Hairless: the ignored antagonist of the Notch signalling pathway

Hereditas 143: 212 221 (2006)

Hairless: the ignored antagonist of the Notch signalling pathway DIETER MAIER Institute for Genetics (240), University of Hohenheim, Stuttgart, Germany Maier, D. 2006. Hairless: the ignored antagonist of the Notch signalling pathway. * Hereditas 143: 212 221. Lund, Sweden. eISSN 1601-5223. Received August 18, 2006. Accepted October 3, 2006 The Notch signalling pathway regulates cell-cell communication in higher eukaryotes. Cellular differentiation and tissue development relies on correct intercellular communication, accounting for the high interest in the Notch signalling pathway. Together with mastermind and CSL (CBF-1, Suppressor of Hairless, lag-2) DNA-binding proteins, Notch forms a complex that mediates transcriptional activation of the respective target genes. This activation is strictly controlled, and deregulation causes extreme developmental defects. In Drosophila , the stringency of the control system is given by the general Notchantagonist Hairless. Hairless assembles in a repressor complex on Notch target genes, which involves Suppressor of Hairless and two corepressors, Groucho and C-terminal binding protein. In mammals, CBF-1 recruits corepressors on its own. In addition Hairless recruits also other proteins. One example is the Pros26.4 AAA-ATPase which specifically destabilises Hairless resulting in a novel positive regulation of Notch signalling. By inhibition of Notch, Hairless not only regulates cellular differentiation but also has anti-apoptotic functions. Moreover, many genetic interactions imply a cross-talk between Hairless and the EGF-receptor pathway, which might act independently of Notch. Surprisingly, no Hairless homologue has been identified in mammals so far, despite the high degree of conservation of other components of the pathway. This discrepancy might be resolved in the future, once all components of the repressor-complex in the different species have been identified. In conclusion, Hairless is a central component of the regulation of the Notch signalling pathway in Drosophila , and is hence essential for cell differentiation and tissue development in the fly. Dieter Maier, Institute for Genetics (240), University of Hohenheim, DE-70599 Stuttgart, Germany. E-mail: maierdie@ uni-hohenheim.de

Tissue development and cellular differentiation requires communication between the respective cells. In higher eukaryotes, cell-cell communication is governed by the Notch signalling pathway. This pathway allows cells with equal potential to meet different developmental fates and hence differentiate to various cell types. Not surprisingly, Notch signalling is involved in a number of human diseases, ranging from cancer to developmental disorders (ARTAVANIS-TSAKONAS et al. 1995, 1999). In this review, I focus on the role of Hairless, a general antagonist of the Notch signalling pathway in Drosophila . Activation of the Notch signalling pathway The Notch (N) signalling pathway was first described in the process of lateral inhibition during neurogenesis in Drosophila melanogaster (reviewed by ARTAVANISTSAKONAS et al. 1995). In this process, a single cell in a cluster of equipotent cells is selected to adopt a specific cell fate by inhibiting the adjacent cells via the Notch signalling pathway to engage in the same cell fate. Intensive studies elucidated additional Notch-regulated developmental processes, e.g. cell fate decisions in asymmetric cell divisions (HARTENSTEIN and POSAKONY 1990; SIMPSON 1997; GHO et al. 1999) or boundary formation of the wing margin (KIM et al. 1995; DE CELIS et al. 1996; MICCHELLI et al. 1997). Accordingly, the Notch signalling pathway is crucial in

many different tissues at all developmental stages in Drosophila (BRAY 1998; ARTAVANIS-TSAKONAS et al. 1999; PORTIN 2002). The pathway is activated by the ligands Delta (Dl) or Serrate (Ser) by binding the transmembrane receptor Notch. The interaction occurs in the extracellular space between two adjacent cells. After ligand binding, the intracellular domain of Notch (Nintra) is proteolytically cleaved by presenilin proteins. Subsequently, Nintra migrates to the nucleus. There it forms a transcriptional activator complex with Suppressor of Hairless [Su(H)] and mastermind (mam). Mam acts as a strong coactivator and can only be bound by Su(H) in the presence of Nintra, resulting in a context specific transcription of Notch target genes (MUMM and KOPAN 2000). Su(H) is a DNA binding protein of the rel family with the mammalian homologues RBP-Jk or CBF-1 (overview in BARRICK and KOPAN 2006). Mammalian CBF-1 recruits corepressors itself and hence acts as a repressor of Notch target genes in the absence of a Notch signal (LUBMAN et al. 2004). In Drosophila , however, direct corepressor binding of Su(H) has not been demonstrated so far. In the case of lateral inhibition, the Notch targets are the genes of the Enhancer of split-Complex [E(spl)-C], which encode basic Helix-loop-Helix (bHLH) proteins. These are inhibitors of genes that determine the primary proneural fate (ARTAVANIS-TSAKONAS et al. 1995; BAILEY and POSAKONY 1995; CASTRO et al.

Hereditas 143 (2006)

