Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Open Access

Peer-reviewed

Research Article

Molecular Analysis of the Notch Repressor-Complex in Drosophila: Characterization of Potential Hairless Binding Sites on Suppressor of Hairless

Abstract

The Notch signalling pathway mediates cell-cell communication in a wide variety of organisms. The major components, as well as the basic mechanisms of Notch signal transduction, are remarkably well conserved amongst vertebrates and invertebrates. Notch signalling results in transcriptional activation of Notch target genes, which is mediated by an activator complex composed of the DNA binding protein CSL, the intracellular domain of the Notch receptor, and the transcriptional coactivator Mastermind. In the absence of active signalling, CSL represses transcription from Notch target genes by the recruitment of corepressors. The Notch activator complex is extremely well conserved and has been studied in great detail. However, Notch repressor complexes are far less understood. In Drosophila melanogaster, the CSL protein is termed Suppressor of Hairless [Su(H)]. Su(H) functions as a transcriptional repressor by binding Hairless, the major antagonist of Notch signalling in Drosophila, which in turn recruits two general corepressors – Groucho and C-terminal binding protein CtBP. Recently, we determined that the C-terminal domain (CTD) of Su(H) binds Hairless and identified a single site in Hairless, which is essential for contacting Su(H). Here we present additional biochemical and in vivo studies aimed at mapping the residues in Su(H) that contact Hairless. Focusing on surface exposed residues in the CTD, we identified two sites that affect Hairless binding in biochemical assays. Mutation of these sites neither affects binding to DNA nor to Notch. Subsequently, these Su(H) mutants were found to function normally in cellular and in vivo assays using transgenic flies. However, these experiments rely on Su(H) overexpression, which does not allow for detection of quantitative or subtle differences in activity. We discuss the implications of our results.

Citation: Kurth P, Preiss A, Kovall RA, Maier D (2011) Molecular Analysis of the Notch Repressor-Complex in Drosophila: Characterization of Potential Hairless Binding Sites on Suppressor of Hairless. PLoS ONE 6(11): e27986. https://doi.org/10.1371/journal.pone.0027986

Editor: Eliana Saul Furquim Werneck Abdelhay, Instituto Nacional de Câncer, Brazil

Received: September 23, 2011; Accepted: October 28, 2011; Published: November 18, 2011

Copyright: © 2011 Kurth et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Funding: This work was supported in part by a National Institutes of Health (NIH) grant CA120199 (http://www.nih.gov) and a Leukemia and Lymphoma Society award (http://www.leukemia-lymphoma.org) to RAK. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

The Notch signalling pathway is highly conserved in metazoans, where it allows for intercellular communication during the specification of cell fates [1]. Notch encodes a single pass transmembrane receptor that is activated by transmembrane ligands presented by the signalling cell. As consequence of receptor activation, the intracellular Notch domain (ICN) is cleaved and migrates to the nucleus. There it binds to the CSL-type DNA-binding protein (C-promoter binding factor 1 [CBF-1] in H. sapiens, [lag-1] in C. elegans, Suppressor of Hairless in D. melanogaster [Su(H)]), and assembles, together with the coactivator Mastermind (Mam), a transcriptional activator complex (overview in: [1]–[4]. Formation of the CSL-ICN-Mam ternary complex, in conjunction with other transcriptional components, results in the activation of Notch target genes, e.g. the Hairy and Enhancer of split (HES) family of genes. HES genes encode transcriptional repressors that function to shut down gene expression for genes that confer the primary cell fate, thereby enforcing a secondary fate within the signal-receiving cell [1]–[2].

The components of the activator complex (CSL-ICN-Mam) are highly conserved from worms and flies to humans in both primary sequence and the overall three-dimensional structure of this complex [5]–[6]. The central molecule of the activator complex is CSL, which contains three functional domains: the N-terminal domain (NTD), beta–trefoil domain (BTD), and C-terminal domain (CTD). Both the NTD and BTD contact DNA. The BTD and the CTD interact with ICN, whereby BTD forms a high-affinity interaction with the RAM domain of ICN and the CTD binds both the ankyrin repeats (ANK) of ICN and Mam [5]–[6], overview in [3].

In the absence of signal, CSL interacts with transcriptional corepressors to turn off transcription from Notch target genes. Similar to the activator complex, CSL is the central component of the repressor complex; however, in contrast to the activator complex, the structure of the repressor complex is still unknown. Human CBF-1 has been shown to interact with several different corepressors, e.g. SMRT/NCOR, MINT/SHARP, KyoT2, and CIR. Most of these corepressors contact a site within the BTD of CBF-1 that likely overlaps where the RAM domain of Notch binds. This has led to a model, in which the repression and activation of Notch target genes is mediated by the competition of ICN and corepressors for binding CBF-1 (overview in [7]). In D. melanogaster, the transcriptional corepressor Hairless is the major antagonist of Notch signalling (reviewed in [8]). Hairless binds to the CTD of Su(H) – the fly CSL protein - and recruits two additional corepressors, the C-terminal binding protein (CtBP) and Groucho (Gro). Together this repressor complex silences expression from Notch target genes ([9]–[12]). Hence, CSL plays a dual role in both activator and repressor complexes.

