SEARCH

SEARCH BY CITATION

Keywords:

  • imaginal disc;
  • Keilin's organ;
  • leg;
  • neural crest;
  • cranial placode;
  • neurogenic ectoderm;
  • cut;
  • escargot;
  • couch potato

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

The thoracic limb primordium of Drosophila melanogaster is a useful experimental model in which to study how unique tissue types are specified from multipotent founder cell populations. The second thoracic segment limb primordium gives rise to three structures: the wing imaginal disc, the leg imaginal disc, and a larval mechanosensory structure called Keilin's organ. We report that most of the limb primordium arises within neurogenic ectoderm and demonstrate that the neural and imaginal components of the primordium have distinct developmental potentials. We also provide the first analysis of the genetic pathways that subdivide the progenitor cell population into uniquely imaginal and neural identities. In particular, we demonstrate that the imaginal gene escargot represses Keilin's organ fate and that Keilin's organ is specified by Distal-less in conjunction with the downstream achaete-scute complex. This specification involves both the activation of the neural genes cut and couch potato and the repression of escargot. In the absence of achaete-scute complex function, cells adopt mixed identities and subsequently die. We propose that central cells of the primordium previously thought to contribute to the distal leg are Keilin's organ precursors, while both proximal and distal leg precursors are located more peripherally and within the escargot domain. Developmental Dynamics 232:801-816, 2005. © 2005 Wiley-Liss, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

The adult appendages of the fruit fly, Drosophila melanogaster, originate from larval structures called imaginal discs. These discs consist of involuted sheets of epithelial cells that evert and undergo terminal differentiation during metamorphosis (reviewed in Cohen, 1993). The thoracic imaginal discs are derived from paired clusters of cells in the ventrolateral ectoderm of the three thoracic segments (designated as T1, T2, and T3 from anterior to posterior) of the embryo. Each of the six clusters ultimately will give rise to three structures: a dorsal imaginal disc (the humeral disc in T1, the wing disc in T2, and the haltere disc in T3), a leg imaginal disc, and a larval sensory organ called the Keilin's organ. The Keilin's organ is proposed to be a vestige of a larval leg similar to those found in more primitive insects (Keilin, 1911, 1915). Collectively, we term the cluster of cells that will give rise to both imaginal and neural tissue the “thoracic limb primordium.” One of the earliest known markers of this primordium is the homeodomain transcription factor Distal-less (Dll). Dll first is expressed in all six thoracic primordia at late embryonic stage 10 (Cohen et al., 1989; Cohen, 1990). Whereas some leg disc, some wing disc, and all Keilin's organ cells express Dll, there appear to be progenitors of both the wing and leg that do not require Dll and may never express it (Campbell and Tomlinson, 1998; Weigmann and Cohen, 1999; Andrews and Boekhoff-Falk, unpublished observations). Consistent with this finding, in in vivo culture experiments, it was demonstrated that tissue from the thoracic segments of Dll null embryos could give rise to adult disc-derived structures, but only of wing and proximal leg identity (Cohen et al., 1993). Thus, Dll is not required for formation of the primordium or for specification of imaginal identity.

By late stage 12, those cells that will give rise to leg and wing imaginal tissue begin to express a second marker, the zinc-finger transcription factor escargot (esg; Fuse et al., 1996; Goto and Hayashi, 1997b). esg is required for the maintenance of diploidy in at least some imaginal tissues in the embryo (Hayashi et al., 1993). At stage 12, cells of the wing and haltere discs can be distinguished from those of the leg both by their dorsal position and by expression of the markers vestigial (vg) and snail (sna; Goto and Hayashi, 1997b). During stage 12, Dll is lost from wing and haltere disc cells. Whereas the wing and haltere disc cells migrate dorsally, cells of the leg disc and Keilin's organ remain intimately associated (Kojima et al., 2000). We refer to the subset of cells within the thoracic limb primordium that give rise to the leg and Keilin's organ as the “leg/Keilin's organ primordium.”

Earlier work has led to some understanding of the mechanisms underlying the specification and subdivision of the thoracic limb primordia. For instance, these primordia are positioned by both signal and selector inputs. Initial activation of Dll in the thorax is dependent upon the Wingless (Wg) signal (Cohen, 1990; Cohen et al., 1993) and a zinc-finger transcription factor encoded by buttonhead (btd; Wimmer et al., 1993; Estella et al., 2003). Later Dll expression is Wg-independent and subject to autoregulation (Cohen et al., 1993). Dorsally, Dll expression is subject to negative regulation by the Decapentaplegic (Dpp) signaling pathway (Goto and Hayashi, 1997b). On the ventral side, the epidermal growth factor (EGF) pathway prevents Dll expression (Raz and Shilo, 1993). Formation of limb primordia in the abdominal segments is prevented by the Hox selectors Ultrabithorax and Abdominal A (Vachon et al., 1992). There is no evidence for regulation of Dll by the thoracic Hox genes Sex combs reduced (Scr) and Antennapedia (Antp), although these genes play essential roles in patterning of the larval leg disc (Struhl, 1981, 1982; Percival-Smith et al., 1997; Casares and Mann, 1998; Percival-Smith and Hayden, 1998; and our observations).

