obstacle may facilitate encounters of RNPs with intersecting microtubules. Switching of RNPs between microtubules, a behaviour observed when intersections occur in in vitro assays, could allow these complexes to explore alternative routes to their destination. Reversals of RNPs at a MAP may also give dynein space to switch to a different lateral position on the same microtubule and thereby provide another opportunity to pass the obstacle following resumption of movement in the previous direction. Compatible with this notion, RNPs were seen moving on single microtubules that can pass obstacles after multiple attempts (Soundararajan, 2014).

    Analysis of the behaviour of RNPs at microtubule ends indicates that plus end encounters always result in a reversal. The behaviour may also be advantageous in a cellular environment by preventing detachment of RNPs at this point, an outcome that would necessitate a rebinding event before motion on the microtubule network can resume. Interestingly, a subset of RNPs undergoes pausing at the minus ends of microtubules, with a mean dwell time of ~1 min for bidirectional complexes. These findings demonstrate that the probability of changing directions is different at the minus end and the plus end of the microtubule and that additional in vivo features, such as the γ-tubulin ring complex or other centrosome-associated factors, are not obligatory for long-term retention of RNPs at the minus end. Intrinsic behaviours of RNPs upon reaching the minus end of the α/β-tubulin polymer may therefore contribute to the dynein-dependent anchorage of RNAs in the vicinity of minus ends in vivo. Analysis of hΔLE, hWT, and hSL1x3 RNAs demonstrates that the probability of an RNP undergoing minus end pausing in vitro, as well as the duration of such events, is not influenced by the addition of more dynein-dynactins through localisation elements. This finding offers an explanation for why inhibition of Egl and BicD following translocation of localising RNAs to the apical cytoplasm of the embryo does not affect dynein-dependent anchorage (Soundararajan, 2014).

    It is intriguing that the average dwell time of pausing events of unidirectional hWT and hSL1x3 RNPs at minus ends is ~ fivefold less than that of bidirectional RNPs harbouring the same RNA species. It was also found that the unidirectional HLE RNPs, which lack features within the h 3′UTR that can recruit non-processive dynein-dynactin, dwell at the minus end for significantly less time than the unidirectional hWT and hSL1x3 RNPs. One explanation for these findings is that the ability of dynein bound to localisation signals to walk processively off the minus end is antagonised by interactions with the microtubule mediated by non-processive dynein bound at other sites in the RNA. Additional, long-term experiments will be required to test this hypothesis. Nonetheless, the data suggest more generally that features that promote bidirectional motion could assist in the retention of RNPs at minus ends (Soundararajan, 2014).

    Collectively, this analysis of encounters of RNPs with MAPs and microtubule ends raises the possibility that the co-existence of unidirectional or bidirectional modes of movement facilitates efficient navigation of an RNA population to its destination in vivo. Processive, unidirectional movement in the minus end direction could be beneficial for rapid, directional movement along regions of the microtubule that are not rich in MAPs. Diffusive motion along microtubules may be valuable for movement of RNPs through an obstacle rich environment and could still contribute to asymmetric sorting as it is associated with long-term retention of complexes at microtubule minus ends. This strategy appears analogous to that used by DNA enzymes and kinesins that depolymerise microtubules, which can employ one-dimensional diffusion to search for their specific target sites (Soundararajan, 2014).

    This in vitro work on RNA motility has provided several new insights into how cargo-motor complexes operate and how their behaviour is modulated by encounters with the environment. The data lead to a model in which discrete cargo-associated features regulate motor processivity, a phenomenon that could not have been recapitulated using minimal motor elements coupled to artificial cargos. The results also illustrate that intrinsic motile properties of cargo populations in vivo are likely to be obscured by the influence of extrinsic factors including MAPs and microtubule ends. Further exploitation of the RAT-TRAP assay is likely to be an effective strategy for shedding light on molecular mechanisms that underpin intrinsic and extrinsic regulation of cargo motility, particularly when combined with powerful Drosophila gene perturbation techniques. In the longer term, it will be important to understand how the behaviours defined in vitro are integrated during sorting of single RNA molecules in vivo, a goal that necessitates the development of new methods to visualise movement of transcripts in the optically challenging embryo system (Soundararajan, 2014).

    Gene-specific targeting of the histone chaperone asf1 to mediate silencing: Asf1 can be coprecipitated with Su(H), the corepressor Hairless, and Groucho

    The histone chaperone Asf1 assists in chromatin assembly and remodeling during replication, transcription activation, and gene silencing. However, it has been unclear to what extent Asf1 could be targeted to specific loci via interactions with sequence-specific DNA-binding proteins. This study shows that Asf1 contributes to the repression of Notch target genes, as depletion of Asf1 in cells by RNAi causes derepression of the E(spl) Notch-inducible genes. Conversely, overexpression of Asf1 in vivo results in decreased expression of target genes and produces phenotypes that are strongly modified (enhanced and suppressed) by mutations affecting the Notch pathway, but not by mutations in other signaling pathways. Asf1 can be coprecipitated with the DNA-binding protein Su(H) and the corepressor Hairless and interacts directly with two components of this complex, Hairless and SKIP. Thus, in addition to playing more general roles in chromatin dynamics, Asf1 is directed via interactions with sequence-specific complexes to mediate silencing of specific target genes (Goodfellow, 2007).

    Modulation of the chromatin structure is a key feature in transcriptional regulation. Chromatin remodeling by ATP-dependent enzymes and posttranslational histone modifications are two important mechanisms that affect transcriptional activity, by influencing the accessibility of upstream regions and promoters. A third mechanism involves the breakdown and reassembly of nucleosomes on the DNA, a process that also allows for the incorporation of histone variants, such as H3.3. Histone chaperones, which bind to histone heterodimers, are required both for nucleosome assembly and for their disassembly. They include the H3/H4 chaperone Anti-silencing factor 1 (Asf1), which has roles in replication-dependent and replication-independent chromatin dynamics (Goodfellow, 2007).

    In yeast, extensive Asf1-mediated exchange of histones that is independent of replication and of transcription has been detected at gene promoters and is likely to be highly significant in maintaining the balance between induction and silencing of genes. Indeed, there are now several examples of yeast Asf1 contributing to chromatin disassembly at promoters to facilitate binding of the RNA-polymerase complex. Conversely, Asf1 also plays important roles in gene silencing when the reassembly of nucleosomes accompanies transcriptional repression. For example, in the absence of Asf1, there is a delay in promoter closure at the PHO5 gene. However, it remains unclear whether Asf1-mediated nucleosome reassembly occurs via a targeted mechanism, involving sequence-specific DNA-binding proteins, or whether it occurs constitutively by default (Goodfellow, 2007).