We have initiated a detailed analysis of the Notch repressor complex in Drosophila. Recently, we have shown that Hairless and Su(H) form a high affinity complex, and that mutations within Su(H) that affect binding of ICN have no effect on Hairless binding. Nonetheless, Hairless and Notch compete for Su(H) in vitro, despite the disparities in affinities of ICN and Hairless for the CTD of Su(H). Moreover, we have mapped a single residue in Hairless that is crucial for binding Su(H) ([10]). To further our understanding of Notch signalling and the repressor complex in Drosophila, we have analysed 17 single, double, and triple mutations in the CTD of Su(H) for their involvement in the binding of Hairless using a yeast two-hybrid assay. A double mutation was identified that strongly reduces interactions with Hairless, but neither affected DNA nor Notch binding by Su(H). In spite of this reduction in binding, overexpression of the Su(H) double mutant in a transcriptional cell culture assay, as well as in the fly, revealed little to no changes in function compared to wild type Su(H). These results were unexpected and we consider two possibilities: (1) potentially other residues in Su(H) contribute to the binding of Hairless, which allows for a sufficiently strong interaction in vivo; or (2) alternatively, the presence of endogenous Su(H) in our cellular and in vivo assays distorts our results.

Results

Identification of potential Hairless binding sites in the CTD of Su(H)

Recently, we have identified the C-terminal domain of Su(H) as the binding domain for Hairless (CTD, amino acids 417–528). Binding to Hairless was enhanced by the presence of the N-terminal α-helix (amino acids 1–119), which helps to stabilize the folding of the CTD. Mutations that affect binding to ICN did not interfere with the binding to Hairless, suggesting that ICN and Hairless do not compete for the same contact sites in Su(H) CTD [10].

To identify the amino acids in CTD responsible for interaction with Hairless, a total of 17 single, double or triple amino acid substitutions were introduced by in vitro mutagenesis. The main criterion for the changes was (1) whether the amino acids were surface exposed, which was based on the orthologous mammalian and C. elegans CSL structures; and/or (2) within a putative protein-protein interaction domain that was determined computationally (http://sppider.cchmc.org) (Fig. 1A,B). The sites of mutation were changed to residues that would likely interfere with Hairless binding (Fig. 1B,C). The mutant constructs were tested in a yeast two-hybrid assay using Hairless or ICN I as bait (Fig. 1 C). In addition, we assayed for the formation of the ternary activator complex consisting of Su(H), Notch Ank and MamN (Fig. 1C; [10]). For the majority of mutants examined no changes in binding were detected. However, four mutations showed reductions in Hairless binding: CTDLEWA (L490E/W491A), CTDWA (W491A), CTDWARE (W491A/R493E) and CTDWVR (W491A/V492R/R493E). Together, this assay revealed that the residues Tryptophan 491 and Arginine 493 are likely important for the binding of Hairless, because the combined mutation CTDWARE nearly abolished Hairless binding in the yeast assay. In addition, Leucine 490 appeared to contribute since CTDLEWA bound less well than CTDWA, whereas mutation of Valine 492 did not further reduce binding in CTDWVR. The single mutations CTDRE (R493E) and CTDVR (V492R) were without effect (Fig. 1C).

thumbnail
Download:
Figure 1. Fine mapping of the Hairless contact sites on Su(H) CTD.

A) Surface representation of CSL-DNA structure with the NTD, BTD, and CTD coloured in dark and light grey, respectively. The DNA is in a stick representation with carbon, oxygen, nitrogen, and phosphorous atoms coloured yellow, red, blue, and orange, respectively. Notch ANK is coloured in blue and MamN in red, and represented as transparent ribbons. The residues on the CTD that interacted with ICN and Mam in the yeast two-hybrid assay are coloured green. Residues that were mutated in the course of this work are coloured magenta. B) Primary sequence of the Su(H)-CTD construct; the CTD is shown in bold. Amino acids shown to contact ANK/Mam are depicted in green; red are those tested for Hairless binding, and brown depicts the AR2 mutation that disrupts the CTD fold. C) Mutant CTD constructs were tested in a yeast two-hybrid assay for binding to full length Hairless (HFL) and to intracellular Notch (ICN I). Moreover, the mutant CTD constructs were tested in a yeast three-hybrid assay for their potential to assemble the ternary activator complex with ANK and MamN. Empty vectors served as negative controls. Relative position of mutations within the CTD is indicated. Note that binding of Hairless but not of ICN I to LEWA or WA is reduced and is nearly abolished in WARE and WVR. However, all these mutants display normal binding to Notch and are capable of forming a ternary activator complex. The constructs GD, GH, VEAD and CR bear mutations at the CTD-ANK and CTD-Mam interfaces, consistent with a strongly reduced binding to ICN I. Hence, Notch and Hairless contact different sites on Su(H). Mutant AR2, in which residues within the hydrophobic core of CTD are mutated, is likely compromised for folding, and fails to bind either HFL or ICN I.

https://doi.org/10.1371/journal.pone.0027986.g001

Two mutations CTDVEAD (V442E/A443D) and CTDCR (C471R) nearly abolished binding to ICN I, and two other mutations CTDGD (G438D) and CTDGH (G439H CTDCR) reduced binding to ICN I; however, none of these mutations affected binding to Hairless (Fig. 1C). CTDCR, CTDGD, and CTDGH lie within the region known to contact the Notch Ankyrin repeats, whereas CTDVEAD is in the vicinity of ANK and MAM, but does not make direct contact with these proteins. Two controls were included, the empty vector and the double mutant CTDAR2 (W459A/F460A). The CTDAR2 mutant affects amino acids buried within the hydrophobic core of CTD and is hence predicted to disrupt CTD folding. As expected, both controls did not bind to either Hairless or ICN I, and failed to assemble the ternary complex with N-Ank and MamN (Fig. 1C).