The Dpp and EGF pathways also are later required to distinguish between wing and leg fates within the limb primordium. High levels of dpp activity are necessary for the specification of wing cells, whereas somewhat lower levels specify the leg disc (Goto and Hayashi, 1997b; Kubota et al., 2000). Establishment of leg disc identity also requires input from the EGF pathway, in part to antagonize dpp function (Kubota et al., 2000). In the absence of the EGF signal, Dpp signaling results in the conversion of the leg precursors to wing or haltere fates (Kubota et al., 2000).

The relative positions of cells that give rise to the various derivatives of the primordium remain uncertain. It has been proposed that fates initially are arranged sequentially along the dorsal–ventral axis (Goto and Hayashi, 1997b). According to this view, cells of the presumptive humeral, wing, and haltere discs occupy the dorsal–most position. The dorsal location of the wing and haltere precursors has been demonstrated using the perdurance of β-galactosidase protein expressed under control of the Dll early embryonic enhancer (Cohen, 1993; Goto and Hayashi, 1997b). In this model, cells occupying an intermediate dorsoventral position become proximal leg, whereas the ventral-most cells become distal leg, and proximal leg precursors subsequently migrate ventrally to surround the presumptive distal leg (Goto and Hayashi, 1997b). An alternative model invokes the specification of a distal organizer that subsequently recruits surrounding cells to become the proximal tissue (Campbell and Tomlinson, 1995). One problem with both models is that the origin of the Keilin's organ is neglected. A second issue concerns the potential roles of Keilin's organ precursors in regulating patterning of the adjacent imaginal tissue and vice versa.

Indeed, the specification of the cells of the presumptive Keilin's organ has received little attention, despite that this structure arises from the thoracic limb primordium, constitutes a distinct tissue type from the imaginal discs, and may play a role in axon pathfinding from the adult legs to the CNS (Jan et al., 1985; Tix et al., 1989). Here, we elucidate the relationships between proneural and imaginal determinants for specifying neural versus imaginal fates.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Thoracic Limb Primordium Arises in the Neuroectoderm and Expresses Members of the ASC

Both the thoracic limb primordia and ventral nerve cord (VNC) of the central nervous system (CNS) originate ventrolaterally in the embryonic ectoderm. However, where these structures are positioned relative to one another was unknown. Given that the thoracic limb primordia will give rise to structures of both ectodermal and neural origin, one could imagine that mechanisms that pattern the VNC also might pattern the neural portion of the primordia. We, therefore, wished to determine the spatial relationship between the thoracic limb primordia and the neurogenic ectoderm. To address this issue, embryos carrying a transgene driving lacZ, under control of the seven-up (svp) promoter, were stained for Dll and β-galactosidase. svp-lacZ drives reporter expression in neuroblasts. Examination of stage 11 svp-lacZ embryos revealed that the ventral two thirds of the thoracic limb primordium lies within the neuroectoderm, whereas the dorsal-most third extends beyond the zone of svp-lacZ expression (Fig. 1A,B,B′). Thus, it seemed likely that mechanisms involved in patterning of the neuroectoderm of the CNS were involved in patterning the thoracic limb primordium.

thumbnail image

Figure 1. The thoracic limb primordium arises within the neurogenic ectoderm and shows spatiotemporal differences in the expression of members of the ASC. A–B′: Expression of Dll (green), which marks the primordium, relative to the svp-lacZ reporter (red), which marks all neuroblasts (NBs) and sensory mother cells (SMCs) in a wild-type stage 11 embryo. Note that the ventral Dll-expressing cells are situated within the neuroectoderm. C–F′: Expression of L'sc (blue) and Dll (green) in a wild-type stage 11 embryo (C–D′) and in a wild-type stage 12 embryo (E–F′). Note the clusters of cells that coexpress Dll and L'sc (asterisks) in the stage 11 animal and that this coexpression is gone by stage 12. The approximate location of cells that coexpressed Dll and L'sc at stage 11 is marked by asterisks in F and F′. G–J′: Expression of Dll (green) and Ac (pink) in a wild-type stage 11 embryo (G–H′) and at stage 12 (I–J′). In contrast to the observations for L'sc, note that there is coexpression of Dll and Ac (asterisks in J, J′) at stage 12 rather than stage 11 and that the number of coexpressing cells is smaller. Approximate locations of cells that will coexpress Dll and Ac at stage 12 are marked with asterisks in H and H′. K–K″: Expression of Dll (green), 22C10 (red), and Cpo (blue) in the primordium of a wild-type stage 15 embryo. Arrowheads mark cells that coexpress Dll and Cpo but do not show 22C10 expression and that we believe will become the support cells for the Keilin's organ nerves. Asterisks mark cells that express Dll, Cpo, and 22C10 that are likely to innervate Keilin's organ. A,C,E,G,I: Views obtained with a ×10 objective. The remaining panels were obtained using a ×40 objective. Anterior is to the left and dorsal is up in all panels. l, lateral column of neuroblasts; i, intermediate column of neuroblasts; m, medial column of neuroblasts.

