BIOLOGICAL OVERVIEW

A recurring, significant theme in insect development is the subdivision of the embryo into ever greater numbers of compartments within segments. At the earliest stages of development segments are defined by pair rule genes, and subsequently, each segment is subdivided into anterior and posterior compartments by the action of segment polarity genes. wingless, as a segment polarity gene, has a role in the establishment of different cell fates, working within and between the anterior and posterior compartments of segments.

Normally, each thoracic and abdominal segment contains an anterior denticle band, and a more posterior region of naked cuticle. In wingless mutants, the naked cuticle is absent, replaced by a disordered array of denticles (Bejsovec, 1991).

The effects of wingless mutation on morphology are mirrored by events inside the embryonic cells. Wingless is secreted by cells in each of 14 posterior compartments of parasegments (embryonic segments). Wingless secretion is dependent on Hedgehog, produced in adjacent compartments. Lack of functional posterior parasegmental compartments (due to a failure to secrete Wingless) results in altered activity just underneath the outer cell membrane. There is an altered distribution of Armadillo, and altered expression of shaggy/zeste-white3. Armadillo is associated with adherens junctions, structures that bind one cell to another, and Shaggy is involved in the transmission of the wingless signal inside the cell. Mutation of wingless also alters the secretion of cuticle and the regulation of denticle production both in the posterior cells of each compartment, and in adjacent cells that would otherwise have responded to wingless signaling.

Wg influences two distinct cellular decisions in patterning the larval ventral epidermis. This segmentally repeating pattern consists of six rows of uniquely shaped denticles arranged in a belt at the anterior of the segment, anterior to the cells that secrete Wingless protein, and an expanse of smooth, naked cuticle form in the posterior portion of the segment. In the absence of wg both the generation of diverse denticle types and the specification of naked cuticle are disrupted, resulting in a lawn of uniform denticles. wg is expressed in one row of cells in each wild-type segment, roughly in the middle of the naked cuticle region. Thus Wg activity influences cell fate decisions many rows of cells away from its source. What then accounts for the two cell fate regulated by Wg signaling in the ectoderm (Moline, 1999)?

Proper pattern formation requires temporal as well as spatial control of Wg activity (Bejsovec, 1991). Analysis of a temperature-sensitive wg allele that is wild type at 18^oC and null for function at 25^oC has shown that Wg activity between 4 and 5.5 hours of development generates diverse denticle types and stabilizes the expression of engrailed. en is a segment polarity gene expressed in the two rows of cells just posterior to the wg domain, at the posterior boundary of each segment. After 6 hours, Wg activity no longer produces these cellular responses, but instead promotes the naked cuticle-secreting cell fate. Thus the population of cells responding to Wg activity changes during development (Moline, 1999 and references therein).

Wg and Wnt molecules tightly associate with membrane and extracellular matrix and appear not to be readily soluble. Thus, it is unlikely that these proteins freely diffuse through extracellular spaces. Rather, Wg appears to be transported via active cellular processes. This phenomenon was first demonstrated using the shibire^ts (shi^ts) mutation to block endocytosis (Bejsovec, 1995). shi encodes the fly dynamin homologue, a GTPase required for clathrin-coated vesicle formation. Rather than the broad, punctate Wg protein distribution normally found over several cell diameters on either side of the wg-expressing cells, shi mutant embryos show high level accumulation of Wg around the wg-expressing cells (Moline, 1999).

Reducing endocytosis in defined domains within the segment, through moderate-level expression of a dominant negative form of Shibire, alters the normal distribution of Wg and changes the domain of cells that respond to Wg. When expressed using the prd-Gal4, shi^D reduces both anterior and posterior movement of Wg protein, causing it to accumulate in and around the wg-expressing row of cells. Driving expression of shi^D with the en-Gal4 reduces movement only in the posterior direction, since the en-expressing cells are a non-overlapping cell population just posterior to the wg-expressing row of cells (Moline, 1999).