Based on the structures of mammalian and worm CSL proteins, Su(H) is expected to bind DNA with its N-terminal and beta-trefoil domains (NTD, BTD; Fig. 1A). Accordingly, mutations in CTD should not interfere with DNA binding, which was confirmed by an electrophoretic mobility shift assay (EMSA) with the relevant Su(H) mutants (Fig. 2). To this end, R486E, W491A, L490E/W491A and W491A/R493E mutations were introduced into full length Su(H) cDNA that was in vitro transcribed and translated. The E(spl) m8 oligo-nucleotide containing a Su(H) binding site (m8-S1, [44]) was used as the target DNA. No difference in DNA-binding was observed between wild type and mutant Su(H) protein (Fig. 2).

thumbnail
Download:
Figure 2. DNA binding is not altered in Su(H) mutants.

Electromobility shift assay for binding of Su(H) protein variants to the radiolabelled E(spl)m8-S1 oligo [44]. Control, no protein added (lane 1). The binding of the mutant Su(H)WA, Su(H)RE, Su(H)LEWA, or Su(H)WARE proteins (lane 3–6) to DNA was similar as the wild-type Su(H) protein (lane 2).

https://doi.org/10.1371/journal.pone.0027986.g002

The CTDWARE double mutation fails to bind truncated forms of Hairless

Based on the near complete loss of binding to full length Hairless, the double mutation CTDWARE (W491A/R493E) was chosen for further analysis. Previously, we defined a subdomain of Hairless, termed NTCT (amino acids 171–375; [10]), which recapitulated all of the binding of Hairless to Su(H) in vitro. Unexpectedly, we found that CTDWARE bound to NTCT similarly to full length Su(H) and only slightly weaker than wild type CTD (Fig. 3). In addition, we tested the ability of two Hairless NTCT mutants, the NT-deletion NTCTΔNT and the single site mutant NTCTLD (L235D) for binding to the Su(H) constructs. Both NTCT mutants fail to bind to full length Su(H) and showed a markedly reduced binding activity towards CTD (Fig. 3; [10]). However, NTCTΔNT and notably NTCTLD completely failed to bind to CTDWARE, strongly indicating that the affected amino acids are involved in the binding of Su(H) and Hairless (Fig. 3).

thumbnail
Download:
Figure 3. Su(H)WARE binding capacity to Hairless mutants.

Yeast two-hybrid assay to test for binding activity of wild type Su(H), CTD and mutant CTDWARE constructs with Hairless NTCT, NTCTΔNT Dand NTCTLD; empty vector served as a negative control. Note the lack of binding of CTDWARE with the ΔNT deletion or the NTCTLD mutation.

https://doi.org/10.1371/journal.pone.0027986.g003

Su(H)WARE gives a normal response in a transcriptional assay

Thus far, our data indicated that the W491A/R493E amino acid substitutions had an effect on the Su(H)-Hairless interaction. In order to test the effect of this mutation in a more physiological setting, it was introduced into the Su(H) full length cDNA [Su(H)WARE] and transiently expressed in S2 cells. Subsequently, we analysed the transcriptional activation and repression of a luciferase reporter construct bearing Su(H) binding sites (NRE-reporter; [13]). We reasoned that Su(H)WARE should behave as a transcriptional activator together with ICN, similar to wild type Su(H), because binding of Notch was unaffected by the mutation. This was indeed observed – transfection of ICN alone strongly activates the NRE-reporter, via endogenous Su(H), which was taken as 100% to normalize the other results [13]. Addition of the wild type Su(H) construct resulted in about three- to four-fold increase of luciferase activity (Fig. 4), which is in agreement with earlier observations [10], [14]. A likewise increase in reporter activity was obtained by adding Su(H)WARE, indicating that the mutant protein can efficiently assemble an activator complex in S2 cells (Fig. 4).

thumbnail
Download:
Figure 4. Activation and repression of a Notch reporter gene by Su(H) variants.

Effects of the mutant Su(H)WARE (WARE) on Notch ICN mediated expression from the NRE-reporter (luciferase reporter containing wild type Su(H) binding sites, [13]) were analysed in Drosophila S2 cell culture. Indicated constructs, Notch ICN, full length Hairless (H), full length Suppressor of Hairless (SuH) or the mutant Su(H)WARE (WARE) were used to transiently transfect S2 cells; empty vector served as a negative control (Con). Luciferase activity is represented on the y-axis and transfection efficiency was normalized by cotransfection of the renilla plasmid. Values for NRE expression in the presence of Notch ICN were normalized to 100% (lane 1). The results confirm published data [10].

https://doi.org/10.1371/journal.pone.0027986.g004

Assembly of the repressor complex was tested by cotransfecting the S2 cells with Hairless and ICN. This caused a strong downregulation of ICN mediated transcriptional activation of the NRE-reporter to about 40% [10], [12], because Hairless can assemble a repressor complex with endogenous Su(H) on the NRE promoter. Interestingly, Hairless is able to abrogate the strong activation mediated by the addition of exogenous Su(H) to near completion (Fig. 4; [10]), indicating that Hairless and Su(H) interact with each other. To our surprise, the same degree of repression was observed with Su(H)WARE (Fig. 4). Apparently, Hairless binds the Su(H) mutant with sufficiently enough affinity to repress transcription as efficiently as the wild type Su(H) protein. This result was unexpected since the yeast two-hybrid data suggested a near complete lack of binding of Su(H)WARE to Hairless. However, at this stage of our analysis, it was unclear whether this effect was specific to S2 cell culture.