Download figure to PowerPoint

The role of the ASC in specifying neural fates in the embryonic central and peripheral nervous systems (CNS and PNS, respectively) is well-documented (reviewed in Goodman and Doe, 1993; Skeath, 1999). Genes of the ASC encode basic helix-loop-helix (bHLH) transcription factors. Each of these is expressed in particular subsets of the embryonic proneural clusters. Within a proneural cluster only one cell (or a few) becomes a CNS neuroblast (NB) or PNS sensory mother cell (SMC). The remaining cells adopt ectodermal fates. A key role of ASC proteins is preventing the ectopic production of neurons by a process termed “lateral inhibition.” ASC proteins achieve lateral inhibition by up-regulating Notch signaling in non-NB/SMC members of each proneural cluster (reviewed in Goodman and Doe, 1993; Skeath, 1999).

Given the overlap of the thoracic limb primordium with the neuroectoderm, we wished to determine the patterns of expression and the functions of ASC gene products within the primordium. We first determined the expression patterns of Lethal of Scute (L'sc) and Achaete (Ac) relative to that of Dll. At early stage 11, approximately eight cells expressing both Dll and L'sc are observed (Fig. 1C,D,D′). The expression of L'sc is lost by stage 12 (Fig. 1E,F,F′). In contrast, Ac is not detectable in the primordium at stage 11 (Fig. 1G,H,H′) but becomes apparent at early stage 12, but in only three to four cells (Fig. 1I,J,J′). This sequential deployment of proneural ASC genes is unprecedented in either the CNS or the PNS. In wild-type embryos stained for Dll and Ac, the latter being expressed in the proneural clusters and neuroblasts of the lateral and ventral columns of the CNS, the center of the thoracic limb primordium is within the lateral column, while its ventral-most tip is at the medial–ventral boundary (Fig. 1G,H,H′). It may be that expression of l'sc marks the Keilin's organ proneural clusters, while ac expression at stage 12 is restricted to the presumptive SMCs. This explanation would be reminiscent of asense (ase) activation in SMCs/NBs after lateral inhibition (Brand et al., 1993; Dominguez and Campuzano, 1993). Examination of the number of neurons at stage 15 (when neural differentiation is more advanced) using the neuron-specific antibody 22C10 (Zipursky et al., 1984) and anti-Couch Potato (Cpo), a marker for all cells of PNS identity (Bellen et al., 1992), indicates that the cpo-expressing cells of the primordium are neurons (Fig. 1K–K″). At least three of the cpo-positive cells have adopted neuronal identity by stage 15 (asterisks in Fig. 1K–K″). This finding is consistent with a previous study in which the number of neurons in Keilin's organ was estimated at five (Dambly-Chaudiere and Ghysen, 1987).

All Cells of the Primordium Adopt Either Neural or Imaginal Identities

At stage 11, no distinction is apparent between cells that will adopt imaginal and neural identities. At this stage, the Dll-expressing cells exhibit uniform expression of Dll and Hth (Fig. 2A–A″) and no expression of a neural marker cut (ct) or the imaginal marker esg (Figs. 2A–A″, 3A,C). By stage 14, Hth is excluded from the presumptive neural portion of the primordium (Fig. 2B–B″). By this point, cells appear to be committed to either neural or imaginal identity (Fig. 3B,D–I). Cells that will adopt leg imaginal identity can be identified by coexpression of Dll and Esg proteins (Fig. 3B,D–G). Conversely, cells that will become neural can be identified by their coexpression of Dll, Cut, and Cpo proteins (Figs. 2C–C″, 3D,E). All of the Dll-expressing cells express either the imaginal marker Esg or the neural markers Ct and Cpo (Figs. 2C–C″, 3D,E,H,). Cells expressing Dll, Ct, and Cpo undergo invagination at stage 14 (Fig. 3D,E). Thus, neural cells internalize before imaginal cells (Figs. 2C–C″, 3E). The basal cells that have Dll and low levels of Ct probably are Keilin's organ precursors. These cells express high levels of Cpo (see Fig. 2C′,C″). Although the leg disc eventually does internalize, we do not observe this movement by stage 17 (not shown).

thumbnail image

Figure 2. hth is excluded from the Keilin's organ precursors, and ct and cpo are spatially restricted within the Keilin's organ precursors. A–B″: Expression of Dll (green), Ct (red), and Hth (blue). C–C″: Expression of Dll (green), Ct (red), and Cpo (blue). A–C″: Views of a wild-type stage 11 embryo (A–A″) and wild-type stage 14 animals (B–C″). In all panels, dorsal is to the left. In A–C, anterior is to the left. In C′ and C″, apical is to the left, basal to the right. All images were collected with a ×40 objective. A–B″: Single slices. C–C″: Z-series data. C–C″: A brightest point projection of a 49-slice Z series collected at 0.4-μm intervals (C) and 85 degrees rotated brightest point projections from C taken between the dotted lines (C–C″). Note that, at stage 11, Dll and Hth are completely coincident with no visible Ct expression within the primordium. At stage 14, there is a hole in the Hth expression pattern at the center of the Dll expression domain. This finding corresponds to the Keilin's organ precursors, marked by Ct. As shown in C–C″, there are spatial differences in the expression of Ct and Cpo: whereas Ct and Cpo expression is found in all Keilin's organ precursors, the highest Ct levels are found more apically, while Cpo expression is highest more basally.