The effects on cuticular pattern elements indicate that Wg moving in an anterior direction from the row of wg-expressing cells defines the domain of cells destined to secrete naked cuticle, whereas posterior movement of Wg is required for correct specification of denticle types in the anterior of the adjacent segment. The patterning defects caused by shi^D expression are reversed by co-expression with wg plus, suggesting that the primary effect of reducing endocytosis in the embryonic epidermis is a disruption of Wg protein transport. Moreover, en-Gal4-driven shi^D reduces endocytosis in a non-wg-expressing group of cells, and causes patterning defects in the cell population posterior to the en domain. Thus, reducing Wg transit through the en cells ‘casts a shadow’, producing patterning anomalies in an otherwise wild-type cell population. This supports the idea that Wg ligand is moved by active cellular processes through cells to arrive at distant target cell populations in the embryo (Moline, 1999).

The results suggest that, during normal development, the temporal changes observed in directionality of Wg protein movement (Gonzalez, 1991) may correlate with the temporal changes in its apparent function (Bejsovec, 1991). In wild-type embryos prior to stage 10, Wg protein is detected over many cell diameters both anterior and posterior to the wg-expressing row of cells (Gonzalez, 1991). Disrupting posterior movement of Wg alters patterning of at least the first three rows of denticles in the segment posterior to the affected source of Wg. Thus, posterior movement of Wg is detectable during the early time period when Wg activity is required in these cells for the generation of diverse denticle types and for the stabilization of en expression (Bejsovec, 1991). At and after stage 10, Wg protein is no longer detected in cells posterior to the wg-expressing row, including the en-expressing cells of that segment, and shows an asymmetric distribution toward the anterior of the segment (Bejsovec, 1991; Gonzalez, 1991). The results reported here correlate this anterior movement with specification of the correct expanse of naked cuticle-secreting cells, presumably through Wg-mediated antagonism of the EGF pathway. This is consistent with previous reports that, after stage 10, Wg is no longer required for maintenance of en expression (Bejsovec, 1991) or for the generation of denticle diversity, and instead promotes specification of naked cuticle cell fate (Bejsovec, 1991, Moline, 1999).

It is unclear by what mechanism Wg is excluded from the posterior cells at stage 10. It is proposed that wild-type naked gene function may contribute to the change in direction of Wg protein movement. Reducing Wg movement through the en-expressing cells eliminates Wg-mediated specification of excess naked cuticle and substantially rescues the nkd mutant phenotype. Thus, posterior movement of Wg from the adjacent segment, and not anterior movement of Wg within the segment, appears to be responsible for the naked mutant phenotype. This observation suggests a role for nkd gene function in restricting posterior Wg transport (Moline, 1999).

Although some aspects of Wg transport appear to be independent of Wg signal transduction, the two processes cannot be completely separated. Overexpression of Dfz2, a Wg signaling receptor, appears to restrict the distribution of the Wg protein, suggesting that it has the capacity to sequester ligand. In contrast, Dfz2 overexpression in the imaginal disc has been shown to enhance the transport of Wg protein and consequently increase its range of activity. This dramatic change in the role of Dfz2 from embryo to imaginal disc suggests that mechanisms controlling Wg distribution may differ between these two developmental stages of Drosophila. For example, recent work has revealed that imaginal disc cells project cytoplasmic extensions, called cytonemes, toward the source of signaling molecules at the center of the discs. These extensions may assist in the broad distribution and long-range activity documented for Wg in the imaginal discs (Moline, 1999 and references therein).

Such cytoplasmic extensions have not been detected in vivo in embryonic epidermal cells. If embryonic cells do produce cytonemes, they may not be functionally relevant to the distribution of Wg signaling activity. Reducing endocytosis in the two rows of en-expressing cells produces Wg-related pattern disruptions in the cells posterior to the affected domain. This suggests that Wg must physically move through the en cells in order to influence cell fate decisions in the posterior cell population. Such an effect would not be predicted if the posterior population were able to extend cytoplasmic projections through the affected 2 cell diameters and directly contact the cells expressing wg (Moline, 1999).