In vivo transcriptional response of Notch target genes during wing development

To analyse the in vivo activity of the mutant Su(H) protein, transgenic flies were established using the PhiC31 method [15]. This system avoids position effects and hence allows the direct comparison of different transgenes at the same location. Su(H)WARE was cloned into an appropriate UAS-vector and integrated at the 96E landing site for comparison with the accordant Su(H) construct [10]. Moreover, the transgenic Su(H)WARE line was recombined with full length Hairless HFL and with mutant Hairless HLD, each integrated at 68E, to allow for a combined overexpression. The latter completely failed to bind wild type Su(H) [10]. The wild type and mutant Su(H) and Hairless transgenes were locally overexpressed using the Gal4/UAS-system [16].

First we analysed the consequences on the expression of the Notch target gene wingless (wg). Wg is expressed in the developing wing imaginal disc in a ring outlining the presumptive wing pouch and along the dorso-ventral boundary, which eventually forms the margin of the wing (Fig. 5) [17], [18]. The constructs were induced singly or in combination in a central area of the wing disc. Overexpression of either Su(H) or Su(H)WARE effected an overproliferation of the affected tissue, which is typical for Notch gain of function, suggesting that both caused the activation of Notch target genes. Accordingly, a subtle expansion of Wg expression was observed compared to the control (Fig. 5). The ectopic Wg expression in the inner and outer rings was not anticipated since wg, according to several publications, is not a Notch target in this part of the tissue [19]–[21]. As expected, overexpression of Hairless HFL antagonized the expression of Wg at the intersection of the HFL expression domain and the presumptive margin and led to less tissue due to cell death. However, induction of the mutant HLD was indistinguishable from wild type confirming complete loss of Su(H) binding in this mutant.

thumbnail
Download:
Figure 5. In vivo influence of the Su(H) and Hairless protein variants on the Notch target wingless.

Wingless (wg) protein is expressed along the dorso-ventral boundary of the wing imaginal disc as a result of Notch signalling (see arrow in lac-Z control) [17], [18]. Wingless protein is shown in green in all panels. Su(H) and Hairless variants were overexpressed using the Gal4/UAS system within the central part of the wing disc (red in lac-Z control). Apart from the beta-galactosidase control, red depicts Su(H) protein. Hairless protein is shown in blue; overlap with Su(H) appears magenta and in addition with wg, it appears white. The control disc shows the expression domain (red, beta-galactosidase) of the omb-Gal4 driver. Wingless (Wg) protein outlines the wing pouch and dissects it along the dorso-ventral boundary (arrow). The latter expression is induced by a Notch signal. Overexpression of wild type Hairless (blue) represses Wg expression (blunt bar). In contrast, the mutant HLD is unable to repress Wg. Overexpression of wild type Su(H) (red) causes proliferation of the wing blade and a subtle expansion of Wg expression. A combined overexpression of Su(H) and Hairless gives a super-additive effect: the expression domain becomes very small and Wg expression is inhibited. The mutant HLD has no such effect; the combination with Su(H) resemble the sole Su(H) overexpression reflecting lack of binding of the two proteins. Overexpression of mutant Su(H)WARE results also in a slight overproliferation of the wing disc and subtle Wg expansion. Also in combination with Hairless, Su(H)WARE strongly impedes proliferation and Wg expression. However, no such effect of HLD on Su(H)WARE can be observed.

https://doi.org/10.1371/journal.pone.0027986.g005

A combined overexpression of Hairless and Su(H) led to a remarkable loss of tissue and repression of wg expression (Fig. 5), which is in accordance with earlier observations and can be explained by the formation of a large surplus of repressor complexes formed [10]–[12], [22]. The mutant HLD fails to bind to Su(H), therefore, the combined overexpression resembled the phenotype of the sole Su(H) overexpression (Fig. 5). Again, in combination with HFL, Su(H)WARE behaved largely identical as wild type Su(H), indicating the normal formation of repressor complexes (Fig. 5), which also confirms our S2 cell culture results.

Activity Su(H) and Hairless protein variants during eye development of the fly

To substantiate these results we extended our analysis to the Drosophila eye, where Notch signalling is required at multiple, subsequent steps (reviewed in [23]). We used the gmr-Gal4 line that drives expression in the differentiating retina [24]. As reported earlier [25], overexpression of Su(H) is characterized by an overproliferation of eye tissue, as expected for a gain of Notch activity (Fig. 6). A likewise phenotype was induced by the overexpression of Su(H)WARE in accordance with its ability to assemble an activator complex together with Notch (Fig. 6a). In contrast, expression of the antagonist Hairless resulted in small, irregular eyes by interference with several Notch dependent processes and subsequent induction of apoptosis (Fig. 6b) [26]–[28]. In combination, Su(H) and Hairless overexpression led to almost eyeless flies: only small slits remained lacking ommatidial structures or eye color (Fig. 6b). Again, a similar result was observed with Su(H)WARE indicating little differences compared to wild type Su(H). In contrast, overexpression of the mutant HLD had little biological effect and did not influence the activity of the Su(H) constructs (Fig. 6b). Similar results were observed in the process of lateral inhibition during bristle formation on the thorax ([10] and not shown).

thumbnail
Download:
Figure 6. Overexpression of Su(H) und Hairless protein variants affect Drosophila eye development.