Download figure to PowerPoint

thumbnail image

Figure 3. Neural and imaginal markers are exclusively expressed in the thoracic limb primordium. A–I: Expression patterns of Dll (green), Ct (red), and Esg (blue) in wild-type Drosophila embryos at either stage 11 (A,C) or stage 14 (B,D–I). Images in A and B were obtained with a ×10 objective, whereas those in C–I were obtained with a ×40 objective. A–D,F,H: anterior is to the left and dorsal is at the top. A,B: Single focal planes. C–I: brightest point projections of 64-slice Z-series stacks collected at 0.3 μm intervals (C,D,F,H) and 85 degrees rotated brightest point projections of the stacks from D taken between the dotted lines (E,G,I). In these panels, the apical surface of the embryo is to the left and the basal surface is to the right. In A and C, note the lack of expression of either the imaginal marker Esg or the Keilin's organ marker Ct within the cluster of Dll-expressing cells. In contrast, in D and E, coexpression of Esg and Dll within the leg disc precursors at the periphery of the Dll expression domain and Dll and Ct expression in the center of the thoracic limb primordium is observed. A population of cells on the ventral side of the primordium that express Esg but not Dll is clearly visible (double arrowheads in D–I). w, wing disc primordium; h, haltere disc primordium; T1, T2, T3, first, second, and third thoracic segments, respectively.

Download figure to PowerPoint

As discussed above, lineage analysis indicates Dll expression probably does not demarcate the full extent of the thoracic primordium (Campbell and Tomlinson, 1998; Weigmann and Cohen, 1999; Andrews and Boekhoff-Falk, unpublished observations). Similar lineage studies have not been done for esg, thus we cannot be certain that esg expression marks all presumptive imaginal tissue. Nor do we know that all of the esg-expressing cells in the vicinity of the thoracic primordium contribute to imaginal discs. Nonetheless, we note that there is a small number of cells that express Esg but not Dll along the ventral edge of the primordium (arrows in Fig. 3D–I), and propose that these cells may constitute a part of the leg disc that contributes to the proximal leg and/or the adult body wall. These cells also appear to lack Btd expression (Estella et al., 2003; Fig. 1D).

Differentiation of the Keilin's Organ Requires the Activity of Dll, the ASC and ct

Dll activity is required for differentiation of several larval sensory structures (Cohen et al., 1989). These structures include the Keilin's organs on the thoracic segments. One or more members of the ASC is required for formation of the Keilin's organs because, in Df(1)sc-B57 animals, the Keilin's organs are missing (not shown). This finding is consistent with earlier work indicating that the ASC is required for formation of Keilin's organ neurons (Dambly-Chaudiere and Ghysen, 1987). The Keilin's organs also are absent in animals lacking ct activity (Wieschaus et al., 1984; Bodmer et al., 1987). There are multiple possible causes for the absence of the Keilin's organ. These causes include loss of the neural cells due to cell death, respecification of neural tissue to other fates, or a failure of the structure to properly differentiate at the transition from embryo to larva. In subsequent sections, we address the levels at which Dll, the ASC, and ct are likely to be involved in Keilin's organ differentiation.

Dll Is Required for the Expression of the Neural Markers l'sc, ac, ct, and cpo and May Repress Expression of the Imaginal Marker esg

To investigate the role of Dll in patterning the thoracic limb primordium, the expression patterns of the Keilin's organ markers ct and cpo and the imaginal marker esg were examined in Dll null embryos. To identify cells that would have expressed Dll in these mutant animals, a reporter construct expressing β-galactosidase under control of a Dll(304) enhancer was placed in a Dll null background. The number of ct-expressing (Fig. 4C,C′) and cpo-expressing (not shown) cells in DllSA1 homozygotes is substantially reduced compared with wild-type (Fig. 4A,A′). The relationship between Dll and esg in the cells where they are normally coexpressed is less clear. In DllSA1 homozygotes (Fig. 4C,C′), the number of esg-expressing cells is reduced compared with wild-type animals (Fig. 4A,A′). In contrast, animals homozygous for second Dll null allele, DllMP, show apparently normal expression of esg in the periphery of the primordium (not shown). Nonetheless, in both cases, the circular gap in the esg-expressing cells, corresponding to the Keilin's organ precursors, is filled in. This finding is consistent with the observations of Kubota et al. (Kubota et al., 2003), who also observed the loss of the hole in the esg pattern in DllMP homozygotes. These results can be explained in at least two ways. First, the Keilin's organ precursors could lose their neural specification and develop along the imaginal pathway. Alternatively, if Dll is required for cell viability, death and loss of the neural cells could cause an apparent derepression of imaginal identity in the center of the primordium.

