Determination of the Wing Imaginal Discs

Determination of discs from ectoderm: distal-less protein

The molecular biology of insect metamorphosis begins with the specification of certain epidermal cells to become imaginal disc precursors. As we discussed in Chapter 9, the organ rudiments in Drosophila are specified on an orthagonal grid by intersecting anterior-posterior and dorsal-ventral signals. In most segments, Hox gene products prevent Distal-less gene expression and the establishment of limb primordia; but in those segments that are specified to be thoracic, limb formation is permitted. Cohen and his colleagues (1993) have demonstrated that the leg and wing originate from the same set of imaginal precursors, specified at the intersections between the anterior-posterior stripes of Wingless (Wg) protein expression and the horizontal band of cells expressing the Decapentaplegic (Dpp) protein. Both proteins are soluble and have a limited range of diffusion. In the early Drosophila embryo (at germ band extension about 4.5 hours after fertilization), a single group of cells at these intersections forms the imaginal disc precursors in each thoracic segment. These cells (and only these cells) express the Distal-less protein. As the cells expressing Dpp are moved dorsally, some of these Distal-less-expressing cells move with them to establish a secondary cluster of imaginal cells (derived from the original ventral cluster). The initial clusters form the leg imaginal disc, while the secondary clusters form the wing or haltere disc. Thus, the leg and wing discs have a common origin (Figure 18.16).

Determination of disc identity: vestigial protein

Despite their common origin, it is obvious that the leg and wing discs are determined to become different structures. The determination of the wing disc appears to be regulated by the vestigial gene. Using a targeted gene expression system, Kim and colleagues (1996) have caused the vestigial gene to be expressed in eye, antenna, and leg discs (Figure 18.17). When this happens, regions of the normal structure are converted into wing tissue.

Determination of the anterior-posterior axis: engrailed and decapentaplegic proteins

The axes of the wing are specified by interactions at their compartmental boundaries (Meinhardt 1980; Causo et al. 1993; Tabata et al. 1995). After these initial interactions, a polar coordinate system may subdivide the wing regions more finely (Held 1995).

During the first larval instar, the leg and wing imaginal discs acquire their anterior-posterior axis. The discs become split into two compartments representing the future anterior and posterior regions of the appendage (i.e., the front of the wing and the rear of the wing). Based on the position of its cells in the segment, the posterior compartment of the wing disc expresses the engrailed gene (Figure 18.18; Garcia-Bellido et al. 1973; Lawrence and Morata 1976). If engrailed function is absent, all the disc cells become anteriorized. The boundary between the posterior and anterior compartments is strictly observed. Cells from one side cannot produce descendants that cross over the boundary to the other.

The Engrailed protein is a transcription factor, and it activates the hedgehog gene in the cells of the posterior compartment. The posterior wing disc cells express the Hedgehog protein, which acts as a short-range signal to induce the expression of Dpp in adjacent anterior cells, while the expression of engrailed in the posterior cells renders them nonresponsive to the Hedgehog they secrete (preventing them from expressing Dpp). Dpp is a TGF-β family paracrine factor that acts as a long-range signal to establish the anterior-posterior axis of the wing (Guillen et al. 1995; Tabata et al. 1995; Nellen et al. 1996). Cells in both compartments close to the area of Dpp expression are exposed to relatively high concentrations of this protein and activate the spalt and oculomotor-blind (omb) genes. Those cells farther away receive lower concentrations of Dpp and activate only the omb gene. These two genes encode transcription factors that specify the region of the wing from the center (where Dpp is expressed) to the periphery.

Dorsal-ventral axis: wingless and apterous proteins

During the second larval instar, the dorsal-ventral axis of the wing disc is determined. The dorsal-ventral boundary lies at the future margin of the wing blade, separating the upper surface of the wing from the lower (Bryant 1970; Garcia-Bellido et al. 1973). The gene involved in this compartmentalization event is apterous. Cells expressing apterous become the dorsal cells (Figure 18.19A; Blair 1993; Diaz-Benjumea and Cohen 1993). When apterous is deleted, all cells in the disc acquire ventral fates. The Apterous protein is a transcription factor that activates the genes for the Serrate and Fringe proteins. Serrate is a ligand for the Notch receptor, and Fringe is involved in regulating Notch ligand binding (Irvine and Wieschaus 1994; Williams et al. 1994; Kim et al. 1995). The Notch receptor is found in the ventral cells, so the binding of Notch on the ventral side with Serrate on the dorsal side stabilizes the wing margin.

