BIOLOGICAL OVERVIEW

Vestigial is a novel nuclear protein with no known homologs, except for an N-terminal domain resembling that of Paired. vestigial expression is evident in thoracic and abdominal segments, in the embryonic primordia of the wing and haltere discs, in discrete cells in the ventral nerve cord, and possibly in progenitors of sense organs of the peripheral nervous system (Williams, 1991). There are no phenotypes associated with vestigial mutation, either in segmentation or neural development, but mutant flies manifest a complete lack of balance organs of the third thoracic segment, as well as a lack of development of wings and halteres.

Cells lacking vestigial undergo extensive cell death in the presumptive wing region of the third-larval instar imaginal discs. This results in the complete elimination of wing structures in adults. Several observations link vestigial function with cell survival and proliferation. Combining vestigial mutation with mutations that cause cell hypertrophy (too many cells) results in a partial reversal of the vestigial phenotype. Mutations in Drosophila tumor suppressor genes giant discs (lgd) and fat (ft) cause epithelial hypertrophy in all imaginal discs. Combining either lgd or ft mutations with vg increases the size of the wing disc and partially restores the bristle pattern (Agrawal, 1995).

vestigial has also been linked to regulation of cell proliferation. Treatment of flies with drugs that inhibit the enzyme dihydrofolate reductase gives rise to a defect that resembles vestigial mutation: nicks in the wings of wild-type flies and a strong vg phenotype in flies heterozygous for a deficiency of the vg locus. The effect of these drugs is to reduce the level of vestigial transcription, suggesting that the level of vestigial transcription is regulated by metabolites of dihydrofolate reductase. Specific mutations in the vestigial locus result in a diminished response to these foliate inhibitors, restoring normal development. These mutations do not knock out vestigial but result in an altered protein (Zider, 1996). The nicking phenomenon induced by drug treatment and its reversal by genetic modifications in vestigial, suggest a role for vestigial in cell proliferation. Thus future research in vestigial biology may reveal an interesting interaction with genes involved in regulation of cell cycle.

The wing margin represents a site of complex developmental interactions resulting in formation of a boundary separating dorsal and ventral cells and induction of cell proliferation generating the familiar adult structure (see the segment polarity and fringe sites for further discussion of events at the wing margin). An intronic enhancer, located between the first and second exons of vestigial, is responsible for vestigial transcription at the wing margin, assuring the presence of vestigial at the margin. Transcription at the margin requires an interaction between Notch and its ligands Serrate and Delta, and appears to be downstream of wingless (Blair, 1994 and Doherty, 1996).

Is vestigial regulated directly by Notch signaling or only via an intervening wingless activity? Notch-dependent activation of wg, cut and vestigial at the wing margin depends on the activity of Suppressor of Hairless. Su(H)-mutant cells lose expression of the vestigial early enhancer, of wingless and of cut in a cell autonomous manner. Clones of Su(H)-mutant cells cause loss of wing tissue and scalloping of the wing, but only in Notch mutant clones at the D/V boundary. vestigial expression at the D/V boundary does not depend on wingless, since misexpression of wild-type wg cDNA, which results in wing margin bristles, does not cause an expansion of vestigial expression. Likewise, wingless expression does not depend on an early function of vestigial. Both Notch and wingless cooperate to activate cut at the D/V boundary. Later expression of vestigial in the wing pouch is, however, wingless dependent. vestigial is expressed in a broad domain throughout the wing. Removing of Wg activity in late second instar larvae leads to almost complete loss of the secondary expression of vestigial in the wing pouch without affecting expression at the D/V boundary. Taken together with the observation that clones of cells lacking shaggy activity show a cell-autonomous increase of Vestigial expression, these results suggest the vestigial is a direct target of the Wg pathway (Neumann, 1996).

