BIOLOGICAL OVERVIEW

The E(spl) locus, located at 96F8-14, harbors 12 transcription units: malpha, mbeta, mdelta, mgamma, and m1, m2, m3, m4, m5, m6, m7 and m8. Molecular analysis has shown that seven of these genes, namely mbeta, mdelta, mgamma, m3, m5, m7 and m8, encode highly related basic-helix-loop-helix (bHLH) proteins, and are turned on in response to Notch signaling, via the transcriptional mediator Su(H). Since E(spl) bHLH proteins are able to act as transcriptional repressors of proneural genes, they probably act as endpoint effectors in the suppression of neural fate mediated by Notch. The pattern of expression of the E(spl)bHLH genes agrees with this model, since it closely matches the domains of neurogenesis, both in the embryo and the larva. The roles of the non-bHLH genes of the E(spl) region have not been fully characterized to date. However, at least one of them, m4, has been long known to have a similar expression pattern to that of E(spl) bHLH genes and appears to also be inducible by Notch signaling in a Suppressor of Hairless-dependent manner (Knust, 1987; Singson, 1994; Bailey, 1995). The whole locus has been scanned for genes showing transcriptional response to activated Notch. In addition to showing that all seven bHLH genes can be turned on by ectopic expression of a constitutively active form of Notch, an identical behaviour has been documented for some of the non-bHLH genes. Could these proteins act in conjunction with the E(spl) bHLH factors to oppose neurogenesis (Wurmbach, 1999).

Among the non-bHLH Notch-inducible genes of the E(spl) Complex are m4 and malpha. The embryonic expression pattern of both m4 and malpha is remarkably similar to that of E(spl)bHLH. In imaginal disks they both show strong expression in areas where peripheral neural elements arise, e.g. the macrochaete proneural clusters of the late larval wing disk, with malpha displaying additional expression at other sites (Wurmbach, 1999). Interestingly, Bearded (Brd) (Leviten, 1997), which maps outside the E(spl) locus, encodes a product with weak structural similarity to the N-terminal half of m4 (31% identical; 48% similar) and malpha (25% identical/37% similar). While it is not known whether Brd is transcriptionally activated by Notch signaling, it has been shown to participate in neural fate acquisition. The products encoded by these genes display significant sequence similarity to each other (44% identity, 66% similarity) (Wurmbach, 1999). Apparent paralogs of E(spl)m4 have been named Bob ( Brother of Bearded) and Tom (Twin of m4), respectively. The Bearded (Brd) family proteins can be classified as Brd-like (Brd and Bob) or m4-like (m4, malpha, and Tom), based on their relative sizes and degree of amino acid similarity (For information about Bob and Tom see the m4 Evolutionary Homologs section). Hypermorphic Brd alleles produce supernumerary external sensory bristles in the adult and display genetic interactions which argue for a model where Brd counteracts the activity of the Notch pathway in sensory organ specification. If m4 and malpha have a function similar to Brd, they should promote neural fate: this is contrary to expectation for a Notch-driven class of genes, which oppose neural fates. Unlike their neighboring E(spl)bHLH genes, overexpression of either m4 or malpha appears to counteract Notch signaling; the overexpression produces supernumerary chaetae, a phenotype similar to that of Brd gain-of-function (gof) alleles. These arise from suppression of the Notch-mediated lateral inhibition process, because their progenitor cells (the sensory organ precursors, or SOPs) are seen in close apposition to one another in imaginal disc epithelia. The possible mechanism for the function of these proteins has been analyzed, as well as their relationships to various components of the Notch signaling pathway (Apidianakis, 1999).

Intercellular signaling mediated by Notch proteins is crucial to many cell fate decisions in metazoans. Its profound effects on cell fate and proliferation require that a complex set of responses involving positive and negative signal transducers be orchestrated around each instance of signaling. In Drosophila the basic-helix-loop-helix (bHLH) repressor encoding genes of the E(spl) locus are induced by Notch signaling and mediate some of its effects, such as suppression of neural fate. A novel family of Notch responsive genes, whose products appear to act as antagonists of the Notch signal in the process of adult sensory organ precursor singularization, has been described. They, too, reside in the E(spl) locus and comprise transcription units E(spl) m4 and E(spl) malpha. Overexpression of these genes causes downregulation of E(spl) bHLH expression accompanied by cell autonomous overcommitment of sensory organ precursors and tufting of bristles. Interestingly, negative regulation of the Notch pathway by overexpression of E(spl) m4 and malpha is specific to the process of sensory organ precursor singularization and does not impinge on other instances of Notch signaling (Apidianakis, 1999).