2005). As a consequence, cells receiving the Notch signal stay in an undefined status open for other differentiation signals, and only the signalling cell adopts the specific fate initiated by the proneural proteins. The discovery that this is not only the case for Drosophila but also true for all higher eukaryotes including humans led to the conclusion that the Notch signalling cascade is one of the most important pathways in animal development (ARTAVANISTSAKONAS 1997; LEWIS 1998; GREENWALD 1998). This is reflected by the high number of excellent review articles, dealing with various aspects of Notch signalling. However, a major component of the Notch signalling pathway in Drosophila , Hairless (H) is obviously a ‘‘hot potato’’, and only rarely discussed in these articles. Hairless acts as common antagonist of the Notch signalling pathway in Drosophila . Hairless acts as a general antagonist by inhibiting the Notch-signal (ARTAVANIS-TSAKONAS et al. 1999). The Hairless locus was already discovered in 1923 by Bridges and Morgan (BRIDGES and MORGAN 1923), as mutations in Hairless like in Notch and Dl are haplo-insufficient. There are many mutants known, and all of them can easily be identified due to the dominant loss of bristles and wing venation defects (LINDSLEY and ZIMM 1992). Based on genetic interactions with all central Notch signalling pathway components, i.e. Notch itself, Dl , Su(H) , mam and the genes of the E(spl)-C, Hairless was classified as pathway member as well (VA¨SSIN et al. 1985; SCHWEISGUTH and POSAKONY 1994). Notch was named according to its dominant phenotype: one or a couple of notches in the wing margin. The dominant wing phenotype of Dl is exhibited by wing veins formed as deltas on the wing margins. Both mutant phenotypes, as well as the Hairless phenotype, are suppressed in double heterozygotes together with Hairless, resulting in almost wild type looking animals (LINDSLEY and ZIMM 1992; MAIER et al. 1992). These genetic interactions demonstrate that Hairless antagonizes the Notch signalling cascade in a dose dependent manner, and only if the signalling proteins and Hairless are in a 1:1 ratio, the pathway functions correctly. Conclusively, a functional antagonist Hairless in the Notch signalling pathway is equally crucial as the other signalling components. Why then is Hairless so rarely considered in review articles? Already in 1999 R. Kopan stated: ‘‘Good things must come to an end: how is Notch signalling turned off ?’’ (KOPAN 1999). One reason might be that only recently various mechanisms that down-regulate Notch-signalling

Hairless: ignored antagonist of the Notch pathway

213

have started to emerge. The other reason might be that Hairless appears to be specific to Notch signalling in insects. Down-regulation of Notch signalling in higher eukaryotes Despite extensive efforts, no Hairless orthologue was found in vertebrates so far. Though a hairless gene in vertebrates exists, mutants thereof lead to nude animals (STRAUSBERG et al. 2002). It seems that this gene has nothing in common with the Drosophila Hairless gene. It was proposed that the Msx2-interacting nuclear target protein (MINT) is a functional homologue of the Drosophila Hairless protein, since it suppresses Notch/ RBP-Jk signalling in the regulation of marginal zone B cell development (KURODA et al. 2003). However, MINT and Hairless show absolutely no structural similarities. Moreover, this hypothesis is weakened by the fact that there are two genes in the Drosophila genome encoding proteins similar to MINT, that belong to the split ends (spen ) class (JEMC and REBAY 2006). They have two domains in common, the RPR (RNA binding domain) and SPOC (Spen Paralogue and Orthologue C-terminal domain), which are both present in MINT, Spen and a second Drosophila protein encoded by the spen like gene spenito (CG2910 ; JEMC and REBAY 2006). Furthermore it has been shown with the help of Drosophila spen mutants, that the Spen protein is a tissue specific antagonist of the Notch signalling pathway (LIN et al. 2003). Numb is another well known Notch antagonist that is conserved in Drosophila and mammals. Numb protein asymmetrically associates with the membrane. It shows binding activity to Notch. Numb may play a role in endocytosis and/or ubiquitination of Notch during asymmetric cell divisions (LE BORGNE et al. 2005). Therefore, Numb can also be regarded as a tissue and context specific Notch antagonist. In contrast, Hairless acts as a general antagonist in the known Notch mediated signalling processes. Up to now, only Hairless orthologues of insects were identified. A detailed understanding of the invertebrate Hairless protein is a good prerequisite to identify the true functional homologues in mammals. Molecular characterisation of Hairless The cloning of the Hairless gene revealed a protein of a theoretical molecular weight of 110 kDa. This novel protein did not show any known structural or functional molecular domains (BANG and POSAKONY 1992; MAIER et al. 1992). Hairless is expressed in all developmental stages and is also maternally contributed. The transcript is enriched in the meso-

214

D. Maier


derm of embryos, and uniformly expressed in all larval tissues. Ubiquitous expression is also found for the Hairless protein that is mainly nuclear but also present in the cytoplasm (MAIER et al. 1999). Although no alternative splicing was found, the anti-Hairless antibody detects two distinct protein isoforms of 150 and 120 kDa on Western blots. It turned out that the 120 kDa protein is a result of a cap independent translation starting at an internal ribosome entry site (IRES) (Fig. 1, 2) (MAIER et al. 2002). A second Hairless isoform is produced by help of an internal ribosome entry site IRES sequences were first identified in picorna viruses within the 5’ untranslated region and allow cap-independent translation start. IRES sequences now are known for a number of cellular mRNAs (reviewed by CARTER et al. 2000). There, the functional significance is not completely understood. It might be used in cell cycle regulated protein synthesis allowing translation during mitosis and/or in cellular stress situations. Normally, IRES sequences are located in the 5’ untranslated gene region. However, in the Drosophila melanogaster Hairless gene, the IRES sequence is located in the 5’ coding region (MAIER et al. 2002) (Fig. 1). A similar situation was reported for the human ornithine decarboxylase gene (CORNELIS et al. 2000). Two Hairless isoforms are produced in Drosophila , one cap dependent translation product of 150 kDa (Hp150) and one of 120 kDa (Hp120) starting the translation at the IRES (Fig. 2). The Hp120 isoform initiates at the AUG codon

1 D. melanogaster D. simulans D. sechellia D. yakuba D. erecta D. ananassae D. pseudoobscura D. persimilis D. willistoni D. mojavensis D. hydei D. virilis D. grimshawi