    A strong correlation between histone loss and gene activation has emerged from genome-wide studies in Drosophila, as it has in yeast, suggesting that transcription in higher eukaryotes is also likely to be regulated by histone loss and replacement at the promoter. However, thus far, the contribution of Asf1 to dynamic gene regulation during cell signaling in multicellular organisms has not been examined. One cell-signaling pathway with very direct effects on transcription is the highly conserved Notch pathway. Activation of the receptor results in the release of a nuclear-targeted intracellular fragment (Nicd), which binds directly to the CSL DNA-binding protein (Suppressor of Hairless, Su(H), in Drosophila) and recruits the coactivator Mastermind, resulting in the activation of target genes. CSL proteins also contribute to the silencing of target genes in the absence of Nicd, through adaptor-mediated recruitment of corepressors such as Groucho (Gro), CtBP, and SMRT. Previous analysis indicates that the activity of Notch target genes correlates with a reduction in histone H3 density, suggesting that nucleosome disassembly and reassembly is likely to be involved in their regulation, and prompting an investigatation of whether Asf1 could play a role (Goodfellow, 2007).

    This study shows that Asf1 contributes to the repression of Notch target genes, and that it is recruited to the DNA through interactions with the Su(H)/H complex. Thus, Asf1 is targeted to specific loci by binding to sequence-specific DNA-binding complexes, where it can promote gene silencing during development (Goodfellow, 2007).

    To investigate whether Asf1 contributes to the regulation of inducible genes in Drosophila, RNA interference (RNAi) was used to deplete S2-N cells and the levels of transcription were analyzed from the 11 well-characterized Notch target genes clustered in the E(spl) complex. Conditions were established for activating Notch in these cells and it was shown that activation results in Su(H)-dependent stimulation of E(spl) gene transcription (Goodfellow, 2007).

    Unlike knockdown of the other chromatin regulators tested, depletion of Asf1 led to a 4-fold increase in E(spl)m7 mRNA levels, but it had no effect on the housekeeping genes rp49 and EF2B. More extensive analysis revealed that mRNA levels for all E(spl) genes were increased after Asf1 depletion in the absence of Notch activation; some showed a greater than 10-fold change in expression, suggesting that these Notch targets are derepressed as they are when the corepressor Hairless is depleted. In contrast, there was little effect of Asf1 depletion on several other repressed genes, including a phagocytosis receptor gene, nimrod. In addition to the derepression observed in resting cells, Asf1 depletion also altered the responsiveness to Notch activation. Many more of the E(spl) genes were susceptible to Notch activation in Asf1-depleted cells; for example, 5 of the 11 genes were expressed at greater than 20-fold higher levels after Asf1 RNAi. There was comparatively little change at the genes, such as E(spl)m3, which normally has the most robust response to Notch and is depleted for histones. Thus, it appears that Asf1 makes important contributions to the silencing of Notch target genes (Goodfellow, 2007).

    Previous studies showed that overexpression of Asf1 in the Drosophila eye (ey::Gal4 UAS::asf1/+) causes a 'small-eye' phenotype in which the eye is reduced in size and ommatidia are disorganized. If these small-eye phenotypes are a consequence of Asf1 altering the transcription of Notch targets, they may be modified when combined with mutations in the Notch pathway. To investigate this possibility, flies overexpressing Asf1 were crossed to alleles affecting genes central to Notch or to other signaling pathways, and the eye size was analyzed in the heterozygous progeny (Goodfellow, 2007).

    The first dramatic result was that the heterozygous combination of a Notch loss-of-function allele (N55e11) and Asf1 overexpression caused a severe reduction in the eye/head capsule ('pin-head') and resulted in lethality. Thus, the effects of Asf1 overexpression were strongly enhanced by a decrease in Notch function. Significant enhancement of the Asf1 phenotype also occurred with Delta loss-of-function alleles, but not with alleles affecting Hedgehog (smo), EGF-R (Egfr), or Wingless (arm, arrow) pathways or with alleles affecting the SET domain protein Trithorax-related (trr), the histone exchange factor Domino (dom), or the cell adhesion protein Pawn (pwn). Complementary results were obtained by using mutant alleles that increase Notch signaling: both a loss-of-function Hairless (H) allele and a gain-of-function Notch allele (NMcd1) suppressed the small-eye defect caused by Asf1 overexpression. These findings are fully consistent with the results of RNAi-mediated Asf1 depletion, and they suggest that Asf1 is involved in repression of Notch target genes. As asf1 mutant cells failed to proliferate, it was not possible to obtain clones of homozygous mutant cells to test the effects of eliminating Asf1 on Notch target genes in the eye (Goodfellow, 2007).

    To investigate whether interactions between Notch and Asf1 occur in other tissues, it was asked whether Asf1 overexpression also perturbed Notch function in the Drosophila wing. Expression of Asf1 in the developing wing pouch (sd::Gal4/+; UAS::asf1/+) resulted in margin loss/wing nicks and mild vein thickening, characteristics of reduced Notch function (Notch/+ heterozygous flies have mild wing nicks due to reduced signaling at the dorsal/ventral (d/v) organizer of the wing). The Asf1 overexpression phenotypes were strongly enhanced when the levels of Notch were reduced; thus, wings had extensive scalloping/margin loss and more extensive vein thickening. Wing phenotypes, similar to the eye phenotypes, produced by Asf1 expression were thus enhanced by reduced Notch (Goodfellow, 2007).

    To further assess whether Asf1 affects expression of target genes regulated by Notch (e.g., cut) or by other pathways (e.g., spalt), the effects of overexpressing Asf1 in wing discs was analyzed. In wild-type discs, Notch-dependent expression of Cut is detected in a stripe along the d/v boundary. This was interrupted and reduced in discs in which Asf1 was overexpressed. In contrast, there was no visible effect on Spalt under these conditions. Similar results were obtained when Asf1 was expressed in a more limited domain (by using ptc::Gal4), where a local loss of Cut, but not Spalt, expression was seen. Stronger expression of Asf1 resulted in more pronounced Notch-like phenotypes and loss of Cut expression, which could be rescued by a reduction in Hairless function. Under these conditions, where Asf1 was expressed more strongly, some more generalized effects of Asf1 were sometimes detected, compatible with its proposed role as a histone chaperone during replication. The replication defects became more severe at even higher levels of expression (29°C). Similarly, clones of cells mutant for asf1 failed to proliferate. Thus, as in yeast, Asf1 appears to have roles in replication-dependent as well as replication-independent chromatin dynamics in Drosophila. By moderating the levels of Asf1 expression, it was possible to uncouple these requirements, revealing a contribution to repression of Notch target genes (Goodfellow, 2007).