A) Ectopic expression of Su(H) (lane 2) and Su(H)WARE (lane 3) cause an increase of eye size compared with a control lac-Z (con, lane 1) ectopic expression using the gmr-Gal4 driver line. Eye size of male flies was measured from 19 to 20 individuals of each genotype. Average area are given in kilo pixel (kpx). Error bar represents standard deviation. B) The UAS transgenes Su(H) and Su(H)WARE were expressed singly or in combination together with full length Hairless HFL or HLD, respectively, using gmr-Gal4 as driver line. As a control, lacZ was overexpressed which gives a wild type looking eye. Su(H) and Su(H)WARE induce a slight overgrowth of tissue resulting in enlarged eyes. In contrast, eye specific overexpression of HFL causes smaller eyes with irregular arrangement of the ommatidia, giving a rough appearance. Overexpression of HLD causes a nearly wild type eye. A combination of Su(H) with HFL results in a complete loss of the ommatidia: only a small eye slit remains that is totally smooth and devoid of any red eye pigment. In contrast a combined overexpression of Su(H) and HLD causes enlarged eyes similar to the sole Su(H) overexpression, demonstrating lack of protein binding. Overexpression of Su(H)WARE alone or in combination with Hairless variants gives similar results suggesting again that the WARE mutation does not influence repressor complex formation.

https://doi.org/10.1371/journal.pone.0027986.g006

The small eyes resultant from Hairless overexpression are partly due to apoptosis induced by the repression of several Notch target genes and the concomitant downregulation of EGFR signalling activity [27]–[29]. We wondered whether the primary cause of the extreme adult phenotypes seen with the combined overexpression of HFL and Su(H) was also due to apoptosis. This was confirmed by staining for the cleaved, active form of Caspase-3 – the final effector Caspase in the apoptotic cascade [30] – which was dramatically increased in the eye discs of the relevant combinations (Fig. 7). In contrast no apoptosis was seen upon ectopic expression of the mutant HLD (Fig. 7). This result is in agreement with the adult eye phenotype. Because HLD fails to bind to Su(H), HLD cannot be recruited to the respective promoters to assemble the respective repression complex, explaining the absence of apoptosis.

thumbnail
Download:
Figure 7. Regulation of apoptosis by Su(H) and Hairless variants during Drosophila eye development.

The indicated UAS transgenes were overexpressed alone or in combination using the gmr-Gal4 driver line. As control the UAS-lacZ gene was likewise overexpressed. Gmr-Gal4 is active in the developing photoreceptor cells that arise posterior to the morphogenetic furrow (MF). Overexpression was visualized with specific antibodies against beta-galacotsidase (red, control), Su(H) (red) and Hairless (blue). In the merge, expression appears magenta. Apototic cells were detected with an antibody directed against activated Caspase 3 (green). A surplus of Hairless induces apoptosis which is strongly enhanced by the combined overexpression with either Su(H) or Su(H)WARE. In contrast, HLD had no such an impact.

https://doi.org/10.1371/journal.pone.0027986.g007

Discussion

While we have detailed knowledge of the structure of the ternary activator complex (CSL-ICN-MAM), repression of Notch signal transduction is less well understood (overview in [4], [7]). We have started a more detailed analysis on the Notch repression complex in Drosophila which contains the CSL-type DNA binding protein Su(H), the bridging platform protein Hairless and the two general corepressors, Groucho and C-terminal binding protein [9], [11], [12], [31]. We have shown recently that Su(H) and Hairless form a high affinity complex that involves the CTD of Su(H) and the NT-domain of Hairless [10]. Moreover, our work demonstrated that Notch can outcompete Hairless for the binding of Su(H). This observation is startling for two reasons: firstly, both Notch and Hairless show comparable affinity for Su(H) which is in the nanomolar range, and secondly, the two molecules contact different sites in Su(H), excluding a simple competition scenario [10]. Presumably, the switch between activator and repressor status is more complicated and may involve structural changes in Su(H) [32], [33].

In order to provide the molecular basis for a deeper understanding of these processes we have started to map the Su(H)-Hairless sites of interaction. In Hairless, we have been able to determine a single amino acid that is crucial for the binding of Su(H) without overtly disturbing Hairless structure [10]. Here we identify two residues W491 and R493 in the CTD of Su(H) that likely contribute to the binding of Hairless based on our yeast two-hybrid data. Strikingly, the WARE mutant behaved similar to wild type in our cellular and in vivo assays. How could this discrepancy in our results be resolved? The simplest explanation is the involvement of one or more additional contact sites in Su(H) located elsewhere that sufficiently stabilize the binding of the full length Hairless and Su(H) proteins in vivo, but not the interaction between CTD and Hairless in the yeast assay. As we have already mutated most of the surface exposed residues in Su(H) CTD without affecting Hairless binding, we must conclude that single mutations are not disruptive and that we have not fortuitously hit upon the right combination of multiple amino acids in Su(H) to completely disrupt binding. Certainly the determination of the Su(H)-Hairless complex crystal structure will clarify the role of these residues in Hairless binding, as well as define other important interaction regions.

