Gene name - hairy
Cytological map position - 66D15
Function - transcription factor
Keywords - pair-rule and patterning
Symbol - h
Genetic map position - 3-26.5
Classification - basic HLH
Cellular location - nuclear
hairy encodes a basic helix-loop-helix protein that functions in at least two distinct stages during Drosophila development: 1) embryogenesis, when it participates as a primary pair-rule gene in the establishment of segments, and 2) the larval stage, when it functions negatively in determining the pattern of sensory bristles on the adult fly. The genes of the Hairy and Enhancer of split HES family encode bHLH transcription factors that are most closely related to two groups of negative regulators of neurogenesis in Drosophila, hairy and the products of the Enhancer of Split complex (Espl). Both hairy and Espl products have been shown to directly repress transcription of the proneural gene achaete, but their activity is required in different contexts. hairy is a prepattern gene. It is required in large areas of the wing and leg imaginal discs to prevent ectopic expression of the proneural gene achaete and the formation of ectopic bristles. The genes of the Espl complex are neurogenic genes that are activated by Notch signalling in a process of lateral inhibition during embryonic and adult neurogenesis. Activation of the Espl genes blocks the accumulation of high amounts of proneural protein in most cells of the proneural clusters, thereby preventing them from adopting a neural fate.
The spatial organization of epidermal structures in the embryo and adult fly constitutes a classicically modelled system for understanding how the two dimensional arrangement of particular cell types is generated. Adult legs are covered with sensory bristles, arranged in longitudinal rows in most segments. Two regulatory genes, hairy and achaete are chiefly responsible for setting up this periodic bristle pattern. achaete is expressed during pupal leg development in a dynamic pattern that develops into narrow longitudinal stripes, 3-4 cells in width, each of which represents a field of cells (proneural field) from which bristle precursor cells are selected. This pattern of gene expression foreshadows the adult bristle pattern established in part through the function of the hairy gene.
In pupal legs, hairy is expressed in four longitudinal stripes, located between every other pair of achaete stripes. In the absence of hairy function, achaete expression expands into the interstripe regions that normally express hairy, fusing the two achaete stripes and resulting in extra-wide stripes of achaete expression. This misexpression of achaete alters the fields of potential bristle precursor cells. This leads to the misalignment of bristle rows in the adult. hairy's role in patterning achaete expression is distinct from that in the wing, in which Hairy suppresses late expression of achaete but has no effect on the initial patterning of achaete expression. Thus, the leg bristle pattern is apparently regulated at two levels: a global regulation of hairy and achaete expression patterns partitioning the leg epidermis into striped zones, and later, a local regulation that involves refinement steps that may control the alignment and spacing of bristle cells within these zones (Orenic, 1993).
What initially regulates the hairy boundaries? The leg is developed from a single parasegment, but the leg has a total of eight proneural stripes. Between the parasegment and the leg there have to be one or more levels of regulation adding additional fields of development, in order to form eight stripes. The pattern must have its origin in the embryo, where a combination of pair-rule and segment polarity genes establish the identity of each cell along the anterior-posterior axis.
Thus, hairy is expressed in a spatially restricted manner in the leg imaginal disc and functions to position adult leg bristle rows by negatively regulating the proneural gene achaete, which specifies sensory cell fates. While much is known about the events that partition the leg imaginal disc and about sensory cell differentiation, the mechanisms that refine early patterning events to the level of individual cell fate specification are not well understood. In the third instar leg imaginal disc, h is expressed along both the D/V and A/P axes. D/V axis expression appears as a single stripe in the anterior compartment of the disc immediately adjacent to the A/P compartment boundary. A/P axis expression appears as two wedge-shaped blocks in the distal leg segments on either side of the A/P compartment boundary. After disc eversion, the D/V axis stripe forms two of the four longitudinal leg stripes. The A/P axis expression forms five circumferential stripes at the first through fifth tarsal segments. The remaining two longitudinal stripes will be positioned along the A/P axis, intersecting the D/V axis stripes at the distal tip of the leg. These stripes do not appear until 2-3 hours after puparium formation (APF). This paper concerns itself with Hairy expression in the D/V axis (Hays, 1999).