    Complexes implicated in repression at Notch targets are formed by the CSL/Su(H) DNA-binding protein in conjunction with adaptor proteins, such as SKIP and Hairless, which recruit general corepressors, including SMTR or Gro and CtBP. On polytene chromosomes from Drosophila salivary glands, Asf1 is detected at most Su(H)-enriched sites, suggesting that these proteins are present at the same loci. Asf1 is also bound at many other loci, and it is strongly enriched at centromeres and telomeres, reflecting its multiple roles in chromatin dynamics (Goodfellow, 2007).

    The colocalization of Su(H) and Asf1 on polytene chromosomes prompted a test of whether Su(H) and/or associated factors could copurify with Asf1 in immunoprecipitation (IP) experiments. For these experiments, extracts prepared from Drosophila embryos, and Su(H) or Asf1 was immunoprecipitated by using moderate salt conditions. Under these conditions, Asf1 was detected in Su(H) IP experiments, and, conversely, Su(H) was precipitated with Asf1, as was the corepressor Gro, but not CtBP. To exclude the possibility that the interaction between Asf1 and the Su(H) complex was mediated by the independent binding of both protein complexes to DNA, IP experiments were performed in the presence of ethidium bromide (EtBr), a DNA-intercalating drug that dissociates proteins from DNA. This treatment did not affect the interaction of Asf1 with Su(H). Thus, these data suggest that Asf1 is present in protein complexes containing the sequence-specific DNA-binding protein Su(H) and the Gro corepressor. A significant suppression of the Asf1-induced small-eye phenotype was observed in flies that were also heterozygous for a strong gro allele (groE48) and an enhancement was seen by Hairless proteins that retained a Gro-binding domain, agreeing with a model linking Gro to Asf1-mediated repression. Therefore whether any of the proteins in the Su(H) repression complex are able to bind to bacterially produced Asf1 (fused to glutathione S-transferase, GST), was examined. Of those tested, both Hairless and the adaptor protein SKIP were bound to GST-Asf1, but not to GST alone or to GST-CAF1p55 (a component of chromatin assembly factor 1). Neither Gro nor Su(H) itself showed direct interactions with Asf1 in this assay (Goodfellow, 2007).

    Finally, to test whether Hairless contributes to the recruitment of Asf1 in vivo, chromatin immunoprecipitation (ChIP) was performed with anti-Asf1 antibodies in cells with and without RNAi-mediated depletion of Hairless and association with two E(spl) genes, m3 and m7, was assayed. The E(spl)m7 gene is silenced in the S2 cells and is strongly affected by Asf1 depletion, whereas E(spl)m3 is expressed in S2 cells, is highly induced by Notch activation, and is more mildly affected by Asf1 depletion. Of the two genes, the greatest effects were seen for E(spl)m7; binding of Asf1 to both enhancer and ORF fragments strongly decreased in ChIP after Hairless depletion. A decrease was also seen at the E(spl)m3 ORF region, but not at the E(spl)m3 enhancer. This enhancer is found to have very low histone coverage in these cells, and it was found that it shows only small Asf1 occupancy levels. The decrease in Asf1 from ORFs of both E(spl)m3 and E(spl)m7 after Hairless depletion may indicate that Asf1 spreads from the site of recruitment. Binding of Asf1 to E(spl)m7 and E(spl)m3 regions was confirmed by using affinity-purified anti-Asf1 antibodies raised in a different species. Loss of Hairless does not affect the binding of Asf1 to other loci that do not require Su(H)/H for their regulation, such as eiger or snRNP69D. Similarly, there was no change in the levels of Polycomb protein associated with bxd-PRE after Hairless knockdown. Together, these data support the model that recruitment of Asf1 to Notch targets requires Hairless (Goodfellow, 2007).

    The density and precise positioning of nucleosomes are important factors in determining the transcriptional activity of a locus. It is now evident that most nonnucleosomal histones in cells are likely to be complexed with chaperones. It is therefore not surprising that the histone chaperone Asf1 is important for chromatin dynamics and has been shown to have multiple roles in transcription as well as in the disassembly and reassembly of chromatin during replication. These include gene-specific roles in repression, activation, and transcription elongation. For example, Asf1 is required for nucleosome disassembly and transcription activation at the yeast PHO5, PHO8, ADY2, and ADH2 promoters. However, the mechanisms responsible for targeting Asf1 to these loci remain unclear. This study has demonstrated that Asf1 can be specifically recruited to target loci by interactions with sequence-specific DNA-binding transcription factors. Asf1 is present in a complex with Su(H), the central DNA-binding protein in the Notch pathway, and that it interacts directly with two proteins found in CSL complexes, Hairless and SKIP. Importantly, it was found that Asf1 plays a significant role in the repression of Notch target genes. Thus, contrary to effects at many of the inducible loci examined in yeast, these data demonstrate a requirement for Asf1 in silencing rather than in activation of these inducible genes (Goodfellow, 2007).

    As the global corepressor Gro is also coprecipitated with Asf1 and is implicated in Asf1-mediated repression through genetic interactions, Gro and Asf1 may cooperate in the repression of Notch target genes. Gro has been postulated to exert long-range repressive effects by nucleating a transcriptionally silent chromatin state, in a similar manner to its yeast relative Tup1. For example, at the STE6 locus, Tup1 recruitment results in increased nucleosomal density and local nucleosome positioning. The recruitment of the histone chaperone Asf1 with Gro to Su(H)/H DNA-binding complexes could facilitate a similar localized increase in histone deposition and participate in the spreading of repressed chromatin. Furthermore, since (H)/H complexes engage in comparatively low-stability interactions with target loci, it is suggested that Asf1 could be critical for translating these transient interactions into stable silencing. However, thus far, the analysis has focused on relatively few targets and tissues; thus, it remains to be determined whether Asf1 is recruited to all targets regulated by Su(H)/H, or whether there are additional factors that influence its recruitment at specific loci. Similarly, it will be important to determine whether other sequence-specific complexes are able to bind directly to Asf1 (Goodfellow, 2007).

    In conclusion, these results show that the histone H3/H4 chaperone Asf1 contributes to selective silencing of genes in Drosophila, through interactions with the Su(H)/H DNA-binding protein complexes. In this way, chaperones can act as gene-selective regulators that contribute to the control of gene expression by developmental signaling pathways (Goodfellow, 2007).