Experiments by Nagel (2000), suggest that the overexpression phenotype of E(spl) m4 and E(spl) malpha obtained by Apidianakis (1999) is likely to be due to a dominant negative effect and does not reflect the biological function of these two genes. In order to elucidate m4/malpha gene function directly, RNAi, which causes sequence-specific transcript degradation, was carried out by injecting either m4 or malpha double-stranded RNA or a mixture of both into pre-blastoderm embryos. In agreement with genetic data, RNAi causes a high incidence of lethality (~50%). Dead embryos develop intermediate to strong neurogenic phenotypes (too many neurons) typical of loss of E(spl) bHLH activity. Surviving embryos hatch into wild type appearing larvae that develop normally to adult flies. From this it is concluded that the m4/malpha genes are required to positively transduce the Notch signal during neurogenesis, and presumably during bristle development as well. Therefore, suppression of lateral inhibition observed after overexpression of either m4/malpha family member must be due to a dominant-negative effect, presumably by titrating out other important Notch pathway components (Nagel, 2000).

A third member of the m4/malpha gene family, bbu/tom, has been identified and has been localized within the Brd-C (see Bearded). It is of interest to know whether bbu/tom has a function during Drosophila development similar to m4/malpha. Overexpression of bbu/tom causes bristle tufting phenotypically indistinguishable from m4/malpha. The molecular mechanism underlying bristle tufting is also based on interference with E(spl) gene activity in a similar manner as occurs for m4/malpha. Ectopic expression of either bbu/tom, m4 or malpha results in the selection of additional SOPs due to silencing of, for example, E(spl) m8 transcription. Repression is specific to those E(spl) genes that are involved in bristle specification and does not include mbeta, which acts mainly in wing vein development. Apparently, bbu/tom overexpression has identical effects on the development of mechanosensory bristles as that of m4/malpha. It was thus of interest to see whether bbu/tom is involved in neurogenesis in the same way, and this was again analyzed by dsRNA injections. Lethality was induced at a similar rate, but neurogenic embryos were much rarer. Instead, a large fraction of the dead embryos did not develop to a stage where they secrete cuticles. The most conspicuous phenotypes of the older embryos were defects during head involution. Thus, bbu/tom is expected to play a role different from m4/malpha during Drosophila embryonic development (Nagel, 2000).

The fact that the activity of bbu/tom differs clearly from either m4 or malpha might be surprising because all three genes are robustly expressed in the neuroectoderm in germ band extended embryos. It is noted, however, that this is the major region of co-expression of the three genes, which show distinct expression patterns during earlier and later stages of embryogenesis. Up to the blastoderm stage, bbu/tom transcripts accumulate in the presumptive ectoderm and are excluded from the presumptive mesoderm. In contrast, m4/malpha are exclusively expressed in the mesectoderm at that stage. Starting at germ band retraction, bbu/tom mRNA is enriched along the midline and in the segmental folds within the lateral ectoderm. The pattern is also quite distinct in wing and leg imaginal discs, whereas in the eye disc, expression bordering the morphogenetic furrow is reminiscent of m4/malpha. No induction of bbu/tom transcription is observed by ectopic expression of the activated Notch receptor in a pair rule pattern in the embryo (prd-Gal4 >>UAS-N^intra). Furthermore, transcriptional activation of bbu/tom is not dependent on Su(H) because Su(H) mutant embryos derived from germ line clones still express bbu/tom normally in the lateral ectoderm. The absence of midline staining is attributed to the hypertrophic nervous system in these embryos (Nagel, 2000).

What is the physiological role of the m4/malpha family? The m4/malpha family contains three members of high structural similarity and up to eight, taking the related Brd gene family into account. Overexpression of various members gives surprisingly similar phenotypes -- ectopic bristles (mechanosensory organs) that result from a failure of the lateral specification of singular bristle precursor cells (SOPs). From these gain of function phenotypes, it has been concluded that this class of gene plays a negative role in Notch-mediated signaling processes. A lack of phenotypes in loss of function mutations of single genes gives little genetic support to this hypothesis and favors functional redundancy between the individual genes (Nagel, 2000).