thumbnail image

Figure 4. Loss of Dll results in loss of ct expression and may result in partial derepression of esg within the Keilin's organ precursors. A,A′,C,C′: Wild-type expression pattern of Dll and Dll(304)-β-Gal (green), Ct (red), and Esg (blue) proteins (A,A′) and of Ct (red) and Esg (blue) relative to the Dll(304) reporter (green) in a DllSA1 homozygote (C,C′). Embryos in these panels are both stage 16. Note that the number of Ct-expressing cells (asterisks) is significantly reduced in the Dll null sample. B,B′: The pattern of cell death in the primordium of a wild-type embryo. D,D′: The pattern of cell death in a DllSA1 homozygote. Both embryos in these panels are stage 14. In these panels, Dll and Dll(304)-β-Gal (B,B′) or the Dll(304)-β-Gal reporter (D,D′) is green, Ct is red, and terminal deoxynucleotidyl transferase [TdT]- mediated dUTP nick end-labeling (TUNEL) staining for apoptotic nuclei is in blue. Note that a small number of cells, marked by white arrowheads, undergo apoptosis in both wild-type and Dll null primordia. The positions of these cells are not reproducible in either genotype. All images were collected using a ×40 objective and are brightest point projections of Z-series data collected at 0.25-μm intervals. A–D′: Composites from 63 slices collected at 0.25-μm intervals (A,A′,D,D′) and Z-series from 57 slices collected at 0.25-μm intervals (B and B′) and from 69 slices at 0.10-μm intervals (C,C′).

Download figure to PowerPoint

To distinguish between these possibilities, terminal deoxynucleotidyl transferase [TdT]- mediated dUTP nick end-labeling (TUNEL) analysis was performed on embryos lacking Dll. In wild-type embryos (Fig. 4B,B′), very little cell death is observed, even at late stages. Similarly, in DllSA1 homozygotes, little cell death was observed within the normal Dll expression domain at stage 11 or until stage 16, when a few cells undergo apoptosis (Fig. 4D,D′). This finding is subsequent to the establishment of the neural component of the primordium at stage 13, as marked by ct and cpo expression. Therefore, Dll likely is required to activate ct and cpo and to repress esg in Keilin's organ cells rather than for cell viability.

Dll is required for the loss of nuclear Exd from the center of the primordium at stage 13 (Gonzalez-Crespo et al., 1998). We have observed the same relationship between Dll and hth (data not shown). This finding was expected, as Hth is required for the nuclear localization of Exd (Rieckhof et al., 1997). It was proposed that cells from which nuclear Exd (and thus Hth) is lost correspond to the presumptive distal leg. However, we find that all of the cells of the primordium from which Hth is excluded are of PNS identity and, thus, are more likely to be the precursors for the Keilin's organ (Fig. 2B–B″). In addition, all Esg-positive cells express nuclear Exd (inferred from the ring of cells at the periphery of the Dll expression domain that also express nuclear Exd [Gonzalez-Crespo et al., 1998] and Hth [Kubota et al., 2003; and this work]), supporting the idea that cells that possess nuclear Exd and Hth are imaginal. This observations indicates that Exd and Hth are expressed in both presumptive proximal and distal domains of the embryonic leg disc and raises the possibility that nuclear Exd and Hth must be lost for the formation of sense organs. Consistent with this concept, it has been noted that, in the antennal disc, Hth must initially be expressed in the distal-most cells that will give rise to the arista, but must subsequently be lost for differentiation of this structure (Dong et al., 2000). This initial expression, then loss, of Hth is very similar to the expression of Hth in the embryonic thoracic limb primordium.

Because the Dll-expressing cells give rise to cells of two distinct fates, imaginal and neural, inputs other than Dll must be required for fate determination. Given the well-characterized role of the ASC in regulating neural identity, members of the ASC expressed within the center of the thoracic limb primordium are candidates for the neuralizing input of the Keilin's organ precursors. To determine the relationship between Dll and ASC genes, the expression of l'sc and ac was examined in Dll null embryos carrying a lacZ reporter marking where Dll would have been expressed (Vachon et al., 1992). The expression of both l'sc and ac were greatly reduced (not shown), indicating that Dll is upstream of the ASC in the limb primordium.