    Spreading of a corepressor linked to action of long-range repressor hairy

    Transcriptional repressor proteins play key roles in the control of gene expression in development. For the Drosophila embryo, the following two functional classes of repressors have been described: short-range repressors such as Knirps that locally inhibit the activity of enhancers and long-range repressors such as Hairy that can dominantly inhibit distal elements. Several long-range repressors interact with Groucho, a conserved corepressor that is homologous to mammalian TLE proteins. Groucho interacts with histone deacetylases and histone proteins, suggesting that it may effect repression by means of chromatin modification; however, it is not known how long-range effects are mediated. Using embryo chromatin immunoprecipitation, a Hairy-repressible gene in the embryo was analyzed during activation and repression. When inactivated, repressors, activators, and coactivators cooccupy the promoter, suggesting that repression is not accomplished by the displacement of activators or coactivators. Strikingly, the Groucho corepressor is found to be recruited to the transcribed region of the gene, contacting a region of several kilobases, concomitant with a loss of histone H3 and H4 acetylation. Groucho has been shown to form higher-order complexes in vitro; thus, the observations suggest that long-range effects may be mediated by a 'spreading' mechanism, modifying chromatin over extensive regions to inhibit transcription (Martinez, 2008).

    In support of the spreading model, Groucho binds hypoacetylated histone H3 and H4 tails, and mutations in its N-terminal oligomerization domain block its repression activity in vivo. The recruitment of histone deacetylases may thus enhance Groucho binding to adjacent histones in a positive-feedback loop. A similar mechanism has been suggested for Tup1, the Groucho homolog in yeast. As with Groucho, histone deacetylases have been shown to be crucial for Tup1 repression. Moreover, Tup1 also has an affinity for hypoacetylated amino-terminal histone tails, and mutations or deletions of the tails cause the derepression of Tup1 targets. As with the Hairy repressor here, Tup1 does not change the methylation status of target genes, and the deletion of histone methyltransferases does not affect Tup1-mediated repression. This suggests that methylation marks may not need to be reversed to achieve repression but they may facilitate ready reactivation seen upon the depletion of Hairy. Regarding the extent of association of Tup1 with target genes in yeast, chromatin immunoprecipitation studies have yielded conflicting pictures. Tup1 has been reported to interact with the a-cell-specific STE6 gene only at the promoter or over a distance from 1 to 3.5 kbp, encompassing the entire gene. This discrepancy may be due to differences in cross-linking or immunoprecipitation conditions, reflecting the difficulty in analyzing indirectly bound factors. Indeed, in the current study, promoter interactions by Groucho were observed only with the use of a double-cross-linking protocol (Martinez, 2008).

    Models of transcriptional repression include direct competition, local 'quenching' (displacement or interference with activators), and interactions with the basal machinery. The finding that activators and coactivators are still present under conditions in which the gene is repressed by Hairy suggests that the third model might apply here. The Gal4 activator might represent a particularly stably bound protein, as it does not show the high rate of exchange noted for other transcriptional activators. Thus, it is possible that Hairy-mediated repression does interfere with the binding of some activators on endogenous loci. However, repression can be quite effective even in the absence of activator displacement, perhaps by targeting the basal machinery, similarly to Tup1-mediator interactions seen in yeast. The promoter-proximal location of the repressor in the current system might bias the system to such interactions, but this arrangement is physiologically relevant, since Hairy is found in such proximal locations on endogenous genes. In addition, the LexA-Hairy repressor is also active when bound at -2 kbp, indicating that promoter proximity is not required for activity. Interestingly, a recent chromatin immunoprecipitation survey of enhancers targeted by the Snail short-range repressor suggests that this repressor can be bound to inactive enhancers simultaneously with activators, raising the possibility that short-range repression might also involve direct interactions with the basal machinery (Martinez, 2008).

    The extensive contacts of Groucho over the repressed locus are strongly reminiscent of the extended nucleoprotein structures deposited on regions repressed by stable, heritably acting systems such as Polycomb group (PcG) proteins in animals and Sir proteins in silent-mating-type loci and subtelomeric regions of yeast. There, chromatin regions are modified and inhibited for the formation of productive transcription complexes. Indeed, the association of activators and components of the transcriptional machinery with repressed loci in these systems mirrors the continued binding of activators and coactivators in the system that was studied here, suggesting that the limiting factor for transcription occurs at a later stage. What sort of inhibitory interaction might be involved in this case? A number of recent reports have raised the possibility that repressed, or nonactivated, promoters feature RNA polymerase II that is blocked for elongation, similar to the paused polymerase found at the hsp70 locus under noninducing conditions. It has been found that RNA polymerase II is not displaced from the slp1 gene upon repression with Runt, a Groucho-binding protein (Wang, 2007). It is possible that Groucho itself, through contacts with histone proteins and/or the recruitment of deacetylases such as Rpd3, establishes a chromatin environment that is inhibitory for transcriptional elongation. However, no reliable signals were obtained for RNA polymerase II at this promoter, precluding a definitive statement about polymerase occupancy in activated and repressed states (Martinez, 2008).

    A difference between the repression complexes assembled by the Hairy repression domain and by these other proteins is the transience of the effect; while PcG regulation is linked to epigenetic modifications that allow repression to persist for an extended time when PcG proteins are depleted, the regulation that is seen in this study is readily reversed upon the loss of the LexA-Hairy repressor. Similar effects are observed with elements regulated by the endogenous Hairy protein; enhancers bearing Dorsal and Twist activator sites that are repressed by Hairy in the blastoderm embryo are reactivated minutes later in the germ band extended stage. Thus, Hairy appears to be designed for highly effective but readily reversible repression, which may be useful in particular developmental settings (Martinez, 2008).

    In contrast to a model of linear spreading, an alternative picture of Groucho interaction that is consistent with the current observations is that the corepressor may be tethered to the promoter region, forming larger multimeric complexes around which proximal and distal portions of the gene are wrapped (the 'turban' model). Groucho would indirectly contact the promoter region via Hairy binding and make direct histone contacts with more-distal regions. The selective cross-linking of Groucho to promoter regions treated only with the additional DSP cross-linker is consistent with such a picture. This model may also explain why downstream interactions, albeit weak ones, are often detected of the LexA-Hairy repressor, particularly when employing the more extensive double-cross-linking protocol. In either case, Groucho itself may be important for interfering with activities of transcription factors or the transcription of distant loci. Both of these models suggest that the extensive spread or extensive contacts of Groucho is mechanistically linked to transcriptional repression; however, it is possible that Groucho's extensive contacts with downstream regions are not the main effector of Hairy-mediated repression. Promoter-proximal activities of Groucho, or of other Hairy corepressor proteins, may play the decisive role in dictating long-range effects. Extensive experimental evidence indicates that Groucho plays a key role in the repression mediated by Hairy; therefore, it seems parsimonious to assume that Groucho activity on the repressed gene is important for repression. In support of a 'turban' model of repression, a recent study of the human Groucho homolog Grg3 showed that the recruitment of Grg3 to chromatin induces the formation of a highly condensed structure in vitro (Sekiya, 2007). In addition, in vivo recruitment of Grg3 by FoxA resulted in Grg3 being detected by chromatin immunoprecipitation analysis up to distances of 1 kbp from the FoxA binding site (Martinez, 2008).