14

position 148 (M148). This was verified with constructs where the presumptive translational start sites (AUG) were changed to isoleucine codons (GUG) in vivo and in vitro as well as in bicistronic constructs. Western blots confirmed that the detected proteins corresponded to the expected sizes. In transgenic flies all tested Hairless IRES mutant constructs failed to compensate for the genetic defect. As only a wild type construct producing both protein isoforms restored a normal phenotype, it was assumed that both isoforms are needed for wild type function. This is surprising since both isoforms comprise all important structural domains yet classified in Hairless (see below and Fig. 2). Particularly interesting is the fact that the IRES dependent isoform Hp120 is enriched after ectopic expression in mitotic cells in contrast to the cap dependent isoform Hp150. These findings supporting the notion that Hp120 plays an important role during mitosis, whereas Hp150 mediates Hairless function primarily during the interphase (MAIER et al. 2002). In summary, Hairless seems to be required throughout the cell cycle. The functional relevance of the IRES is supported by homology data: two protein isoforms are detected in D. hydei , D. obscura and Musca domestica on Western blots. A corresponding methionine was experimentally verified in the Hairless protein sequence of D. hydei (MARQUART et al. 1999). In addition, not only a second start codon for a potential short isoform is highly conserved in all Drosophila species sequenced to date, but also the surrounding residues (analysed by FlyBase tblastn, Fig. 1). Typically, multiple serine residues are located N-terminally and several alanine residues are Cterminally of the respective methionine (Fig. 1).

Subgenus

Structure-function analysis of Hairless Sophophora

Drosophila

Fig. 1. The internal ribosome entry results in translation start at the third methionine start codon (M148) in Drosophila melanogaster (arrow). A high degree of conservation of the surrounding residues is predicted in several Drosophila species suggesting that two Hairless isoforms are produced in these flies as well. Notably, the start methionine is followed in all species by multiple alanine residues. In blue are identical residues, red similar, yellow less conserved residues.

The first functional protein domain which was defined in D. melanogaster Hairless was the Suppressor of Hairless [Su(H)] binding domain (BROU et al. 1994). Su(H) alleles were genetically identified as suppressors of the dominant Hairless phenotype (ASHBURNER 1982; LINDSLEY and ZIMM 1992). Hence it is not surprising that Hairless and Su(H) proteins show mutual binding activity as confirmed by pull down experiments. The Su(H) binding domain (SBD) was mapped to the first half of the Hairless protein to a region encompassing residues 212 to 293 (BROU et al. 1994). The mere fact that Hairless has 1076 residues implies that there are more functional domains. An in vivo approach using deletion constructs was chosen for a systematic structure-function analysis. Overexpression of a Hairless transgene resulted in the


Hereditas 143 (2006) A

215

Minor cap dependent isoform Main cap dependent isoform IRES dependent isoform 12

3

SBD

Acid

GBD

S4BD

CBD

B FLH C1 C2 C3 CX C6

Fig. 2. (A) The Hairless gene encodes three protein isoforms. A minor protein isoform is translated from the first AUG (1). From the second AUG (2) the main cap dependent isoform Hp150 is produced. This start codon is linked to a good Drosophila consensus site for translation initiation (CAVENER 1987). The third IRES dependent isoform is initiated at the AUG (3) at codon M148 (MAIER et al. 2002). A schematic view of the Hairless protein highlights in colour the important functional domains. These include the domains interacting with Su(H) (SBD, red), with Gro (GBD, light blue) and with CtBP (CBD, yellow), respectively, (BAROLO et al. 2002; NAGEL et al. 2005) as well as the putative region interacting with Pros26.4 (S4BD, magenta) (MU¨LLER et al. 2006). The acidic domain (Acid) that has a putative silencing activity is shown in deep blue (MAIER et al. 1997). (B) Deletion constructs C1-C6 and CX that were used in a systematic structure-function analysis, are schematically depicted. They were functionally tested in vivo for their residual Hairless activity in comparison to full length Hairless (FLH) (MAIER et al. 1997).

opposite phenotype of the loss of function mutation, namely the formation of ectopic bristles (antiHairless phenotype) (BANG and POSAKONY 1992; MAIER et al. 1997). Several Hairless constructs missing different parts of the protein were generated and tested in vivo. It was expected that those constructs that lacked important structural elements of Hairless should result in weaker or no anti-Hairless phenotypes and lack the ability to complement Hairless mutants (MAIER et al. 1997). The results of this structurefunction analysis of Hairless are well supported by data from an evolutionary comparison between the homologous proteins of D. hydei and D. melanogaster. The D. hydei gene was isolated via RT-PCR and the cDNA was sequenced (MARQUART et al. 1999). The protein comparison revealed an overall amino acid identity of /69%. This is a relatively weak homology compared to other members of the Notch signalling pathway. For example Notch, Dl and Su(H) are much higher conserved even between vertebrates and Drosophila (ARTAVANIS-TSAKONAS et al. 1999). The Su(H) orthologue in mouse (CBF-1 or RBP-Jk) shares an amino acid identity to the Drosophila melanogaster protein of 82% over large portions (FURUKAWA et al. 1991). Although the Hairless proteins are widely diverged, highly conserved protein stretches are found in all functionally important regions except for the very N-terminal portion and the acidic domain (Fig. 2).

Characterisation of functional domains in Hairless The most dramatic reduction of Hairless activity was noted for the construct missing the SBD (Fig. 2, C2; MAIER et al. 1997). This domain also comprises the largest conserved sub-region in Hairless, which was used to narrow down the SBD more precisely. It contains approximately 130 residues with an identity of /87% between D. melanogaster and D. hydei . Unexpectedly, overexpression of the C2-construct produced a net-like wing venation phenotype arising from cross-talk with EGF signalling (JOHANNES and PREISS 2002). This result demonstrates on one hand that Su(H) is a very important interaction partner and suggests on the other hand that there are Su(H) independent functions of Hairless (see below). There are two other deletions in Hairless that dramatically reduce its function. One of these truncated versions misses the 15 most C-terminal residues (Fig. 2, C6), the other one lacks a large portion of the C-terminal region but keeps the extreme C-terminus (Fig. 2, CX). Of the 15 very C- terminal residues, the last 10 are identical between melanogaster and hydei . A close inspection revealed a binding motif for the C- terminal Binding Protein (CtBP), which was confirmed in subsequent studies (MOREL et al. 2001). Moreover, the C-terminal third of Hairless (CX) contains several highly conserved regions with yet unknown functions. At least one further protein binds to Hairless within this domain and was identified in a yeast two-hybrid