Despite the similarity between these genes, there is a major difference between the members of the m4/malpha and Brd gene families, in that only m4 and malpha are direct transcriptional targets of Notch signaling whereas genes in the Brd-C are not. De-repression of bbu/tom transcription within the ventral ectoderm after driving Notch with a Kr-Gal4 line appears to conflict with the lack of transcriptional activation by Notch that has been observed. One possibility to resolve this conflict is to note that Kr-Gal4 is expressed earlier than prd-Gal4. Since Notch is notorious for its ability to affect cell fate at many stages, it is not unlikely that the earlier ectopic expression might cause a cell fate change of the ventral ectoderm to lateral ectoderm, where bbu/tom is strongly expressed. Put more parsimoniously, the ectopic activation of bbu/tom expression is most likely a secondary indirect effect, since it is seen when N^intra is driven by one Gal4 line and not by another. Because the transcriptional regulation of these genes is different, it is quite likely that their physiological roles might also be different. The RNAi experiments now reveal a role for m4/malpha as positive factors in the Notch pathway during neurogenesis in the early embryo, whereas bbu/tom appears to function differently. This interpretation is corroborated by genetic evidence using a combination of small deletions in the E(spl)-C that result in failure of lateral inhibition during neurogenesis as well as bristle development. Whereas deficiencies in m4 increase bristle density, those in Brd and bbu/tom do not, even in the background of E(spl)-C deficiencies. Hence, despite the structural similarities between members of the m4/malpha family, bbu/tom has lost a number of their characteristics, most notably Notch responsiveness in agreement with its remote chromosomal position. The phenotypes achieved by overexpression of any of the three genes, however, are very similar. They are likely a result of dominant-negative effects and thus do not allow conclusions to be drawn on a role for this gene family in normal bristle development. They rather indicate an ability of this class of proteins to interact with other factors specifically required during the process of bristle precursor specification (Nagel, 2000).

GENE STRUCTURE

Exons - 1

Bases in 3' UTR - 165

PROTEIN STRUCTURE

Amino Acids - 152

Structural Domains

The five genes described in this paper are arrayed between mbeta and m7, both coding for bHLH proteins. Two other bHLH genes, m3 and m5 are intermingled with the five. Bearded and M4 are 16% identical. Furthermore, in transcripts of both Brd and m4 (Knust, 1997 and Klambt, 1989) there are three common regulatory sequence motifs within the 3' UTR. These are known as the 'Brd box', the 'GY box' and the 'K box'. As in m4, the sequence motif of the Brd box is found twice in the 3'-UTR of malpha mRNA at similar positions but without a GY box. None of the other four non-bHLH E(spl)-C genes contains either a Brd or a GY box. The K box appears to be more common. It is found twice in the 3'-UTR of malpha and once each in the 3' UTRs of m2 and m6 (Wurmbach, 1999).

Gain-of-function alleles of the Drosophila gene Bearded (Brd) cause sensory organ multiplication and loss phenotypes indistinguishable at the cellular level from those caused by loss-of-function mutations in the genes of the Notch pathway. Molecular analysis has been carried out on the structure and expression of both wild-type and mutant Brd transcription units. Brd transcript is truncated and accumulates to substantially higher levels in the gain-of-function mutants, due to the insertion of a transposable element of the blood family in the Brd 3' untranslated region (UTR). The wild-type Brd 3' UTR includes three copies of a 9-nucleotide sequence (CAGCTTTAA) that is referred to as the 'Brd box'. Moreover, the 3' UTRs of Brd and of the m4 transcription unit of the Enhancer of split gene complex [E(spl)-C] exhibit an unusually high degree of sequence identity that includes not only Brd box sequences but also a second motif referred to as the 'GY box' (GTCTTCC). Both the Brd box and the GY box are also present in the 3' UTRs of several basic helix-loop-helix repressor-encoding genes of the E(spl)-C, often in multiple copies, suggesting that a novel mode of post-transcriptional regulation applies to Brd and many E(spl)-C genes. The fact that the more abundant Brd mutant mRNA lacks the GY box and two of the Brd boxes present in wild-type Brd mRNA suggests that either or both of these elements may confer instability on transcripts that contain them. Brd encodes a novel small protein of only 81 amino acids that is predicted to include a basic amphipathic alpha-helix. The deduced Brd protein shows sequence similarity to the E(spl)m4 protein, which is likewise expected to include a basic amphipathic alpha-helix, suggesting that the two proteins have related biochemical functions (Leviten, 1997).

date revised: 5 January 2001

Home page: The Interactive Fly © 1995, 1996 Thomas B. Brody, Ph.D.

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