    The current study demonstrates that repression by Hairy is associated with histone deacetylation, which is certainly consistent with the nature of cofactors associating with this protein. Interestingly, this modification appears to be restricted to regions close to the repressor binding sites, which places them close to the transcriptional initiation site. How might this be related to the long-range effects mediated by Hairy? One possibility is that Hairy, regardless of where it is bound, induces characteristic changes on chromatin close to the transcriptional start site, that would induce a dominant (and hence long-range) effect on target genes. Alternatively, the local chromatin deacetylation may reflect the reversal of promoter-localized histone acetylases (e.g., SAGA), and acetylation levels on other portions of the gene are already too low to show robust deacetylation. A third possibility is that other Hairy-induced chromatin modifications, not assayed in this study, are more extensive than the deacetylation (Martinez, 2008).

    This study strongly supports a model for Hairy repression that involves contacts between the Groucho corepressor and extended regions of the silenced gene, providing a basis for the long-range repression observed for this protein that is independent of activator displacement. Interesting questions for future studies are how Groucho spreading is limited and whether specific chromatin signals modulate this activity. In addition, such extended repression complexes might be specific to subsets of Hairy targets. A recent study identified a mutation in gro that blocks multimerization but not repression of some genes, suggesting that this cofactor is likely to employ distinct activities at different genes (Jennings, 2008). In addition, genomic surveys indicate that Hairy is likely to associate with distinct cofactors at different loci. Future work will focus on identifying the roles of individual cofactors of this repressor at genes that represent the diversity of Hairy targets in Drosophila (Martinez, 2008).

    Drosophila CK2 phosphorylates Hairy and regulates its activity in vivo
    Hairy is a repressor that regulates bristle patterning, and its loss elicits ectopic bristles (neural hyperplasia). However, it has remained unknown whether Hairy is regulated by phosphorylation. This study describes the interaction of protein kinase CK2 and Hairy. Hairy is robustly phosphorylated by the CK2-holoenzyme (CK2-HoloE) purified from Drosophila embryos, but weakly by the catalytic CK2alpha-subunit alone, suggesting that this interaction requires the regulatory CK2beta-subunit. Consistent with this, Hairy preferentially forms a direct complex with CK2-HoloE. Importantly, genetic interactions were demonstrated between CK2 and hairy (h). Thus, flies trans-heterozygous for alleles of CK2alpha and h display neural hyperplasia akin to homozygous hypomorphic h alleles. In addition, similar phenotypes are elicited in wild-type flies upon expression of RNAi constructs against CK2alpha/beta, and these defects are sensitive to h gene dosage. Together, these studies suggest that CK2 contributes to repression by Hairy (Kahali, 2008. Full text of article).

    Degringolade, a SUMO-targeted ubiquitin ligase, inhibits Hairy/Groucho-mediated repression

    Transcriptional cofactors are essential for proper embryonic development. One such cofactor in Drosophila, Degringolade (Dgrn), encodes a RING finger/E3 ubiquitin ligase. Dgrn and its mammalian ortholog RNF4 are SUMO-targeted ubiquitin ligases (STUbLs; see Model for SUMO-directed ubiquitination by the conserved STUbL family). STUbLs bind to SUMOylated proteins via their SUMO interaction motif (SIM) domains and facilitate substrate ubiquitylation. This study shows that Dgrn is a negative regulator of the repressor Hairy and its corepressor Groucho [Gro/transducin-like enhancer (TLE)] during embryonic segmentation and neurogenesis, as dgrn heterozygosity suppresses Hairy mutant phenotypes and embryonic lethality. Mechanistically Dgrn functions as a molecular selector: it targets Hairy for SUMO-independent ubiquitylation that inhibits the recruitment of its corepressor Gro, without affecting the recruitment of its other cofactors or the stability of Hairy. Concomitantly, Dgrn specifically targets SUMOylated Gro for sequestration and antagonizes Gro functions in vivo. These findings suggest that by targeting SUMOylated Gro, Dgrn serves as a molecular switch that regulates cofactor recruitment and function during development. As Gro/TLE proteins are conserved universal corepressors, this may be a general paradigm used to regulate the Gro/TLE corepressors in other developmental processes (Abed, 2011).

    Transcriptional cofactors are essential for the function of sequence-specific transcription factors and are part of the machinery required to execute temporally coordinated gene expression programs. Regulation of cofactor recruitment and activity is emerging as a major level of gene expression regulation. For example, Hairy/Enhancer of split/Deadpan (HES) family repressors are the primary transducers of the Notch signalling pathway that has a central role in patterning, stem cell development, and is misregulated in cancers. A well-studied case is the Drosophila repressor Hairy, a typical HES family member, which encodes a basic helix-loop-helix (bHLH) Orange repressor required for embryonic segmentation and adult peripheral nervous system (PNS) specification. Hairy-mediated repression is dependent on its ability to recruit cofactors. For example, Hairy recruits the corepressor Groucho (Gro) through it C-terminal WRPW domain, an interaction that is essential for periodic repression of fushi tarazu (ftz). In addition, Hairy recruits dCtBP and dSir2 through its PLSLV and basic domains, respectively. While these cofactors are required for Hairy-mediated repression, they exhibit context-dependent recruitment and function. Interestingly, some cofactors enhance Hairy-mediated repression (e.g., Gro and dSir2), whereas others are required to refine Hairy's function (e.g., dCtBP and dTopors). Consistent with this, it was found that most of the genomic loci bound by Hairy in the context of Kc cells exhibit corecruitment of dSir2 and dCtBP, but are not co-bound by Gro. However, the mechanisms that regulate context-selective cofactor association with Hairy or that may regulate cofactor activities are largely unknown (Abed, 2011).

    A possible mechanism is that post-translational modification of Hairy regulates its association with a given cofactor and determines its overall function. One such modification is ubiquitylation that in many cases regulates the stability of transcription factors. However, ubiquitylation can also serve as a regulatory modification that does not lead to degradation, but affects protein-protein interaction or intracellular localization. Similarly, SUMOylation is a post-transcriptional modification that is involved in the regulation of gene expression and is mediated by the SUMO-specific E1-, E2-, and E3-SUMO ligase enzymes. Both ubiquitin and SUMO modifications are highly regulated. These two modifications can also be connected through proteins collectively termed SUMO-targeted ubiquitin ligases (STUbLs). STUbLs are RING proteins that bind non-covalently to the SUMO moiety of SUMOylated proteins via their N-terminal SUMO interaction motif (SIM) domains, and subsequently target the SUMOylated protein for ubiquitylation via their RING domain. Thus, STUbLs are able to 'sense' SUMOylated targets and modify them by ubiquitylation. The observation that STUbLs are associated with transcription complexes suggests that their function is directly linked to regulation of gene expression. For example, the STUbL protein RNF4 was found to be a positive regulator of steroid hormone transcription. Importantly, STUbLs are structurally and functionally conserved, as the mouse and human RNF4 proteins can substitute for their yeast orthologs in functional assays. STUbLs are required for the correct assembly of kinetochores, for the cell's ability to cope with genotoxic stress, and for genome stability. RNF4 is highly expressed in the stem cell compartment of the developing gonads and brain, and its expression is enriched in progenitor cells, likely representing its role in 'stemness'. Recently, RNF4 was shown to regulate the SUMO- and ubiquitin-mediated degradation of PML and PML-RAR. However, the role of STUbL proteins in transcription during development of higher eukaryotes is largely unknown (Abed, 2011).