216

D. Maier

screen: the S4 subunit (Pros26.4) of the proteasome (MARQUART et al. 1999; MU¨LLER et al. 2006). Pros26.4, an AAA-ATPase plays an important role in the regulatory subunit of the 26S proteasome. The Drosophila gene Pros26.4 is the orthologue of yeast Rpt2 . The Hairless protein interacting domain is located in the N-terminal part of Pros26.4, but not in the C-terminal AAA-ATPase domain. Reduction of the cellular Pros26.4 level using RNA interference stabilizes the Hairless protein, however, neither Notch nor Su(H). This interaction between Hairless and Pros26.4 reveals a novel positive regulation route of the Notch signalling pathway (MU¨LLER et al. 2006). A further interesting domain comprises an acidic protein sequence following the SBD (Fig. 2; C3). As its deletion enhances Hairless activity, the intact acidic domain may act as a silencer (MAIER et al. 1997). An evolutionary comparison between the Hairless orthologues from D. melanogaster, D. hydei and Anopheles gambiae reveals no conservation in this region. However, in all three species these regions are remarkably acidic in the otherwise highly basic Hairless protein (MARQUART et al. 1999; unpubl. data). This leads to an isoelectric point (pI) of less than five in this protein part, whereas the overall pI is above ten. The Nterminal protein part was found to be of functional importance as overexpression of a Hairless construct missing this domain (Fig. 2; C1) results in strong antiHairless phenotypes. However, this construct lacks rescue capacity at ambient temperatures (MAIER et al. 1997). This truncated construct should initiate with M148, which is also the initiation methionine of the IRES dependent Hairless isoform Hp120 (Fig. 2, MAIER et al. 2002). Hence the observed reduced rescue capacity compared to the corresponding wild type Hp120 isoform is unexpected. The following explanation comes to mind: for proper ribosomal recognition, the IRES needs RNA-stem loops in front of the start methionine codon. Such stem loops are predicted by computer analysis in the H-IRES. However, C1 lacks these sequences (Fig. 2). Therefore, translation of Hp120 might be reduced to a minimum (summarized by EHRENFELD and SEMLER 1995). After heat shock stimulation of the hsp 70 heat shock promoter that drives expression of the Hairless construct, ample mRNA is available to translate sufficient amounts of protein to exhibit an anti-Hairless phenotype. How does Hairless antagonize the Notch signalling pathway? Su(H) is thought to act jointly with Notch and mam in Drosophila as an activator of Notch target genes like for example the genes of the E(spl)-C. Hence, Su(H)


acts in the same direction as Notch, Dl, mam and E(spl) in the Notch signalling pathway (ARTAVANISTSAKONAS et al. 1999). This view is supported by gainand loss of function phenotypes of Su(H) . Su(H) loss of function in germ line clones leads to a so called neurogenic phenotype which is characterized by a hypertrophic nervous system and loss of epidermis (LECOURTOIS and SCHWEISGUTH 1995). This is the same phenotype that typifies mutations in the so called ‘‘neurogenic loci’’ Notch , Dl , E(spl) , mam, neuralized (neur) and big brain (bib ) (POULSON 1937; LEHMANN et al. 1983; VA¨SSIN et al. 1985). Ectopic expression of Su(H) during pupal development results in a loss of bristles, a phenotype which is also observed for example after ectopic expression of E(spl) (SCHWEISGUTH and POSAKONY 1994; NAGEL et al. 2000; MOREL et al. 2001). Since Hairless gainand loss of function cause exactly the opposite phenotypes, a threshold model was proposed (BANG et al. 1995). According to this model, Hairless blocks Su(H) and only an excess of activated Notch is able to overcome Hairless activity. This excess of activated Notch eventually starts the signalling process and specifies the sensory organ precursors in the process of lateral inhibition. This model was questioned since loss of Hairless leads to an activation of Notch target genes also in tissues devoid of Notch activity (KLEIN et al. 2000). In accordance, we found that Su(H) concentrations in the nucleus depend on Hairless: nuclear Su(H) is less abundant in cells mutant for Hairless, however, is enriched when Hairless is overexpressed (MAIER et al. 1999). The threshold model implies that Hairless keeps Su(H) in the cytoplasm. In fact, the threshold model had to be expanded due to recent findings showing that a Su(H)- H- protein complex binds on DNA and acts as transcriptional repressor (MOREL et al. 2001). Hairless acts as a corepressor of Su(H) The vertebrate Su(H) homologue RBP-Jk/CBF-1 has intrinsic repressor activity. It assembles a repressor complex together with corepressors e.g. the silencing mediator of retinoid and thyroid hormone receptors (SMRT), Ski-interacting protein (SKIP) or histone deacetylase 1 (HDAC1). Only by binding to the Notch intracellular domain is it transformed into an activator (LUBMAN et al. 2004). This was an unexpected functional difference as the vertebrate proteins are highly conserved and the Drosophila Su(H) had been characterized as transcriptional activator. However, Su(H) can act as repressor in Drosophila as well. This was irrevocably shown by MOREL et al. (2001), though the threshold model was strongly doubted by these data. Ectopic expression of Hairless causes opposite