A small number of major regulatory (selector) genes have been identified in animals that control the development of particular organs or complex structures. In Drosophila, the vestigial gene is required for wing formation and is able to induce wing-like outgrowths on other structures. Because ectopic expression of Vg in many imaginal discs induces the outgrowth of wing tissue, the expression of various wing patterning genes was examined to see if they are induced in ectopic growths. Vg is expressed in the entire developing wing pouch whereas Spalt and Serum response factor (SRF) have specific expression patterns within this domain but are not expressed in wild-type leg discs. Targeted expression of Vg with the Gal4-UAS system induces ectopic expression of Sal and SRF in developing leg imaginal discs. Similarly, the nubbin (nub) gene (which is also expressed and required during wing development ) is ectopically induced in leg discs by Vg expression. In each case, only a subset of the cells expressing Vg activate the target gene, which suggests that additional factors control the expression pattern of each gene. In a first step toward elucidating the molecular mechanism by which Vg regulates gene expression, the response of wing-specific enhancers to ectopic Vg expression was examined. Attention was focused on both the boundary and quadrant enhancers of the vg gene and the enhancer from the SRF gene that drives expression specifically in the intervein region between veins three and four. All three enhancers are induced by ectopic Vg expression in leg and other imaginal discs. Importantly, ectopic expression of Vg in clones of cells induces the enhancers only within the clones. However, gene expression is not induced in all cells within clones nor in all clones. In addition, each individual enhancer is expressed in different regions of these discs that appear to correlate with the spatial distribution of the different signaling inputs known to be required for activation of these enhancers (Halder, 1998).

Scalloped is required for Vg function. In the notum primordia of the wing disc, the vg enhancers, as well as the sal, SRF, and nub genes are not induced by ectopic Vg even though the known required extracellular signals are present. Target gene activation could depend then on the function of another gene(s). One candidate for such a factor is the product of the sd gene, which is expressed in a pattern similar to Vg in the wing disc and is required for wing formation and the proper expression of Vg and other genes. In other discs, such as the leg and eye discs, sd is endogenously expressed and is upregulated wherever ectopic Vg is able to induce wing-specific gene expression and trigger wing development. It is noted, however, that a sd enhancer trap line and the SRF-intervein C enhancer transgene are also ectopically induced by Vg in the presumptive notum, although at levels lower than those observed in the wing pouch. This is consistent with the inability of Vg to trigger wing development and induce other wing patterning genes in the developing notum. Indeed, mis-expression studies show that Sd function is required in parallel with Vg in order for Vg to exert its wing inducing activity. The three wing-specific enhancers from the SRF and vg genes are activated synergistically when Sd and Vg are coexpressed in Drosophila S2 cells. Although each individual protein has some effect on reporter gene expression, this is significantly less than that observed in the presence of both Vg and Sd. Titration of the amounts of transfected Vg and Sd plasmids with all enhancers shows that the relative concentration of the two factors is critical and, at any given Vg concentration, high levels of Sd reduce activation (Halder, 1998).

To define the sequences of the enhancers that respond to Vg/Sd, the activation of smaller fragments from the 704-bp SRF intervein C enhancer, the 806-bp vg quadrant enhancer, and the 754-bp vg boundary enhancer in tissue culture were analyzed. A 125-bp fragment (SRF-A) derived from the 5' end of the SRF enhancer is activated, whereas an adjacent 131-bp fragment (SRF-B) is not activated. A 65-bp fragment from the vg quadrant enhancer (MD2) has been identified that, when multimerized, produces an expression pattern very similar to the full-length enhancer in wing discs. When assayed in tissue culture, MD2 is activated by Vg and Sd. Within the vg boundary enhancer, a 120-bp fragment sufficient to drive reporter gene expression in the wing pouch (vg-A) as well as a nonoverlapping 90-bp fragment (vg-B) are also activated synergistically by cotransfection of Vg and Sd. Sd was shown, using mobility shift and DNase I footprinting assays, to bind specifically to essential sites for target gene activation (Halder, 1998).