    This study shows that Degringolade (Dgrn), the only Drosophila STUbL protein, physically and genetically interacts with Hairy and its cofactor Gro and antagonizes Hairy/Gro-mediated repression during segmentation and neurogenesis. Ubiquitylation of Hairy by Dgrn affects choice of cofactor by preventing Gro, but not dCtBP, from binding to Hairy. It was also found that Dgrn specifically targets SUMOylated Gro, alleviates Gro-dependent transcriptional repression, and suppresses Gro functions in vivo throughout development. DamID chromatin profiling experiments revealed that the antagonism between Dgrn and Gro is aimed at a broad array of genomic loci, suggesting that Gro-Dgrn antagonism is of general importance beyond Dgrn's interaction with Hairy (Abed, 2011).

    Dgrn binds directly to Hairy and is capable of ubiquitylating Hairy in a reconstituted system and in cells. The recognition motif for Dgrn within Hairy maps to Hairy's basic region and requires a specific positive charge (Arg33). This motif is transferable and functionally conserved, not only in Hey and other HES proteins (e.g., E(spl)m8 and Dpn), but also in dMyc and other bHLH proteins including the activator Sc. Therefore, it may reflect a general property of bHLH recognition by STUbL proteins. No evidence was found for direct SUMOylation of the HES and bHLH proteins: bacterially purified Hairy and Dgrn proteins interact, anti-SUMO antibodies fail to detect SUMOylated Hairy, Hairy's mobility in SDS-PAGE is not altered upon incubation with the dUlp1 SUMO peptidase, and mutating putative SUMOylation sites within Hairy does not alter its recognition or ubiquitylation by Dgrn. Accordingly, this study found that Dgrn's interaction with Hairy is mediated through Dgrn's RING motif independent of the SIM domains. Similarly, the yeast STUbL Slx5-Slx8 recognizes the MATα2 repressor independent of SUMOylation (Xie, 2010). Hairy recognition by Dgrn/RNF4 is also different from its recognition of substrates, such as GST-SUMO or PML, that involves direct SUMOylation of the targeted protein and requires the Dgrn/RNF4 SIM domains (Abed, 2011 and references therein).

    Importantly, SUMOylation and the SIM motifs are necessary for Dgrn to target SUMOylated Gro and for Dgrn's suppression of HES/Gro repression in vivo, it is likely that the SIM domains interact with the poly-SUMO chain itself (Geoffroy, 2010). Dgrn possessing two separate recognition modules is reminiscent of the dual recognition properties described for the RING protein UBR1 (E3alpha). As the current dogma is that STUBLs recognize (via their SIM domains) poly SUMO chain(s) rather than the substrate, the dual recognition mechanism observed with Dgrn may further substantiate substrate recognition and specificity (Abed, 2011).

    The contribution of each SIM domain is additive, and a Dgrn mutant harbouring a single SIM domain is capable of binding to GST-SUMO, as well as conjugating Hairy, although to a lesser extent than wild-type Dgrn. Correspondingly, it was found that elevated levels of SUMOylated proteins are detected in dgrn null embryos (Barry, 2011; Abed, 2011).

    As an ubiquitin ligase, Dgrn catalyses the formation of mixed poly-ubiquitin chains on Hairy. This ubiquitylation does not map to Hairy's basic region, its putative SUMOylation sites, or to a single Lys residue. Importantly, this poly-site ubiquitylation does not affect Hairy protein stability or integrity, but rather selectively inhibits Gro binding to Hairy. Furthermore, in cells in which Dgrn protein levels are reduced via RNAi, Hairy protein levels are also decreased compared with control cells, suggesting that Dgrn is likely required for Hairy expression. This is different from dTopors, a Hairy-associated PHD-RING finger protein, which catalyses Lys48-linked chains and regulates Hairy turnover. Further work will be required to determine the exact molecular events and the role that specific ubiquitin chain linkage has in Dgrn's ability to inhibit Gro from binding to Hairy in vivo (Abed, 2011).

    Despite extensive efforts, ubiquitylated Gro forms were not identifed in this study. Nonetheless, the data suggest that Dgrn specifically targets the SUMO chains on Gro, which likely serve as a signal for Gro sequestration by as yet to be identified machinery (Abed, 2011).

    In transcription assays, Dgrn is a potent activator of ac and Sxl transcription, a function that requires its catalytic activity. Dgrn antagonizes Hairy-, Dpn-, and Gro-mediated repression in vivo. Dgrn specifically targets SUMOylated Gro, Dgrn function inversely correlates with SUMOylation, and a reduction in SUMO levels impairs Dgrn's ability to fully alleviate repression. Thus, Dgrn's activity suppresses the local repressive chromatin structure generated by repressors, their associated cofactors, and the SUMO pathway. It was also found that expression of DgrnHC/AA can inhibit the activation mediated by Da/Sc, suggesting that Dgrn is required to alleviate repression by endogenous repressors and/or corepressors. This fits well with the observation that reduction in Dgrn protein levels via RNAi impairs Da/Sc-mediated activation. While this study focused on Dgrn's effects on the repressive machinery, it is also possible that part of Dgrn ligase activity enhances the function of activators and/or coactivators. For example, Dgrn efficiently ubiquitylates the pro-neural activator Sc, and significant activation of the ac or Sxl promoters requires only Dgrn along with either Da or Sc (Abed, 2011).