The expression pattern of the D/V axis h stripe (D/V-h) is highly reminiscent of the patterning of known Hh target genes; therefore, constituents of the Hh signal transduction pathway were examined as potential regulators of the stripe. Hh is secreted from the posterior compartment of imaginal discs and influences gene expression in anterior compartment cells. Ci, a zinc finger transcription factor, is expressed throughout the anterior compartment and is thought to mediate Hh signaling by direct transcriptional regulation of Hh target genes. The region of high-level, full-length Ci overlaps with D/V-h, making Ci a strong candidate for regulation of this h stripe. The influence of Hh on h expression was investigated by generating somatic clones which lack functional Smoothened (Smo). Within smo clones, both dorsally and ventrally, D/V-h expression is lost in a cell autonomous manner, implicating Hh signaling in D/V-h regulation. This analysis was extended to Ci, the only known transcriptional mediator of Hh signaling, by assaying the expression of h in legs misexpressing full-length Ci. In the leg disc, the 30A-GAL4 insertion is expressed dorsally in cells of the presumptive femur, in the distal tip of the leg, and in a central ring corresponding to the fifth tarsal segment. Misexpression of a ci transgene under control of the 30A-GAL4 driver results in ectopic expression of h throughout the 30A expression domain, suggesting that ci can serve as a positive regulator of h (Hays, 1999).
In order to independently assay the response of D/V-h to Ci, the leg-specific enhancer elements which govern the expression of D/V-h were isolated and cloned into a lacZ reporter vector. A 9 kb genomic fragment 30-40 kb 3' of the h transcription start site contains sequences that direct beta-gal expression in the D/V-h stripe domain. This reporter construct is referred to as D/V-h- lacZ. The D/V-h-lacZ stripe lies in the anterior compartment of the disc adjacent to the compartment boundary and superimposes with endogenous H protein. As with endogenous D/V-h, D/V-h-lacZ expression is lost in smo mutant clones, suggesting that this enhancer element is a target of Hh signaling. Misexpression of ci with the 30A-GAL4 driver results in ectopic expression of D/V-h-lacZ, throughout the 30A domain, though the pattern differs somewhat from that seen with endogenous H. This confirms the specific activation of D/V-h expression by exogenously supplied Ci. In addition to ectopic activation by Ci, a dominant-negative form of ci impedes expression of D/V-h (Hays, 1999).
The D/V-h enhancer sequences are separable into discrete dorsal and ventral components, referred to as D-h and V-h, respectively. Both of these elements are responsive to Ci. The separability of these enhancers suggests that D/V-h expression is regulated by dorsal- and ventral-specific factors, such as Dpp and Wg, and not by ci alone. Dpp and Wg are known to specify dorsal and ventral leg fates, respectively, and have been shown to antagonize each otherís function to maintain dorsal and ventral leg territories. In the absence of one signaling molecule, expression of the other and its target genes expands, resulting in the duplication of dorsal or ventral leg structures (Hays, 1999).
To assess the roles of Dpp and Wg in the regulation of the D-h and V-h enhancer elements, the expression of D-h-lacZ and V-h-lacZ was assayed in legs that were mutant for either dpp or wg. Expression from D-h-lacZ is reduced in leg discs that are mutant for dpp, and is expanded to produce a full D/V axis stripe in discs that are mutant for wg. Conversely, V-h-lacZ expression is severely reduced in wg mutant discs, and duplicated in dpp mutant discs. These findings are in keeping with what has been demonstrated for other Dpp and Wg target genes and suggest that the D-h and V-h enhancers are targets of Dpp and Wg signaling, respectively. The roles of Dpp and Wg in D/V-h regulation were further examined by making somatic clones lacking components of the Dpp and Wg signaling pathways. Given the antagonism between Dpp and Wg in the leg, it was necessary to analyze clones mutant for a component of each pathway. D-h expression was examined in clones mutant for wg and Mothers against dpp (Mad). Mad is a downstream effector in the Dpp signaling pathway that has been shown to bind DNA and transcriptionally regulate some Dpp target genes directly. Dorsal Mad;wg clones that intersect the D-h stripe show loss of h expression, except for variable low level expression in a single row of cells immediately adjacent to the A/P boundary. It can reasonably be concluded that loss of D-h results from the loss of Mad, since there is no duplicated Wg in these clones. These results not only support the finding that D-h is a target of Dpp signaling, it identifies Mad as a potential transcriptional regulator of D-h. Thus it is proposed that D/V-h expression is regulated in a non-linear pathway in which Ci plays a dual role. In addition to serving as an upstream activator of Dpp and Wg, Ci acts combinatorially with them to activate D/V-h expression (Hays, 1999).