Members of the ASC Are Necessary but Not Sufficient for Expression of the Neural Markers ct and cpo and Repression of Imaginal Identity Within the Keilin's Organ Precursors

To test whether one or more ASC genes might cooperate with Dll to establish neural identity and to repress imaginal fates, we examined the expression of ct and cpo in animals hemizygous for Df(1)sc-B57. In these animals, Keilin's organ expression of both ct (Fig. 5A,A′) and cpo (not shown) is greatly reduced at stage 13. In contrast, the imaginal marker esg appears to be derepressed in the center of the primordium at this stage (Fig. 5A,A′). Furthermore, cells were occasionally observed that expressed both neural and imaginal markers (n = 12 cells in 32 primordia examined, Fig. 5B,B′), consistent with these cells having adopted an intermediate fate. Thus, it is likely that one or more ASC genes are required for establishment of neural identity within the Keilin's organ precursors. In contrast to what we observed for Dll, the ASC is required both for establishment of neural identity and for survival of the Keilin's organ precursors. At stage 13, the number of cells in the ASC null primordium is normal based on Dll expression (Fig. 5A,A′,C,C′). At later stages, however, the number of Dll-expressing cells is reduced (Fig. 5B,B′,D,D′). When animals carrying the ASC deficiency were analyzed by TUNEL assay, no apoptotic nuclei were observed in the primordia at stage 13 (Fig. 5C,C′), despite extensive neural cell death in the PNS and VNC. By stage 15 or later, however, several apoptotic nuclei were clearly visible in the center of the primordium (Fig. 5D,D′).

thumbnail image

Figure 5. The ASC is necessary but not sufficient for both Keilin's organ identity and repression of imaginal identity. A–D′: Embryos in A,A′,C,C′ are stage 13, whereas the embryos in B,B′,D,D′ are stage 16. E,E′: The embryo is approximately stage 15. A,A′,C–D′: The data are brightest point projections of Z series data from a series of 45 sections collected at 0.4-μm intervals(A,A′), from a series of 53 sections taken at 0.25-μm intervals (C,C′), and from a series of 49 sections collected at 0.3-μm intervals (D,D′). B,B′,E,E′: Single focal sections chosen for clarity. A–B′: Expression of Dll (green), Ct (red), and Esg (blue) in Df(1)sc-B57 hemizygous embryos. A–A′: Note the loss of ct-expressing cells in the center of the primordium in the region that would give rise to the Keilin's organ and an increase in the number of esg-expressing cells in this region. B,B′: Although less common than a simple reduction in the number of ct-expressing cells, some embryos had cells that expressed both ct and esg, indicating that these cells were indistinct with respect to their identity as imaginal or Keilin's organ (arrows). C–D′: Patterns of Dll (green), Ct (red), and cell death by terminal deoxynucleotidyl transferase [TdT]-mediated dUTP nick end-labeling (TUNEL, blue) in Df(1)sc-B57 hemizygotes at stages 13 (C,C′) and 16 (D,D′). Compare these results with those for Dll null embryos in Figure 5D, D′. There is very little cell death observed in Dll null embryos, whereas a significant amount is seen in embryos lacking the ASC (arrowheads in C–D′). Note the cell in D,D′ that expresses both Dll and Ct and is dying (arrows). Together, these data indicate that the absence of cells expressing the Keilin's organ markers at stage 13 in ASC nulls is not due to cell death and that, unlike Dll, the ASC is required for cell viability. E, E′ show the expression of Dll (green), Ct (red), and Esg (blue) in an embryo expressing Sc throughout the primordium using the Dll-GAL4 driver. No expansion in the number of cells expressing Keilin's organ markers or loss of imaginal cells is observed. F: Phenotypically, animals of this genotype show the normal number of bristles in the Keilin's organ. The inset is an enlargement of the Keilin's organ from the same cuticle. KO, Keilin's organ.

Download figure to PowerPoint

To determine whether or not ASC genes were capable of converting the imaginal precursors to a neural identity, l'sc or sc was ectopically expressed throughout the primordia using Dll-GAL4. This driver led to expression of the ASC family member from stage 10 onward in cells fated to give rise to leg and wing as well as the Keilin's organ. Neither ectopic l'sc (Fig. 5E,E′) nor ectopic sc (not shown), perturbed esg expression. Similarly, neither the number of ct-expressing (Figs. 5E,E′) nor cpo- expressing (not shown) cells is detectably increased. Consistent with these results, ectopic expression of ASC family members throughout the Dll domain does not result in an increased number of sensory bristles in the Keilin's organ (Fig. 5F). Thus, members of the complex are required for establishment of neural fates within the primordium but are not sufficient to cause otherwise imaginal cells to develop along a neural path.