phenotypes than ectopic expression of Su(H), namely additional bristles (MOREL et al. 2001). Since Hairless is an antagonist of the Notch signalling pathway one would expect that a combined overexpression of Su(H) and Hairless resulted in wild type looking flies, similar to those observed upon combined overexpression of Nintra and Hairless (SCHWEISGUTH and POSAKONY 1994; BAILEY and POSAKONY 1995; GO et al. 1998). However, the opposite phenotype is observed and emerging flies are extremely furry (MOREL et al. 2001). In other words, Su(H) operates together with Hairless as a super-repressor. Therefore it was a not surprising finding that DNA-bound Su(H) is able to bind to Hairless which may act there within a repressor complex (Fig. 3, MOREL et al. 2001). This finding may also explain the observation that down-regulation of Hairless activates Notch-target genes like vestigial (KLEIN et al. 2000; NAGEL et al. 2005). Lack of Hairless would result in the absence of the repressor complex and consequently

in the reduction of transcriptional repression (SCHWEISGUTH 2004; NAGEL et al. 2005). Hairless recruits two corepressors CtBP and Groucho The fact that Hairless has a conserved C-terminal Binding Protein consensus sequence led to the hypothesis that Hairless may recruit CtBP as a corepressor. Indeed, a Hairless construct missing the CtBP binding domain is unable to cause the super-additive bristle phenotypes when overexpressed together with Su(H). In addition, the protein interaction between Hairless and CtBP was demonstrated in vivo and in vitro (MOREL et al. 2001). Recently it was shown that Hairless recruits Groucho (Gro) as a second corepressor (BAROLO et al. 2002; NAGEL et al. 2005). Since Hairless is a fast evolving gene, orthologues in insects only have been identified yet. However, this high evolutionary rate was crucial to identify small conserved and therefore, most likely important protein domains. In

Dl

N

S4

intra

H Su(H) S4 Gro intra

mam

r

H Re pr es so

r ato tiv Ac

N

Su(H)

217

CtBP

?

X

Fig. 3. Hairless antagonizes the Notch signalling pathway. Su(H) may act inside the nucleus (dashed line) as activator together with Nintra and mam or as a repressor together with Hairless and the corepressors Gro and CtBP on hypothetical target genes (blue). This finding suggests that Su(H) functions as molecular switch: Su(H) exhibits activator function by interaction with the cleaved intracellular domain of Notch (Nintra) together with mam. This may lead to competition with Hairless for binding to Su(H). Hairless has repression activity also without the corepressors Gro and CtBP. Possibly additional corepressors are involved (?). Alternatively, according to the threshold model, Hairless binding to Su(H) in the cytoplasm and/or the nucleus maybe obstructive, thus antagonizing the activation by Notch. A negative regulator of Hairless is Pros26.4 (S4). Binding of Pros26.4 leads to the preferential degradation of Hairless (MU¨LLER et al. 2006), thus resulting in the activation of the Notch signalling pathway.

218

D. Maier

a comparison between the Hairless proteins of D. melanogaster, D. hydei and the protein predicted from the recently published A. gambiae database, a small conserved motif YSIxxLLG was identified with similarities to an ‘eh1’ type Gro-binding site (SMITH and JAYNES 1996; BAROLO et al. 2002). A direct binding between Gro and Hairless was demonstrated in pull down experiments (BAROLO et al. 2002). Co-immunoprecipitation experiments confirmed this interaction in vivo and revealed that Hairless binds CtBP and Gro at the same time (NAGEL et al. 2005). Furthermore, mutations of the corepressors Gro and CtBP repress Notch and enhance Hairless mutant phenotypes. The studies concentrated on mechano-sensory bristles in adult flies (BAROLO et al. 2002). Ectopic expression of Hairless results in additional bristles (BANG and POSAKONY 1992; MAIER et al. 1997). If Hairless constructs that lacked Gro or CtBP binding domains were overexpressed, the number of additional bristles is remarkably reduced (CASTRO et al. 2005; NAGEL et al. 2005). The effects of depleting components of the Su(H), Hairless, Gro and CtBP repressor complex were examined in RNA interference studies during the formation of the wing margin (NAGEL et al. 2005). As a target, the expression of the vestigial -lacZ reporter gene construct was monitored. Vestigial is a well defined target gene of Notch. The expression of vestigial is restricted by Notch signalling to the dorso-ventral boundary of wing discs (KIM et al. 1996). Reduction of one of the repressor complex components led to a transcriptional activation of the vestigial -lacZ reporter gene. Only CtBP RNAi did not cause any effect. Conclusively, Gro seems to be the more important corepressor in this context. Another interpretation is that RNAi was not sufficient to eliminate CtBP mRNA completely and left residual CtBP activity (NAGEL et al. 2005). This view is supported by the observation that ectopic expression of Hairless represses expression of vestigial completely only in the presence of CtBP. Hence, CtBP is required for boundary formation in the wing imaginal disc. In Hairless constructs missing the Gro or CtBP binding domain or both, reduced repression is found, however, a basal repression is still detected. These experiments demonstrated that on one hand both, Gro as well as CtBP, are necessary in the corepressor complex, but on the other hand, Hairless without the two retains repressive activity (NAGEL et al. 2005). Hairless needs both corepressors also during formation of sensory organ precursor cells in the process of lateral inhibition. In cell culture assays it was shown that Hairless is able to repress Notch activity up to 80%. This repression is dependent on Su(H), since Hairless constructs missing the SBD lost repressive activity.