Mutant Wg molecules that are secreted properly, but fail to signal, are transported as if by default (Bejsovec, 1995). Initially, these mutant embryos show a wild-type distribution of Wg protein, but over time they begin to accumulate Wg-containing vesicles in tissues that do not express the gene and in which the protein is not normally detected. This indicates that most, if not all, embryonic cells have the ability to internalize Wg, and that this process does not require signal transduction. Moreover, it suggests that the mutant Wg ligand is able to bind to a cell surface receptor that does not transduce signal. This is consistent with a multiple-receptor model for Wg, where some Wg-binding receptors are dedicated exclusively to the transport process. Thus the dynamic distribution of Wg during development may reflect an interplay between signaling receptors and other cell surface molecules essential for ligand transport (Moline, 1999). These results suggest that a single signaling molecule, in this case Wingless, can determine multiple cell fates. These alternate cell fates depend on cell autonomous temporal changes in responsiveness to the Wg ligand and on regulated transport across adjacent cell populations that facilitate or interfere with this transport differently.

The effects of wingless signaling in the margin of the wing are fairly well understood. Here decapentaplegic is not expressed adjacent to Wingless producing cells, as is the case in embryonic segmentation. Any possible compounding effects attributable to DPP are removed, due to its absence, thus demonstrating a pure wingless effect. In the case of the wing, wingless expression is independent of hedgehog while dpp expression remains dependent on hh. The anterior edge of the wing is marked by stout, slender, and chemosensory bristles, all three types of which are innervated. Bristles and epidermal hairs are not innervated. Thus in the wing margin one can more easily observe the effect of the presence or the absence of wingless on bristle cell production and innervation, without having to contend with the effects of dpp production.

Both achaete and cut are involved in the specification of sensory bristles, the peripheral sense organs of the wing margin. wingless is expressed in a narrow band of cells. Adjacent cells which do not produce wingless serve as precursors of both sensory and non-sensory elements. Cut protein is expressed in a wingless dependent fashion in cells expressing wingless; achaete is expressed in the adjacent cells, those not expressing wingless. Both cut and achaete expression are dependent on wingless. The wings of flies carrying conditional lethal mutations of wingless show an absense of bristles; mechanoreceptors are transformed into chemoreceptors and the arrangement of chemoreceptors is altered. Thus the wingless signal modifies the production of achaete and cut resulting in altered sensory cell and bristle production (Couso, 1994). In summary, wingless critically regulates the production of bristles and sensory cells on the wing margin. It does this as a secreted molecule acting locally on adjacent cells, modifying the production of Cut and Achaete, two proteins involved in neurogenesis.

It has been suggested that wingless expression at the dorsal-ventral boundary of the wing disc depends on a signal from dorsal to ventral cells mediated by Serrate and Notch. Wingless expression is lost from the wing margin and the size of the wing is significantly reduced when Notch activity is removed from the third instar larva using a temperature sensitive allele of Notch. Therefore, it is likely that wingless is regulated by the Notch pathway acting through Suppressor of Hairless (Diaz-Benjumea, 1995).

Wingless has an earlier role in specification of the wing. Wing discs arise during embryonic development from a region of the epidermis devoid of wg expression. Ten to thirteen cells in each wing primordium express engrailed but not wingless. Thus, the obligitory role of wingless in leg disc formation does not appear to hold for wing disc formation.

During the second larval instar wg expression is first detected in the anterior compartment of wing discs. wingless appears to have a primary role in specifying the wing primordium. This conclusion is based on the observation that ectopic expression of wg can induce supernumary wings in the portion of the disc normally fated to give rise to body wall. Thus WG protein can reprogram cells in the notum to wing pouch identity very early in wing development. An important target of WG in this function is the gene pdm-1 which is involved in specifying the proximal-distal axis of the wing (Ng, 1996).

Thus, two distinct roles for wingless in wing morphogenesis have been identified: a primary role in specifying the wing primordium, and subsequent role mediating the patterning activities of the dorso-ventral compartment boundary (Ng, 1996).

GENE STRUCTURE

Bases in 5' UTR - 417

Exons - five

Bases in 3' UTR - 1085

PROTEIN STRUCTURE

Amino Acids - 468

Structural Domains

The WG protein has an N-terminal hydrophobic region characteristic of a signal sequence whose function is to expedite secretion. There is one potential N-linked glycosylation site. The protein is rich in conserved cysteine residues (Rijsewijk, 1987).

date revised:  2 January 2001 

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

Please e-mail comments/corrections to brodyt@codon.nih.gov