    These data suggest that part of Dgrn's activity is aimed specifically at the Gro corepressor that is shared by all HES proteins. First, Dgrn-mediated ubiquitylation of Hairy prevents Gro recruitment to Hairy. Second, Dgrn specifically targets SUMOylated Gro and its associated Gro oligomers for sequestration. Specifically, it was found that the detected level of Gro protein is dependent on Dgrn and the method of protein extraction. For example, in embryos that lack Dgrn (dgrnDK) and when protein extracts are made in RIPA buffer, the detected levels of Dgrn in dgrnDK embryos is higher compared with that of wild type. However, if the extraction is performed in 4% SDS buffer, the detected levels of Gro protein in wild-type and dgrnDK embryo extracts is equal. Likewise, the signal detected for Gro using immunostaining in embryos is highly complementary to the milder RIPA extraction. dgrnDK embryos show an increased signal compared with wild-type embryos (as in the absence of Dgrn, less Gro is sequestered and more Gro molecules are available for detection by the antibody). The majority of Gro appears to be sequestered. Since only 90% of Gro can be recovered after co-transfection of Dgrn using SDS extraction, the possibility cannot be ruled out that a fraction of the SUMOylated Gro is degraded. All together, these data suggest that Dgrn is required for Gro sequestration and that loss of Dgrn 'liberates' sequestered Gro (Abed, 2011).

    While the data support a model in which Dgrn targets SUMOylated Gro for sequestration, Dgrn may also regulate the molecular machinery that is required for Gro SUMOylation and subsequently sequestration. Furthermore, while it is established that STUbL targets SUMOylated proteins for ubiquitylation and degradation, it is also possible that Dgrn has an impact on the SUMO pathway and SUMO isopeptidases (Abed, 2011).

    Gro and its mammalian orthologs, the transducin-like enhancers of split (TLE1-4) proteins, repress transcription via several mechanisms, including oligomerization to generate local repressive chromatin structures, and are negatively regulated by phosphorylation. This study found that site-specific phosphorylation used by RTK signalling to inactivate Gro is not a prerequisite for Dgrn activity. However, the details surrounding other phosphorylations, the role of site-specific SUMOylation of Gro, and the molecular machinery mediating sequestration, as well as Dgrn's effects on specific Gro-dependent repressive mechanisms await further studies (Abed, 2011).

    In vivo, it was found that Dgrn antagonism of Gro is highly relevant for embryonic segmentation, PNS development, and sex determination, processes that are regulated by Gro (Barry, 2011). Indeed, Dgrn can suppress the gain-of-function phenotypes of Gro, as well as rescue the phenotypes associated with tissue-specific inactivation of Gro using RNAi transgenes. The genomic targets of Gro and Dgrn are distinct from that of dCtBP or dSir2, and that 38% of Gro direct targets are shared with Dgrn. Thus, it is predicted that Dgrn will be involved in other HES-independent, but Gro-regulated, processes as well. It is likely that both proteins have unique regulatory roles during early development. This notion stems from observations that each of the factors has exclusive, non-overlapping, genomic binding sites, and that neither of the two genes can functionally rescue the embryonic lethality associated with mutants of the other protein (i.e., Gro cannot rescue the female sterility associated with dgrn null females, and reducing the dose of Dgrn does not rescue the lethality associated with the groE48 mutant) (Abed, 2011).

    Finally, an open question is how can the activity of a general corepressor be temporally and spatially regulated during development. The data to date suggest a model in which Dgrn has a regulatory role. Since it is suggested that SUMOylation enhances Gro-mediated repression, one can imagine that ATP-dependent SUMOylation of Gro within the repressor complex will result in local augmented repression. However, concomitantly, SUMOylation will promote Dgrn recruitment, and subsequent inactivation of the repression complex on chromatin or in its vicinity, ensuring that local SUMO-augmented repression is limited in time and space. It is speculated that this type of transcriptional regulation will be instrumental to define and sharpen patterning borders throughout development (Abed, 2011).

    The Drosophila STUbL protein Degringolade limits HES functions during embryogenesis

    Degringolade encodes a Drosophila SUMO-targeted ubiquitin ligase (STUbL) protein similar to that of mammalian RNF4. Dgrn facilitates the ubiquitylation of the HES protein Hairy, which disrupts the repressive activity of Hairy by inhibiting the recruitment of its cofactor Groucho. This study shows that Hey and all HES family members, except Her, interact with Dgrn and are substrates for its E3 ubiquitin ligase activity. Dgrn displays dynamic subcellular localization, accumulates in the nucleus at times when HES family members are active and limits Hey and HES family activity during sex determination, segmentation and neurogenesis. Dgrn interacts with the Notch signaling pathway by antagonizing the activity of E(spl)-C proteins. dgrn null mutants are female sterile, producing embryos that arrest development after two or three nuclear divisions. These mutant embryos exhibit fragmented or decondensed nuclei and accumulate higher levels of SUMO-conjugated proteins, suggesting a role for Dgrn in genome stability (Barry, 2011).

    A common theme among DNA-bound transcriptional regulators is the recruitment of co-activators and/or co-repressors to carry out their function. An important aspect of all HES family regulation is their recruitment of the co-repressor Groucho. Abed (2011) has shown that the ubiquitylation of Hairy does not lead to its degradation, but rather interferes with the ability of Hairy to recruit Gro, thereby antagonizing Hairy's repressive activity. This study found that dgrn mutant embryos show defects in segmentation. It is suggested that, similar to Dgrn's interaction with Hairy in segmentation, the ubiquitylation of other HES family members or Hey leads to their inability to recruit the cofactor Gro and thus antagonizes the repressor activity of this protein family. Consistent with this, it was found that loss of Dgrn function affects known Hey and HES family early functions, including sex determination and nervous system development (Barry, 2011).

    It was surprising that Dgrn is female sterile rather than exhibiting zygotic lethality. Another, as yet unidentified, STUbL protein might function redundantly to Dgrn postzygotically. As the early Drosophila embryo develops essentially as a closed system running on maternally provided mRNA and proteins, the early syncytial embryo relies heavily on translational and post-translational modifications to control protein activity. Both Dgrn and Gro are maternally contributed and ubiquitously distributed. Thus, Dgrn might be recruited to the nucleus at different times during these early developmental stages to attenuate Gro's ability to be a potent co-repressor in the absence of active transcription, thereby modulating Hey and HES family activity (Barry, 2011).

    Dgrn's human homolog is the transcriptional cofactor and STUbL protein RNF4. Indeed, human RNF4 acts as a functional homolog of Dgrn. RNF4 has also been shown to be a functional ortholog of the Rfp1/Rfp2-Slx8 heterodimer (from now on referred to as Rfp-Slx8) in S. pombe and the Slx5-Slx8 heterodimer in S. cerevisiae. RNF4 and the yeast Rfp-Slx8 and Slx5-Slx8 heterodimers have been shown to be important for DNA repair, kinetochore assembly and genome stability, with the loss of these proteins leading to fragmented chromosomes, elongated nuclei, asymmetric positioning of the nuclei and an accumulation of SUMOylated proteins (Barry, 2011 and references therein).