However, Hairless constructs that were defective in either Gro or CtBP binding had a reduced repressive activity of about 50% that remained the same in the double mutation. These observations confirm the results of the ectopic expression experiments. Since Su(H) binding is sufficient for Hairless repressive activity, Hairless must have an additional, Gro and CtBP independent mode of inhibition (NAGEL et al. 2005). Involvement of Hairless in the crosstalk with other signalling pathways Several independent genetic screens were performed to identify new Hairless functions. First, Hairless was ectopically expressed under the control of the enhancer of the sevenless gene (sev-H). The result was a weak but robust rough eye phenotype. Using EMS, flies were mutagenised and screened for enhancers of this rough eye phenotype. A number of mutants affecting the Notch and the epidermal growth factor receptor (EGFR)-signalling pathway were identified. Interestingly, with rugose (rg), an A kinase anchor protein was identified, acting as molecular link of the EGFR and the Notch signalling pathways (SCHREIBER et al. 2002; SHAMLOULA et al. 2002; WECH and NAGEL 2005). Four new alleles of rg were isolated. The small rough eye phenotype of rg mutants is caused primarily through cell type specific apoptosis of cone cells (WECH and NAGEL 2005), which is rescued by Hairless mutations and enhanced by Dl mutations (SCHREIBER et al. 2002). Increased Notch activity rescued the rg eye phenotype almost completely suggesting an anti- apoptotic function of Notch (WECH and NAGEL 2005). The fact that Hairless acts in the opposite direction, supports the idea of a pro-apoptotic function of Hairless. An additional indication that Hairless is involved in the crosstalk with EGFR came from further genetic screens that identified Hairless as a genetic interaction partner of EGFR (MU¨LLER et al. 2005). Accordingly, the Hairless loss of bristle phenotype was enhanced if EGFR is reduced (PRICE et al. 1997). Moreover, Hairless mutants were identified in a screen for modifiers of a furrow stop phenotype linked to hedgehog-signalling (CHANUT et al. 2000). Second, a Su(H) independent function of Hairless was used to identify new Hairless interaction partners. This independent function was observed with a heat shock inducible Hairless construct missing the SBD (Fig. 2, C2). Its overexpression results in additional wing veins (see above, MAIER et al. 1997; JOHANNES and PREISS 2002). A large number of candidates was screened for modifiers which alter this Su(H) independent phenotype. Again a crosstalk between the


Hairless and the EGFR signalling pathways was found. It was shown that ectopic vein material depends on veinlet (rhomboid), an activator of EGFR signalling, since in a homozygous mutant veinlet1 (ve1 ) the formation of ectopic veins was completely suppressed. Recently, a large scale screen was performed for modifiers of a rough eye phenotype that is generated by overexpression of Hairless in the gmr pattern through increased apoptosis. A large number of additional interacting factors which are involved in growth, differentiation, and cell death were discovered (MU¨LLER et al. 2005). The multi-functional gene Hairless In the first decades of the last century, Hairless was identified as a gene that is necessary to establish the correct bristle number in Drosophila . At first glance, this may not be very spectacular. With the mounting experimental data from a number of laboratories, the Hairless gene became more and more exciting. First genetic results led to grouping it together with very interesting neurogenic loci: their mutation results in embryos packed with nerve cells at the expense of epidermis and of course, in embryonic lethality. Recent genetic and molecular studies showed that Hairless acts as an antagonist of the Notch signalling pathway. This pathway is necessary at essentially all steps of development in the fly as well as in humans. Whenever cell-cell communication takes place, Notch is involved and this is true for the embryo as well as for the imaginal tissue. New results suggest that Su(H) has a key position in this pathway in Drosophila . Together with Nintra and mam, Su(H) transcriptionally activates target genes of the Notch signalling pathway like e.g. the genes of the E(spl)-C. By displacing this activator complex, Hairless is able to assemble a repressor complex together with Su(H), Gro and CtBP. Hence, Su(H) functions as a molecular switch as shown in Fig. 3, though the molecular mechanism is largely unknown. It may involve binding competition or other mechanisms and is still subjected to further studies. Hairless has also Su(H) independent roles, which makes the functional analysis even more interesting. In genetic screens using misexpression of Hairless leading to reduced tissue most likely by apoptosis, new genetic interaction partners of Hairless were identified. These interacting factors are involved in many diverse cellular processes including differentiation, cell growth and apoptosis (MU¨LLER et al. 2005). Further analysis of these interaction partners will stimulate the discussion about Hairless and may bring Hairless into an adequate scientific focus. Why is this not already the case? There is the striking oddness that Hairless is


219

not found in vertebrates. To date numerous orthologues are known in insects, from flies, to bees, to beetles (unpubl data). Is there no need for a Hairless-like protein in vertebrates? It is conceivable that repressorcomplex assembly in vertebrates differs from insects because CBF-1/ RBP-Jk recruits corepressors itself (LUBMAN et al. 2004). However, the picture may simply be incomplete yet. Presumably the corepressor complexes that are assembled on CSL-type proteins are rather large and may even contain different components depending on the cellular context. Not unlikely, there may be a Hairless-like protein in mammals that directly contacts CBF-1/RBP-Jk and mediates binding to corepressors like the vertebrate homologues of Gro and/or CtBP. And plausibly, other corepressors will be found in Drosophila that complex with Su(H). A first start has been made with the skiphomologue Bx42 that was shown to be involved in the Notch signalling cascade (NEGERI et al. 2002). Hairless is central to the regulation of Notch signalling processes in Drosophila , and is hence essential for cell differentiation and tissue development in the fly. Moreover, many genetic interactions suggest that Hairless may be involved in linking Notch- and EGF-receptor signalling pathways. The fact that no Hairless homologue has been identified in mammals so far may explain why this interesting gene is largely ignored. However, once the fine details of negative regulation of the Notch signalling pathway have been worked out, we expect to find a functional homologue in vertebrates as well. The way is being paved by the model system Drosophila once again. Acknowledgements I am grateful for critical reading of the manuscript by Dr. Anette Preiss, Anja C. Nagel and Dorothee Kiefer. I apologize to all authors whose work I was not able to consider in this review article due to space limitations.