ct Is Required for Specification of External Sensory Class on the Keilin's Organ Neurons, but Not for Neural Identity and Does Not Repress Imaginal Identity

ct is required for the specification of ES class neurons in the developing PNS. Because ct is expressed in the presumptive Keilin's organ precursors and is required for the formation of this structure, we wished to address whether or not ct is required for either the establishment of neural identity or the repression of imaginal identity in the Keilin's organ precursors. The expression patterns of cpo and esg in male embryos hemizygous for the null ctC145 allele, therefore, were examined. In these embryos, cpo expression is normal (Fig. 6A,A′) and the expression of esg was not expanded (Fig. 6B,B′). Thus, it appears unlikely that ct is directly involved in specification of neural identity within the Keilin's organ precursors. Nonetheless, ct is required for the differentiation of this structure in the larva (Wieschaus et al., 1984; Bodmer et al., 1987). In the absence of ct, PNS cells that normally would give rise to ES class sensilla sometimes are transformed to chordotonal structures (Bodmer et al., 1987; Blochlinger et al., 1991). atonal (ato) is a proneural gene required for specification of chordotonal organs (Jarman et al., 1993). To determine whether the role of ct in the Keilin's organ precursors is to prevent the expression of ato, we stained ct null embryos for Ato. No ato expression was observed in the thoracic limb primordia of these embryos between stages 10 and 17 (n = 30 animals; data not shown). However, it has been proposed that a critical function of ato during chordotonal differentiation is to repress ct (Jarman and Ahmed, 1998; Brewster et al., 2001). Thus, loss of ct might be expected to lead to transformation of Keilin's organ cells to chordotonal identity even in the absence of ato derepression. We, therefore, examined the morphology of the Keilin's organ precursors in stage 17 ct null embryos using rhodamine–phalloidin staining to visualize actin bundles. Using this method, we did not observe (n = 45 primordia of stage 16 or later in ct null embryos) any cells of the Keilin's organ that adopted the elongated rod-like actin bundles characteristic of scolopales of chordotonal organs (Fig. 6C,C′). In these stainings, the pentascolopidial chordotonal organs in the abdominal segments of both ct null and wild-type siblings were clearly visible (Fig. 6D). This finding suggests that loss of ct is not sufficient to convert the Keilin's organ precursors to chordotonal fate either at the level of ato expression or morphologically. On the basis of Cpo expression, Keilin's organ cells of ct null embryos remain viable throughout embryogenesis and maintain PNS identity.

thumbnail image

Figure 6. Loss of ct does not cause either loss of neural identity or an obvious shift of Keilin's organ precursors to chordotonal identity. Embryos in A–B′ are stage 14, whereas the embryo in C–D is at stage 16. All are single confocal sections. A–C′ are second thoracic sections, whereas D is a third abdominal segment. A,A′: Expression of Dll (green) and Cpo (red) in a ctC145 hemizygote. The expression of both is indistinguishable from wild-type expression (compare with Fig. 3D–I). B,B′: Expression of Dll (green) and Esg (blue) in a ctC145 hemizygote. No derepression of Esg is observed in the absence of ct (compare with Fig. 3H). C,C′:ctC145 hemizygous embryo stained for Dll (green) and actin (phalloidin–purple). D: Small concretions of actin are visible in the Keilin's organ precursors (white arrowheads in C,C′) but bear no similarity to the scolopidia of the pentascolopidial organ found in the abdominal segments (bracketed by white arrows). lch5, lateral chordotonal 5 cluster (pentascolopidial organ).

Download figure to PowerPoint

Imaginal Determinant escargot Is Sufficient to Repress Keilin's Organ Development

Both Dll and members of the ASC are required for Keilin's organ fate. Yet ectopic expression of ASC members is not sufficient to convert the leg disc cells to neural identity. Thus, some factor or constellation of factors could be present in the imaginal cells to prevent this from occurring. Given its previously described role in specifying imaginal tissues, a likely inhibitory factor is esg. To test whether esg can prevent neural differentiation, esg was expressed throughout the Dll domain using the Dll-GAL4 driver. In these animals, expression of the Keilin's organ markers ct (not shown) and cpo (Fig. 7A,A′) were both markedly reduced. The loss of ct and cpo expression may be direct or by means of repression of the ASC family members, as earlier l'sc expression in the primordia of these animals also is severely reduced (Fig. 7B,B′). Consistent with the loss of Keilin's organ marker expression, cuticles from animals ectopically expressing esg throughout the Dll domain lack Keilin's organs (Fig. 7C).

thumbnail image

Figure 7. Ectopic expression of esg is sufficient to repress Keilin's organ formation and perturbs the morphology of the late leg/Keilin's organ primordium. In all panels, anterior is to the left and dorsal is up. A,A′: Expression of Cpo (blue) and Dll (green) in the primordium of a stage 16 embryo in which Esg has been uniformly expressed using the Dll-GAL4 driver. These images are a brightest point projection from a Z-series of 68 images collected at 0.3-μm intervals. Note the smaller number of Cpo-expressing cells (asterisks) compared with those in Fig. 2C–C″. B,B′: Expression of L'sc (blue) and Dll (green) in a stage 11 embryo expressing Esg using the Dll-GAL4 driver. Note the absence of any cells expressing L'sc compared with the normal complement of eight cells shown in the wild-type embryo in Fig. 1D and D′. Loss of Cpo (A,A′) and Ct (not shown) expression probably results from loss of ASC expression. The shape of the primordium at stage 11 versus stage 15 (A–B′) is quite different. Note the normal shape of the primordium at stage 11 (B, B′), and the disorganization at later stages (A,A′). The level of Dll expression is also lower than normal at later stages: the gain on the Dll channel in A is approximately five times higher than required to visualize it in a wild-type embryo. C: View of the ventral left T3 segment of a first instar larva in which Esg was driven throughout the primordium using Dll-GAL4. Note the absence of the Keilin's organ (white arrow), while the ventral black dot (in which the Dll-GAL4 driver is not expressed) is still present. VB, ventral black dot.