    The budding yeast Slx5-Sx8 proteins were identified as a complex of proteins required for the viability of SGS1 (a gene encoding the only RecQ helicase involved in genomic integrity in S. cerevisiae) mutant cells. In Drosophila, loss of RecQ5 function leads to the loss of synchronous divisions in the syncytial embryo, an increased number of double strand breaks and a slight increase in the number of abnormal nuclei falling from the surface of the embryo. Mutations of the RecQ family member DmBlm (mus309 -- FlyBase; the Drosophila ortholog of human BLM, which leads to the human disorder Bloom Syndrome when mutated) are female sterile with severe defects in embryogenesis: syncytial embryos frequently include anaphase bridges, gaps in the normally uniform monolayer of nuclei and asynchronous mitoses (Barry, 2011 and references therein).

    Recently, smt3 (SUMO) mutant embryos were shown to display embryonic nuclear cycle defects, including irregular size and distribution of nuclei, chromosome clustering, chromosome bridges, fragmentation and reduced number of nuclei in relation to the centrosome pairs. Several cell cycle factors were identified that are substrates for SUMOylation, it was and proposed that SUMOylation of these factors is important for controlling the cell cycle. The fragmented and de-condensed nuclei observed in the early arrest phenotypes of dgrnDK null embryos are reminiscent of the Slx8-Slx5 mutant phenotypes, fly RecQ mutant phenotypes and smt3 mutant phenotypes, suggesting that the role of STUbL proteins in genome stability and DNA repair might be a conserved function (Barry, 2011).

    Alternatively, mutations in actin cytoskeleton and cell cycle checkpoint components in Drosophila have also been shown to exhibit nuclear arrest phenotypes. Defects in cell cycle checkpoint proteins, including Pan Gu, Plutonium and Giant Nuclei affect the S-phase checkpoint in the early embryo such that mutation of any of these genes leads to unregulated S-phase, resulting in giant polyploid nuclei. Disruption of the actin cytoskeleton can also lead to nuclear division abnormalities of cortical nuclei. For example, the scrambled and nuclear fallout mutants exhibit severe abnormalities in the appearance and localization of cortical nuclei. Further experiments will be needed to determine the molecular mechanism(s) underlying Dgrn's early arrest phenotype. However, regardless of the mechanism, this represents a new function for Hey or HES family proteins or a function for Dgrn that is not HES-dependent (Barry, 2011).

    Interestingly, nuclear cycles during which Dgrn accumulates in the nucleus correspond to times when HES family members are active, which would be necessary for Dgrn to interact physically with HES proteins and subsequently affect their functions. One exception to this is Dgrn nuclear localization at nuclear cycle 9. There are no known HES family activities at nuclear cycle 9; however, several HES family members are yet to be characterized molecularly and genetically. Dgrn also exhibits a novel accumulation pattern during the gastrulation stages where it prefigures morphogenetic furrows, suggesting a possible role for Hey or HES family members in morphogenesis. Chromatin profiling experiments identifying direct transcriptional targets of Hairy identified a number of targets important for morphogenesis, suggesting that Hairy might play a role in morphogenesis. Consistent with this, a new hairy allele (h674) has been reported to affect the early stages of salivary gland morphogenesis. Thus, although Dgrn might work with Hey, Hairy and/or other HES family members during these times, it also remains possible that these Dgrn activities are Hey- and HES-independent (Barry, 2011).

    Dpn is a negative regulator of Sxl. dpn mutants have a modest effect on Sxl in males leading to ectopic expression from the Sxl-Pe promoter that is sufficient to induce the inappropriate female fate in some cells. The Hey and HES family co-repressor Gro has also been shown to act as a negative regulator of Sxl; the loss of maternal Gro results in severe misexpression of Sxl in males leading to female fate. The relatively mild effect of Dpn on Sxl regulation compared with Gro led to a search for additional HES family proteins involved in Sxl regulation. Hey was identified as a maternal repressor of Sxl-Pe, albeit in a spatially variable pattern in males. Unlike the mammalian homologs of Hey, which are unable to bind Gro presumably owing to its C-terminal YRPW domain, this study found that Drosophila Hey binds Gro in GST pulldown assays. The data suggests that Dgrn is an important player in sex determination where it interferes with the repressive activities of Dpn and Gro; both Dpn and Hey are substrates for Dgrn's E3 ubiquitin ligase activity. In addition, Sxl protein staining and in vitro transcription assays demonstrate that Dgrn antagonizes the repression of Sxl. As proposed for the interaction of Dgrn with Hairy during segmentation (Abed, 2011), this study found that Dgrn provides a new level of control over the activity of Hey and the HES family members in sex determination. This control is mediated by ubiquitylation that probably disrupts the ability of these repressors to recruit Gro, thereby antagonizing their ability to repress transcription of Sxl-Pe in males (Barry, 2011).

    It has been proposed that sex in Drosophila is not determined by the ratio of X-chromosomes to sets of autosomes (X:A ratio), but rather by X chromosome dose. It was speculated that a feedback mechanism in females is caused by the acetylation of chromatin, which inhibits Gro-mediated repression. Interestingly, the finding that Dgrn antagonizes Gro activity via the ubiquitylation of Hey and HES family repressors and targets SUMOylated Gro for sequestration provides an alternate scenario for this feedback mechanism in females (Barry, 2011).

    Notch (N), through the E(spl) proteins (its downstream targets), heads one of the major developmental signaling pathways that functions in progenitor cell fate determination and differentiation. Recently, Sxl has been shown to inhibit Notch RNA translation and to negatively regulate the Notch signaling pathway in females. The notched wing phenotype of N was shown to be sensitive to Sxl, such that reducing the dose of Sxl suppressed the lethal effects of N hypomorphic alleles. This study found that reducing the dose of dgrn can also partially rescue the lethal effects of N hypomorphic alleles suggesting that Dgrn antagonizes N signaling. More specifically, it is hypothesized that the rescue of N1 male lethality is due to a decrease in Sxl expression. Dgrn heterozygosity also suppresses the vein patterning phenotype associated with NAX1682, suggesting that it is required for N signaling in this context also. Interestingly, Dgrn could be antagonizing N by two distinct mechanisms (or a combination of the two): the first an indirect antagonization of N signaling through Dgrn's control of Sxl expression, and the second by direct inhibition of the repressor activities of the E(spl)-C protein by ubiquitylation, thus blocking the repressive arm of the N pathway. The second mechanism has implications in regulating crosstalk between N and EGFR signaling pathways. Further studies will be required to determine the role of Dgrn's STUbL activity and whether Dgrn's activity on E(spl)-C proteins is redundant to EGFR signaling or whether both of these activities are required to antagonize Notch signaling (Barry, 2011).

  • hairy: Biological Overview | Evolutionary Homologs | Regulation | 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.