REFERENCES Artavanis-Tsakonas, S. 1997. Alagille syndrome-a notch up for the Notch receptor. Nature Genet. 16: 212 213. Artavanis-Tsakonas, S., Matsuno, K. and Fortini, M. E. 1995. Notch signaling. Science 268: 225 232. Artavanis-Tsakonas, S., Rand, M. D. and Lake, R. J. 1999. Notch signaling: cell fate control and signal integration in development. Science 284: 770 776. Ashburner, M. 1982. The genetics of a small autosomal region of Drosophila melanogaster containing the structural gene for alcohol dehydrogenase. III. Hypomorphic and hypermorphic mutations affecting the expression of Hairless. Genetics 101: 447 459. Bailey, A. M. and Posakony, J. W. 1995. Suppressor of Hairless directly activates transcription of the Enhancer of split complex genes in response to Notch receptor activity. Genes Dev. 21: 2609 2622.

220

D. Maier

Bang, A. G. and Posakony, J. W. 1992. The Drosophila gene Hairless encodes a novel basic protein that controls alternative cell fates in adult sensory organ development. Genes. Dev. 6: 1752 1769. Bang, A. G., Bailey, A. M. and Posakony, J. W. 1995. Hairless promotes stable commitment to the sensory organ precursor fate by negatively regulating the activity of Notch. Dev. Biol. 172: 479 494. Barolo, S., Stone, T., Bang, A. G. et al. 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. Barrick, D. and Kopan, R. 2006. The Notch transcription activation complex makes its move. Cell 124: 883 885. Bray, S. 1998. A Notch affair. Cell 93: 363 374. Bridges, C. B. and Morgan, T. H. 1923. Carnegie Institute of Washington, publ. no. 327 Brou, C., Logeat, F., Lecourtois, M. et al. 1994. Inhibition of the DNA-binding activity of Drosophila Suppressor of Hairless and its human homolog, KBF2/RBP-J kappa, by direct protein-protein interaction with Drosophila Hairless. Genes Dev. 8: 2491 2503. Carter, M., Kuhn, K. M. and Sarnow, P. 2000. In: Hersley, J. W. B., Mathews, M. B. and Sonenberg, N. (eds), Translational control. Cold Spring Harbor Lab. Press, pp. 615 635. Castro, B., Barolo, S., Bailey, A. M. et al. 2005. Lateral inhibition in proneural clusters: cis-regulatory logic and default repression by Suppressor of Hairless. Development 132: 3333 3344. Cavener, D. R. 1987. Comparison of the consensus sequence flanking translational start sites in Drosophila and vertebrates. Nucleic Acid Res. 15: 1353 1361. Chanut, F., Luk, A. and Heberlein, U. 2000. A screen for dominant modifiers of ro (Dom), a mutation that disrupts morphogenetic furrow progression in Drosophila , identifies Groucho and Hairless as regulators of atonal. Genetics 156: 1203 1217. Cornelis, S., Bruynooghe, Y., Denecker, G. et al. 2000. Identification and characterization of a novel cell cycleregulated internal ribosome entry site. Mol. Cell 5: 597 605. de Celis, J. F., de Celis, J., Ligoxygakis, P. et al. 1996. Functional relationship between Notch, Su(H) and the bHLH genes of the E(spl) complex: the E(spl) genes mediate only a subset of Notch activities during imaginal development. Development 122: 2719 2728. Ehrenfeld, E. and Semler, B. L. 1995. Anatomy of the poliovirus internal ribosome entry site. Curr. Top. Microbiol. Immunol. 203: 65 83. Furukawa, T., Kawaichi, M., Matsunami, N. et al. 1991. The Drosophila RBP-J kappa gene encodes the binding protein for the immunoglobulin J kappa recombination signal sequence. J. Biol. Chem. 266: 23334 23340. Gho, M., Bellaiche, Y. and Schweisguth, F. 1999. Revisiting the Drosophila microchaete lineage: a novel intrinsically asymmetric cell division generates a glial cell. Development 126: 3573 3584. Go, M. J., Eastman, D. S. and Artavanis-Tsakonas, S. 1998. Cell proliferation control by Notch signaling in Drosophila development. Development 125: 2031 2040. Greenwald, I. 1998. LIN-12/Notch signaling: lessons from worms and flies. Genes Dev. 12: 1751 1762.


Hartenstein, V. and Posakony, J. W. 1990. A dual function of the Notch gene in Drosophila sensillum development. Dev. Biol. 142: 13 30. Jemc, J. and Rebay, I. 2006. Characterization of the split ends-like gene spenito reveals functional antagonism between SPOC family members during Drosophila eye development. Genetics 173: 279 286. Johannes., B. and Preiss, A. 2002. Wing vein formation in Drosophila melanogaster : Hairless is involved in the cross-talk between Notch and EGF signaling pathways. Mech. Dev. 115: 3 14. Kim, J., Irvine, K. D. and Carroll, S. B. 1995. Cell recognition, signal induction and symmetrical gene activation at the dorsal-ventral boundary of the developing Drosophila wing. Cell 82: 795 802. Kim, J., Sebring, M., Esch, J. J. et al. 1996. Integration of positional signals and regulation of wing formation and identity by Drosophila vestigial gene. Nature 382: 133 138. Klein, T., Seugnet, L., Haenlin, M. et al. 2000. Two different activities of Suppressor of Hairless during wing development in Drosophila . Development 127: 3553 3566. Kopan, R. 1999. Good things must come to an end: how is Notch signaling turned off ? Sci. STKE 9: 1 3. Kuroda, K., Han, H., Tani, S. et al. 2003. Regulation of marginal zone B cell development by MINT, a suppressor of Notch/RBP-J signaling pathway. Immunity 18: 301 312. Le Borgne, R., Bardin, A. and Schweisguth, F. 2005. The roles of receptor and ligand endocytosis in regulating Notch signaling. Development 132: 1751 1762. Lecourtois, M. and Schweisguth, F. 1995. The neurogenic Suppressor of Hairless DNA binding protein mediates the transcriptional activation of the Enhancer of split complex genes triggered by Notch signalling. Genes Dev. 9: 2598 2608. Lehmann, R., Jimenez, F., Dietrich, U. et al. 1983. On the phenotype and development of mutants of early neurogenesis in Drosophila melanogaster. Wilhelm Roux’s Arch. 192: 62 74. Lewis, J. 1998. Notch signalling and the control of cell fate choices in vertebrates. Semin. Cell Dev. Biol. 9: 583 589. Lin, H. V., Doroquez, D. B., Cho, F. et al. 2003. Split ends is a tissue/promoter specific regulator of Wingless signaling. Development 130: 3125 3135. Lindsley, D. L. and Zimm, G. G. 1992. The genome of Drosophila melanogaster. Academic Press. Lubman, O. Y., Korolev, S. V. and Kopan, R. 2004. Anchoring Notch genetics and biochemistry: structural analysis of the ankyrin domain sheds light on existing data. Mol. Cell 13: 619 626. Maier, D., Stumm, G., Kuhn, K. et al. 1992. Hairless, a Drosophila gene involved in neural development, encodes a novel, serine rich protein. Mech. Dev. 38: 143 156. Maier, D., Marquart, J., Thompson-Fontaine, A. et al. 1997. In vivo structure-function analysis of Drosophila Hairless. Mech. Dev. 67: 97 106. Maier, D., Nagel, A. C., Johannes, B. et al. 1999. Subcellular localization of Hairless protein shows a major focus of activity in the nucleus. Mech. Dev. 89: 195 199. Maier, D., Nagel, A. C. and Preiss, A. 2002. Two isoforms of the Notch antagonist Hairless are produced by differential translational initiation. Proc. Natl Acad. Sci. USA 99: 15480 15485.