Several questions remain open regarding the placement of leg sensory organs. There is much to be learned about the regulation of the eight Ac leg stripes. In the absence of H function, ac is expressed in four broad stripes, which may represent four zones that are competent to express ac. The expression of ac in these zones could be under the control of independent cis-regulatory elements as are D-h and V-h. The D/V axis stripes would likely be regulated by high level Dpp and Wg signaling. The A/P axis stripes could be sensitive to low levels of Dpp and Wg signaling. This may also be the case for the A/P-h stripes, whose regulation is also not yet understood. Alternatively, the whole leg may be competent to express ac with the interstripes established by H and an unknown factor that represses the expression of ac in the four non-H-expressing interstripe regions (Hays, 1999).
The precellular Drosophila embryo contains approximately 10 well characterized transcriptional repressors. At least half are short-range repressors that must bind within 100 bp of either upstream activators or the core transcription complex to inhibit (or quench) gene expression. The two long-range repressors can function over distances of 1 kilobase or more to silence transcription. Previous studies have shown that three of the five short-range repressors interact with a common corepressor protein, dCtBP. In contrast, the two long-range repressors, Hairy and Dorsal, recruit a different corepressor protein, Groucho. Hairy also was shown to interact with dCtBP, thereby raising the possibility that Groucho and dCtBP are components of a common corepressor complex. To investigate this issue, wild-type and mutant forms of Hairy were misexpressed in transgenic embryos. Evidence is presented that Hairy-mediated repression depends on the Groucho interaction sequence (WRPW) but not the weak dCtBP motif (PLSLV) present in the native protein. Conversion of the PLSLV motif into an optimal dCtBP interaction sequence (PLDLS) disrupts the activity of an otherwise normal Hairy protein. These results suggest that dCtBP and Groucho mediate separate pathways of transcriptional repression and that the two proteins can inhibit one another when both bind the same repressor (Zhang, 1999).
The removal of the weak dCtBP interaction motif (PLSLV) does not impair Hairy-mediated repression of Sex lethal, forkhead, huckebein, and tailless. If anything, removal of this motif augments Hairy function. This observation suggests that the binding of dCtBP somehow interferes with Groucho-mediated repression. Additional support for this view stems from the observation that the PLDLS/WRPW protein, which contains an optimal dCtBP motif, is inactive and fails to repress any of the target genes that were examined. The simplest interpretation of these results is that the dCtBP and Groucho corepressors interfere with one another when both are bound to Hairy. Such antagonistic interactions are supported by previous genetic studies, which suggest that lowering the dose of maternal dCtBP products can partially suppress the embryonic phenotypes of hairy mutants. The P-SLV-K and WRPW motifs are separated by just nine amino acid residues within the C terminus of the Hairy protein. When dCtBP and Groucho both bind, they might be unable to interact with additional corepressors or with their target proteins in the core transcription complex (Zhang, 1999).
Bases in 5' UTR - There are two hairy transcripts that differ in the length of the 5' end (Rushlow, 1989).
Exons - three
Bases in 3' UTR - 828
Amino Acids - 337
Hairy possesses a bHLH domain (Rushlow, 1989).
date revised: 10 August 97
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.