The future wing margin also becomes a signaling center, analogous to the Dpp-secreting band that forms the anterior-posterior axis (Zecca et al. 1996; Neumann and Cohen 1997). The Fringe and Serrate proteins act to promote the transcription of the Wingless gene at the border of the dorsal and ventral compartments (Figure 18.19B-D). Wingless is secreted by the cells adjacent to the border in both compartments, and it acts as a long-range morphogen. Those cells that are close to this border and receive high concentrations of Wingless protein express the achaete-scute genes. The next group of cells receives moderate concentrations of Wingless and expresses the distal-less gene. Those wing cells receiving relatively low amounts of Wingless express the vestigial gene. However, only those cells expressing the Apterous protein (i.e., the dorsal cells) are able to respond to these signals. Thus, Wingless protein acts as a morphogen on the dorsal surface only.

One of the problems with this model of diffusible gradients is that although the influence of Wingless extends to the edges of the wing disc, neither Hedgehog nor Wingless protein travels well in the extracellular environment. The transport of these proteins does not seem to be due to simple diffusion. One possibility is that the cells actively transport Wingless in a particular direction (Pfeiffer and Vincent 1999). Alternatively, Wingless might be transported locally by diffusion and more widely by cytoplasmic filaments that extend from the peripheral cells into the central region producing the Wingless protein. Evidence for this latter model comes from Ramírez-Weber and Kornberg (1999), who identified extremely thin processes extending from the peripheral cells, across the wing disc, to the sites of Wingless synthesis (Figure 18.20). These actin-based extensions are called cytonemes, and they are similar to the thin filopodia of sea urchin mesenchymal cells. Evidence has yet to show that they are active in transporting the Wingless protein to the peripheral cells. However, this is an exciting and unexpected mechanism for some types of long-range cell-cell communication.

Proximal-distal axis: wingless protein

In addition to acting as a morphogen defining the dorsal-ventral axis, the Wingless protein also acts to promote cell division and the extension of the wing (Neumann and Cohen 1996). The interaction between the dorsal-ventral and anterior-posterior axes at their boundaries is critical for the outgrowth along the proximal-distal axis. During metamorphosis, the “distalization” of the proximal-distal axis from the base of the thorax outward to the tip of the wing or leg is accomplished by cell interactions at the boundaries between the other two axes.*


18.1 Creation of the dorsal-ventral wing surfaces. The juxtaposition of those cells expressing Apterous with those that do not initiates a cascade of gene expression that results in markedly different cell types. These events were predicted by theoretical biologists years before the molecules were discovered.


18.2 Homologous specification. If a group of cells in one imaginal disc are mutated such that they give rise to a structure characteristic of another imaginal disc (for instance, cells from a leg disc giving rise to antennal structures), the regional specification of those structures will be in accordance with their position in the original disc.

Figure 18.16. Schematic model for the initiation and separation of the leg-wing disc in the Drosophila thorax.

Figure 18.16

Schematic model for the initiation and separation of the leg-wing disc in the Drosophila thorax. The embryo is divided into an orthagonal grid by vertical stripes of Wingless (Wg) and a horizontal band of Decapentaplegic (Dpp) synthesis and secretion. (more...)

Figure 18.17. The vestigial gene determines wing disc identity.

Figure 18.17

The vestigial gene determines wing disc identity. If the vestigial gene is expressed in some of the cells of an eye imaginal disc, those cells will be determined as wing cells, and that part of the eye will form wing tissue (arrows). (After Kim et al. (more...)

Figure 18.18. Compartmentalization and anterior-posterior patterning in the wing disc.

Figure 18.18

Compartmentalization and anterior-posterior patterning in the wing disc. (A) In the first-instar larva, the anterior-posterior axis has been formed and is manifested by the expression of the engrailed gene in the posterior compartment. The Engrailed protein (more...)

Figure 18.19. Determining the dorsal-ventral axis.

Figure 18.19

Determining the dorsal-ventral axis. (A) The prospective ventral surface of the wing is stained by antibodies to the Vestigial protein (green), while the prospective dorsal side is stained by antibodies to the Apterous protein (red). The region of yellow (more...)

Figure 18.20. Cytonemes in the third-instar wing disc of Drosophila.

Figure 18.20

Cytonemes in the third-instar wing disc of Drosophila. (A) The GFP reporter gene was expressed in cells of the anterior and posterior flanks of the wing disc, but not among the central cells that are synthesizing Wingless. (B) Enlargement of a peripheral (more...)


Yes, this process is complex, and likely to get more so as we learn more about it. This situation is not without its humor. Sydney Brenner (1996) recalls Nobel laureate Francis Crick being frustrated by this complexity and saying, “God knows how these imaginal discs work.” Brenner fantasized a meeting wherein Crick asked the Deity how He constructed these entities, only to have God bewildered by their complexity as well. Eventually, all God could do was to reassure Crick that “we've been building flies here for 200 million years and we have had no complaints.”

From: Metamorphosis: The Hormonal Reactivation of Development

Cover of Developmental Biology
Developmental Biology. 6th edition.
Gilbert SF.
Sunderland (MA): Sinauer Associates; 2000.
Copyright © 2000, Sinauer Associates.

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.