However, we also need to consider the quantitative differences in the approaches. In the yeast the molecules are tested in a near 1∶1 molar ratio (assuming equal expression and stability of the proteins), whereas both in vivo approaches were based on overexpression and hence assayed with an excess of Su(H). S2 cells lack Notch but express both Su(H) and Hairless [13], [34], [35]; and the endogenous levels of Su(H) are sufficient for a strong response to experimental ICN doses [13], [14], [36]. Addition of Su(H) enhances the ICN response nearly fourfold [10], [14], [36], similar to the in vivo situation, where overexpression of Su(H) elicits Notch gain of function phenotypes [12], [22], [37]. We have no information on the amount of Su(H) in a cell that is freely available for binding to either Notch or Hairless, and the above observations suggest that Su(H) is limiting. Clearly, Su(H) occupancy on Notch target gene promoters is highly dynamic and enhanced by the presence of ICN [35]. However, Drosophila cells express high levels of Su(H) in the cytoplasm which is rather unconventional for a transcription factor [22], [38], [39]. The mechanisms underlying Su(H) nuclear import/export are little understood and may involve Notch signalling and repression, respectively [38]–[41]. Most likely cytoplasmic Su(H) is unavailable for transcriptional complex formation, be it repressor or activator complex. Moreover, Su(H) from the cytoplasm may resupply the nucleus once Su(H) is bound in complexes such that a steady level of free nuclear Su(H) is achieved. Overexpression of Su(H) may raise this level considerably, explaining the increase of Notch output in the cell culture as well as in fly tissue. Hairless would access the same free pool of Su(H), however, only bind to wild type protein to build up repressor complexes and silence Notch target genes, whereas Notch could access both mutant and wild type Su(H). Assuming a large enough pool of wild type Su(H), then repressor complex formation might in fact be as efficient in the presence of mutant as of wild type overexpressed Su(H) (Fig. 8). This rather speculative model would be in agreement with an in vivo reduced or lack of binding between Hairless and Su(H)WARE and concomitant repression by a complex consisting primarily of Hairless and endogenous Su(H). Experiments addressing the exact composition of the repressor complexes, i.e. the presence of mutant Su(H), may help support this model.

thumbnail
Download:
Figure 8. Modell of a Su(H) cytoplasmic pool.

A) The large cytoplasmic pool of Su(H) is not available for complex formation in the nucleus, however, is used to replenish nuclear Su(H). This explains the apparent limitation of Su(H) when ICN is overexpressed. The export mechanism is currently unknown and may represent a further level of Su(H) regulation. B) Overexpression of Su(H) increases the cytoplasmic protein pool, thereby raising the level of available nuclear Su(H). This may cause Notch gain of function phenotypes in tissues, where Notch signalling takes place, or in the presence of exogenous ICN. The darker colour represents mutant Su(H) protein to exemplify the model; wild type Su(H) would behave identical. C) Combined overexpression of Su(H) and Hairless allows formation of repressor complex, as long as endogenous wild type Su(H) is available from the cytoplasmic pool. Lack of binding of Hairless to mutant Su(H) predicts an enrichment of activator complexes containing the mutant Su(H) and ICN.

https://doi.org/10.1371/journal.pone.0027986.g008

Materials and Methods

Yeast two-hybrid experiments

Single, double or triple missense mutations in CTD were introduced using the QuickChangeR II XL site-directed mutagenesis kit (Stratagene). All mutants were sequence verified (StarSeq, Mainz). CTD mutants were cloned into pJG vector [42] and tested for protein interactions with pEG constructs as described previously [10]. Primer sequences are available upon request. The yeast three-hybrid experiments were performed with the N-ANK domain cloned in the pESC-Leu vector (Stratagene) and pEG-MamN as outlined in [10].

The CTDWARE mutant DNA was excised with Msc I and Eco 52I from the CTDWARE pJG-construct and reintroduced in likewise digested Su(H) cDNA to generate the mutant full length construct Su(H)WARE. It was shuttled into pRmHa-3 [43] and pUAST-attB- vectors [15] for subsequent in vivo analyses.

Electro-mobility shift assays - EMSA

DNA binding assays of Su(H) and Su(H) mutants were performed according to standard protocols using a double stranded DNA-oligomer (made by hybridization of primers 5′ GGT TCT TTC AGC TCG GTT CCC ACG CCA CGA GCC AC 3′ and 5′ TTG GGT GGC TCG TGG CGT GGG AAC CGA GCT GAA AG 3′ and labelled with Klenow polymerase) containing the E(spl)m8-S1 Su(H) binding site [44] and Su(H) proteins produced from cDNA by in vitro transcription/translation using the TNT®Coupled Reticulocyte Lysate System (Promega).

Cell culture assays

For cell culture experiments Drosophila Schneider S2 cells, obtained from the Drosophila Genomics Resource Centre DGRC (Indiana University, Bloomington USA), were transfected with the respective constructs and the activity measured with a Notch responsive luciferase reporter (NRE-reporter) as described previously [13]. Renilla expression plasmid (tk-Renilla; Promega) was cotransfected as internal control. Reporter activation elicited by transfection with pMT-ICN was taken as 100% [45]. Cotransfection with Su(H) and Hairless constructs were analysed as described before [10]. CuSO4 was used to induce protein expression 6 h after transfection. Luciferase activity was measured 18 h later in duplicate (Lumat LB 9507, EG & Salem, MA) using the dual-luciferase reporter assay system (Promega).

Analysis of mutant Hairless and Su(H) transgenes in vivo

Transgenic Su(H)WARE flies were generated with the PhiC31 integrase-based integration system [15] to avoid position effects and allow for a direct comparison with likewise integrated wild type Su(H) [10]. For co-overexpression experiments, the Su(H)WARE insertion located at 96E was recombined with HFL and HLD, respectively, located at 68E [10]. Tissue specific overexpression was achieved with the Gal4/UAS-system [16] using omb-Gal4 and gmr-Gal4 driver lines (http://flybase.org).