Marquart, J., Alexief-Damianof, C., Preiss, A. et al. 1999. Rapid divergence in the course of Drosophila evolution reveals structural important domains of the Notch antagonist Hairless. Dev. Genes. Evol. 209: 155 164. Micchelli, C. A., Rulifson, E. J. and Blair, S. S. 1997. The function and regulation of cut expression on the wing margin of Drosophila : Notch, Wingless and the dominant negative role for Delta and Serrate. Development 124: 1485 1495. Morel, V., Lecourtois, M., Massiani, O. et al. 2001. Transcriptional repression by Suppressor of Hairless involves the binding of a Hairless-dCtBP complex in Drosophila . Curr. Biol. 11: 789 792. Mu¨ller, D., Kugler, S.J., Preiss, A. et al. 2005. Genetic modifier screens on Hairless gain-of-function phenotypes reveal genes involved in cell differentiation, cell growth and apoptosis in Drosophila melanogaster. Genetics 171: 1137 1152. Mu¨ller, D., Nagel, A. C., Maier, D. et al. 2006. A molecular link between Hairless and Pros26.4, a member of the AAA- ATPase subunits of the proteasome 19S regulatory particle in Drosophila . J. Cell Sci. 119: 250 258. Mumm, J. S. and Kopan, R. 2000. Notch signaling: from the outside in. Dev. Biol. 228: 151 165. Nagel, A. C., Maier, D. and Preiss, A. 2000. Su(H)independent activity of Hairless during mechano-sensory organ formation in Drosophila . Mech. Dev. 94: 3 12. Nagel, A. C., Krejci, A., Tenin, G. et al. 2005. Hairlessmediated repression of Notch target genes requires the combined activity of Groucho and CtBP corepressors. Mol. Cell. Biol. 25: 10433 10441. Negeri, D., Eggert, H. and Saumweber, H. 2002. Inducible RNA interference uncover the Drosophila protein Bx42 as an essential nuclear cofactor involved in Notch signal transduction. Mech. Dev. 117: 151 162. Portin, P. 2002. General outlines of the molecular genetics of the Notch signalling pathway in Drosophila melanogaster: a review. Hereditas 136: 89 96.


221

Poulson, D. F. 1937. Chromosomal deficiencies and the embryonic development of Drosophila melanogaster. Proc. Natl Acad. Sci. USA 23: 133 137. Price, J. V., Savenye, E. D., Lum, D. et al. 1997. Dominant enhancers of Egfr in Drosophila melanogaster : genetic links between the Notch and Egfr signaling pathways. Genetics 147: 1139 1153. Schreiber, S. L., Preiss, A., Nagel, A. C. et al. 2002. Genetic screen for modifiers of the rough eye phenotype resulting from overexpression of the Notch antagonist Hairless in Drosophila . Genesis 33: 141 152. Schweisguth, F. 2004. Notch signaling activity. Curr. Biol. 14: R129 138. Schweisguth, F. and Posakony, J. W. 1994. Antagonistic activities of Suppressor of Hairless and Hairless control alternative cell fates in the Drosophila adult epidermis. Development 120: 1433 1441. Shamloula, H. K., Mboqho, M. P., Pimentel, A. C. et al. 2002. rugose (rg) a Drosophila A kinase anchor protein, is required for retinal pattern formation and interacts genetically with multiple signaling pathways. Genetics 161: 693 710. Simpson, P. 1997. Notch signaling in development: on equivalence groups and asymmetric developmental potential. Curr. Opin. Genet. Dev. 7: 537 542. Smith, S. T. and Jaynes, J. B. 1996. A conserved region of engrailed, shared among all en-, gsc-, Nk1-, Nk2- and msh-class homeoproteins, mediates active transcriptional repression in vivo. Development 122: 3141 3150. Strausberg, R. L., Feingold, E. A., Grouse, L. H. et al. 2002. Generation and initial analysis of more than 15,000 fulllength human and mouse cDNA sequences. Proc. Natl Acad. Sci. USA 99: 16899 16903. Va¨ssin, H., Vielmetter, J. and Campos-Ortega, J. A. 1985. Genetic interactions in early neurogenesis of Drosophila melanogaster. J. Neurogenet. 2: 291 308. Wech, I. and Nagel, A. C. 2005. Mutations in rugose promote cell type-specific apoptosis in the Drosophila eye. Cell Death Differ. 12: 145 152.