One possible reason for the importance of the concentration of Sd on Vg function concerns the localization of the Vg protein. It was observed that in S2 cells transfected with the vg expression plasmid alone, the Vg protein appears to be localized to both the cytoplasm and the nucleus. In contrast, in cells cotransfected with the Vg and Sd expression plasmids, Vg is clearly localized to nuclei. Vg localization is more diffuse in sd mutant clones than in sd+ cells; this is also true of ectopic Vg localization in regions of imaginal discs that lack endogenous Sd expression. Furthermore, deletion of the Sd interaction domain of Vg results in cytoplasmic accumulation of Vg in vivo. Thus, Sd may facilitate the transport or retention of Vg protein in the nucleus and, coupled with the concentration-dependent, synergistic effects of Vg and Sd on target gene expression, these results suggest that the proteins form a complex in vivo (Halder, 1998).

These results demonstrate that the activation of several genes in the wing field requires Vg/Sd function. It is also known that for each of the cis-regulatory elements analyzed here, direct input(s) of particular signaling pathways are also required. Specifically, the activation of the SRF intervein C element requires both Vg/Sd and Hh signaling; the activation of the vg boundary enhancer requires Vg/Sd and N signaling, and the activation of the vg quadrant enhancer requires Vg/Sd and Dpp signaling. Because these regulatory elements are not expressed in all tissues in which the signals are active, nor in all wing cells in which Vg/Sd are active, it is deduced that neither the input of various signals nor of Vg/Sd alone are sufficient for gene activation in vivo. Rather, the results suggest that the various wing-specific cis-regulatory elements require a combination of direct inputs, comprising the Vg/Sd selector function, which restricts expression to the wing field, and at least one signal transducer that mediates signaling inputs and hence, the pattern of gene expression within the wing field. One prediction of this model is that gene expression patterns within the wing field may be changed by altering the signal-transducer binding sites within a cis-response element. To test this, the Suppressor of Hairless [Su(H)] binding site that mediates the N input in the vg boundary enhancer was changed to sites for the Cubitus interruptus (Ci) protein that transduces Hh signaling. This switches the pattern of gene expression from a N-induced dorsoventral stripe to a Hh-induced anteroposterior stripe while retaining the restriction of gene activation to the wing disc (Halder, 1998 and references).

These results demonstrate that the role of the Vg/Sd selector function is to directly regulate wing-specific cis-regulatory elements that also require particular signaling inputs. The patterns of gene expression induced in the wing disc are limited to cells in which both the selector genes and specific signaling pathways are active. The response of the SRF-A, vg boundary, and vg quadrant enhancers to Hh, N, and Dpp signaling are limited to the wing pouch by Vg/Sd and occur in different patterns because of their direct regulation by the Ci, Su(H), and Mad proteins, respectively. Furthermore, the finding that the changing of the Su(H) binding site into a Ci binding site in the vg boundary enhancer switches the pattern from a wing-specific dorsoventral N-regulated stripe to a wing-specific anteroposterior Hh-regulated stripe suggests that spatial expression patterns are determined by the sites for individual DNA-binding signal transducers. One corollary of this model is that for any given signaling protein, different selector proteins may be involved in directing tissue-specific responses in different organs and tissues. For example, other studies have shown that tissue-specific enhancers in the embryo that are regulated by Dpp also require the action of the Labial/Extradenticle or Tinman selector proteins to limit expression to the endoderm or mesoderm, respectively. It is suggested that, in general, combinatorial control by selector proteins and common signal transducers at a cis-regulatory level is required for the tissue- and organ-specific responses of target genes to widely deployed signaling systems (Halder, 1998 and references).

GENE STRUCTURE

Exons - 8

Base pairs in 3' UTR - 549

PROTEIN STRUCTURE

Amino Acids - 453

Structural Domains

Vestigial is a novel protein with no known homologs. The protein regions flanking the exon 4 glycine repeat are probably functional, because strong protein sequence conservation between D. melanogaster and D. virilis is seen in the intervals flanking the exon 4 glycine-rich interval. Two regions containing alternating histidine residues are present in the amino-terminal domain and are similar to the paired or His-Pro repeat. Similar regions are found in paired, bicoid, daughterless, Deformed and E74. Homology in this region is most similar to Deformed (Williams, 1991).

date revised:  3 July 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.