Antibody staining of imaginal discs was performed as described before using the following antisera: anti H-A [46], anti-Su(H) (Santa Cruz Biotech), anti-cleaved Caspase 3 (NEB Cell Signaling Technologie) anti-wg as well as anti-beta-galactosidase (developed by M. Cohen and J.R. Sanes, respectively, and obtained from Developmental Studies Hybridoma Bank developed under the auspices of the NICHD and maintained by The University of Iowa, Department of Biology, Iowa City, IA 52242). Secondary antibodies coupled to DATF, Cy3 or Cy5 were purchased from Jackson Laboratory (Dianova). Samples were mounted in Vectashield (Vector Lab) and analyzed on a Zeiss Axiophot linked to a Bio-Rad MRC1024 confocal microscope. Flies were monitored using an ES120 camera (Optronics), with Pixera Viewfinder Version 2.0 software. Pictures were assembled with Corel-PhotoPaint and CorelDRAW Version 9.0 software.

Acknowledgments

We acknowledge S. Artavanis-Tsakonas, S. Bray and C. Delidakis for sharing DNA-constructs and F. Karch for sending the PhiC31 fly lines and vectors. We thank A. Schulz and I. Wech for invaluable technical help.

Author Contributions

Conceived and designed the experiments: AP RAK DM. Performed the experiments: PK DM. Analyzed the data: PK AP RAK DM. Contributed reagents/materials/analysis tools: PK AP RAK DM. Drafted and revised the manuscript: AP RAK DM. Critically read the article: PK.

References

  1. 1. Artavanis-Tsakonas S, Matsuno K, Fortini ME (1995) Notch signaling. Science 268: 225–268.
  2. 2. Bray SJ (2006) Notch signalling: a simple pathway becomes complex. Nat Rev Mol Cell Biol 7: 678–689.
  3. 3. Kopan R, Ilagan MXG (2009) The canonical Notch signaling pathway: Unfolding the activation mechanism. Cell 137: 216–233.
  4. 4. Kovall RA, Blacklow S (2010) Mechanistic insights into Notch receptor signaling from structural and biochemical studies. Curr Top Dev Biol 92: 31–71.
  5. 5. Nam Y, Sliz P, Song L, Aster JC, Blacklow SC (2006) Structural basis for cooperativity in recruitment of MAML coactivators to Notch transcription complexes. Cell 124: 973–983.
  6. 6. Wilson JJ, Kovall RA (2006) Crystal structure of the CSL-Notch-mastermind ternary complex bound to DNA. Cell 124: 985–996.
  7. 7. Borggrefe T, Oswald F (2009) The Notch signaling pathway: Transcriptional regulation at Notch target genes. Cell Mol Life Sci 66: 1631–1646.
  8. 8. Maier D (2006) Hairless, the ignored anatagonist of the Notch signalling pathway. Hereditas 143: 212–221.
  9. 9. Barolo S, Stone T, Bang AG, Posakony JW (2002) Default repression and Notch signaling: Hairless acts as an adaptor to recruit the corepressors Groucho and dCtBP to Suppressor of Hairless. Genes Dev 16: 1964–1976.
  10. 10. Maier D, Kurth P, Schulz A, Russell A, Yuan Z, et al. (2011) Structural and functional analysis of the repressor complex in the Notch signaling of D. melanogaster. Mol Biol Cell 22: 3242–3252.
  11. 11. Morel V, Lecourtois M, Massiani O, Maier D, Preiss A, et al. (2001) Transcriptional repression by Suppressor of Hairless involves the binding of a Hairless-dCtBP complex in Drosophila. Curr Biol 11: 789–792.
  12. 12. Nagel AC, Krejci A, Tenin G, Bravo-Patiño A, Bray S, et al. (2005) Hairless mediated repression of Notch target genes requires combined activity of Groucho and CtBP co-repressors. Mol Cell Biol 25: 10433–10441.
  13. 13. Bray SJ, Musisi H, Bienz M (2005) Bre1 is required for Notch signaling and histone modification. Dev Cell 8: 279–286.
  14. 14. Matsuno K, Go MJ, Sun XS, Eastman DS, Artavanis-Tsakonas S (1997) Suppressor of Hairless-independent events in Notch signaling imply new pathway elements. Development 124: 4265–4273.
  15. 15. Bischof J, Maeda RK, Hediger M, Karch F, Basler K (2007) An optimized transgenesis system for Drosophila using germ-line-specific phiC31 integrases. Proc Natl Acad Sci U S A 104: 3312–3317.
  16. 16. Brand AH, Perrimon N (1993) Targeted gene expression as a means of altering cell fates and generating dominant phenotypes. Development 118: 401–415.
  17. 17. Días-Benjumea FJ, Cohen SM (1995) Serrate signals through Notch to establish a Wingless-dependent organizer at the dorsal/ventral compartment boundary of the Drosophila wing. Development 121: 4215–4225.
  18. 18. Rulifson EJ, Blair SS (1995) Notch regulates wingless expression and is not required for reception of the paracrine wingless signal during wing margin neurogenesis in Drosophila. Development 121: 2813–2824.
  19. 19. Del Álamo Rodríguez DA, Terriente J, Galindo MI, Couso JP, Díaz-Benjumea FJ (2002) Different mechanisms initiate and maintain wingless expression in the Drosophila wing hinge. Development 129: 3995–4004.
  20. 20. Perea D, Terriente J, Díaz-Benjumea FJ (2009) Temporal and spatial windows delimit activation of the outer ring of wingless in the Drosophila wing. Dev Biol 328: 445–455.
  21. 21. Terriente J, Perea D, Suzanne M, Díaz-Benjumea FJ (2008) The Drosophila gene zfh2 is required to establish proximo-distal domains in the wing. Dev Biol 320: 102–112.
  22. 22. Furriols M, Bray S (2000) Dissecting the mechanisms of Suppressor of Hairless function. Dev Biol 227: 520–532.
  23. 23. Voas MG, Rebay I (2004) Signal integration during development: insights from the Drosophila eye. Dev Dyn 229: 162–175.
  24. 24. Hay BA, Maile R, Rubin GM (1997) P element insertion-dependent gene activation in the Drosophila eye. Proc Natl Acad Sci U S A 94: 5195–5200.
  25. 25. Müller D, Nagel AC, Preiss A (2006) A molecular link between Hairless and Pros26.4, a member of AAA-ATPase subunits of the proteasome 19S regulatory particle in Drosophila. J Cell Sci 119: 250–258.
  26. 26. Müller D, Kugler SJ, Preiss A, Maier D, Nagel AC (2005) Genetic modifier screens on Hairless gain-of-function phenotypes reveal genes involved in cell differentiation, cell growth and apoptosis in Drosophila. Genetics 171: 1137–1152.
  27. 27. Nagel AC, Preiss A (2011) Fine tuning of Notch signaling by differential co-repressor recruitment during eye development of Drosophila. Hereditas 148: 77–84.
  28. 28. Protzer CE, Wech I, Nagel AC (2008) Hairless induces cell death by downregulation of EGFR signalling activity. J Cell Sci 121: 3167–2176.
  29. 29. Wech I, Nagel AC (2005) Mutations in rugose promote cell type-specific apoptosis in the Drosophila eye. Cell Death Differ 12: 145–152.
  30. 30. Yu SY, Yoo SJ, Yang L, Zapata C, Srinivasan A, et al. (2002) A pathway of signals regulating effector and initiation caspases in the developing Drosophila eye. Development 129: 3269–3278.
  31. 31. Brou C, Logeat F, Lecourtois M, Vandekerckhove J, Kourilsky P, et al. (1994) Inhibition of the DNA-binding activity of Drosophila Suppressor of Hairless and of its human homolog, KBF2/RBP-J kappa, by direct protein-protein interaction with Drosophila Hairless. Genes Dev 8: 2491–2503.
  32. 32. Friedmann DR, Wilson JJ, Kovall RA (2008) RAM-induced allostery facilitates assembly of a Notch pathway active transcription complex. J Biol Chem 283: 14781–14791.
  33. 33. Kovall RA (2008) More complicated than it looks: assembly of Notch pathway transcription complexes. Oncogene 38: 5099–5109.
  34. 34. Fehon RG, Kooh PJ, Rebay I, Regan CL, Xu T, et al. (1990) Molecular interactions between the protein products of the neurogenic loci Notch and Delta, two EGF-homologous genes in Drosophila. Cell 61: 523–534.
  35. 35. Krejci A, Bray SJ (2007) Notch activation stimulates transient and selective binding of Su(H)/CSL to target enhancers. Genes Dev 21: 1322–1327.
  36. 36. Eastman DS, Slee R, Skoufos E, Bangalore L, Bray S, et al. (1997) Synergy between Suppressor of Hairless and Notch in regulation of Enhancer of split mgamma and mdelta expression. Mol Cell Biol 17: 5620–5628.
  37. 37. Schweisguth F, Posakony JW (1994) Antagonistic activities of Suppressor of Hairless and Hairless control alternative cell fates in the Drosophila adult epidermis. Development 120: 1433–1441.
  38. 38. Fortini ME, Artavanis-Tsakonas S (1994) The Suppressor of Hairless protein participates in Notch receptor signaling. Cell 79: 273–282.
  39. 39. Gho M, Lecourtois M, Geraud G, Posakony JW, Schweisguth F (1996) Subcellular localization of Suppressor of Hairless in Drosophila sense organ cells during Notch signalling. Development 122: 1673–1682.
  40. 40. Frise E, Knoblich JA, Younger-Shepherd S, Jan LY, Jan YN (1996) The Drosophila Numb protein inhibits signaling of the Notch receptor during cell-cell interaction in the sensory organ lineage. Proc Natl Acad Sci U S A 93: 11925–11932.
  41. 41. Maier D, Nagel AC, Johannes B, Preiss A (1999) Subcellular loclaization of Hairless protein shows a major focus of activity within the nucleus. Mech Dev 89: 195–199.
  42. 42. Gyuris J, Golemis E, Chertkov H, Brent R (1993) Cdi1, a human G1 and S phase protein phosphatase that associates with cdk2. Cell 75: 791–803.
  43. 43. Bunch TA, Grinblat Y, Goldstein LS (1988) Characterization and use of the Drosophila metallothionein promoter in cultured Drosophila melanogaster cells. Nucleic Acids Res 16: 1043–1061.
  44. 44. Bailey AM, Posakony JW (1995) Suppressor of Hairless directly activates transcription of Enhancer of split Complex genes in response to Notch receptor activity. Genes Dev 9: 2609–2622.
  45. 45. Matsuno K, Diederich RJ, Go MJ, Blaumüller CM, Artavanis-Tsakonas S (1995) Deltex acts as a positive regulator of Notch signaling through interactions with the Notch ankyrin repeats. Development 121: 2633–2644.
  46. 46. Maier D, Nagel AC, Preiss A (2002) Two isoforms of the Notch antagonist Hairless are produced by differential translation initiation. Proc Natl Acad Sci U S A 99: 15480–15485.