Download figure to PowerPoint

Whereas the size and shape of the thoracic limb primordia is normal at early stages in the Dll-GAL4; UAS-esg animals, the size of the leg/Keilin's organ primordium appears to be reduced in these animals by stage 13 (compare the Dll stainings of the embryo in Fig. 7A,A′ with that of the stage 11 embryo in 7B,B′). This reduction in size also was observed in the Dll-expressing head limb primordia (not shown) and is in contrast to what has been reported previously (Kubota et al., 2003). The wild-type levels of Esg protein expression in the wing and haltere discs are significantly higher than in the leg/Keilin's organ primordium. We, therefore, tested whether ectopic esg expression caused the leg/Keilin's organ cells to be respecified to wing or haltere identity. We stained Dll-GAL4; UAS-esg embryos for both Dll and Snail (Sna) to mark the leg/Keilin's organ and wing/haltere precursors, respectively. The number of sna-expressing cells was not detectably increased in animals with ectopic esg (data not shown), indicating that these cells are unlikely to have been shifted to wing and haltere fates.

Consistent with the ectopic expression results, loss of esg leads to an apparent increase in the number of Keilin's organ precursors (not shown). However, phenotypically, the Keilin's organs of esg null larvae are almost always (n = 43/47 Keilin's organs) indistinguishable from wild-type, although occasionally (n = 4/47 Keilin's organs), there is a small increase in the number of bristles (to four or five bristles from the wild-type 3). Not all Keilin's organs could be scored in these cuticle preparations due to orientation. TUNEL analysis of esg null embryos reveals no increase in apoptosis with the primordium, even at the end of embryogenesis (not shown).

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Thoracic Limb Primordium Arises in the Neurectoderm and Is Divided Into Distinct Domains That Will Give Rise to Imaginal and Neural Derivatives

An early marker for the thoracic limb primordium is the homeodomain transcription factor Dll. Dll is first detectable at late stage 10. Between stages 12 and 13, the population of Dll-expressing cells is subdivided into domains that express either neural or imaginal markers. The neural derivative gives rise to the Keilin's organ, whereas the imaginal components give rise to the leg and wing discs. Using the svp and ASC neural markers, we found that the ventral half of the primordium arises within the neurogenic ectoderm. Because the dorsal-most cells of the primordium are those that become the wing disc, the portion of the primordium that lies outside of the neurogenic ectoderm likely gives rise to the wing disc. Cells that lie within the neurectoderm probably give rise to the leg disc and Keilin's organ. Based on analysis of Dll expression in several phyla, it has been proposed that the earliest role of Dll was in the development of the nervous system and that its role in patterning the proximodistal axes of appendages evolved later (Panganiban et al., 1997). The relationship between the developing nervous system and the appendages in more primitive organisms than Drosophila is unknown, but it may be that the relationship between these structures is of broader evolutionary significance.

Not all Dll-expressing cells of the primordium are born at the same time. Instead, cells that activate the early embryonic Dll enhancer de novo are produced continuously between stages 10 and 12 in the ventral portion of the primordium (Goto and Hayashi, 1997a). We have shown here that the Dll-producing domain lies within the neurogenic ectoderm. Dll-expressing cells subsequently migrate dorsally. As they move away from this hypothetical “source,” they cease utilization of the Dll early embryonic enhancer and either switch to the late embryonic enhancer or cease transcription of Dll entirely. The latter population gives rise to the wing disc, whereas the former gives rise to the leg disc and Keilin's organ. The location of the primordium at the boundary between the neurectoderm and non-neural ectoderm and the early expression of Dll in the migrating cells resembles Dlx expression in the specification and migration of neural crest cells in vertebrates. Dlx family members are among the earliest markers for the neural crest (reviewed in Merlo et al., 2000). A second intriguing parallel between the Drosophila thoracic limb primordium and the vertebrate neural crest is that the Esg-related transcription factors Snail and Slug have been implicated in the specification and migration of the neural crest, whereas Sna and Esg have been implicated in the development of the wing cells that migrate away from the leg disc and neuroectoderm (reviewed in Hemavathy et al., 2000). We, therefore, propose that the vertebrate neural crest and the thoracic imaginal discs may be derived from a common set of cells present in the last common ancestor of protostomes and deuterostomes.

Fate Mapping of the Thoracic Limb Primordium

Fate maps of the leg/Keilin's organ limb primordium have been proposed (Goto and Hayashi, 1997b; Kubota et al., 2003). However, these maps do not distinguish between distal leg and Keilin's organ precursors. Our data indicate that the primordium has three distinct populations of cells that are likely to correspond to proximal leg disc, distal leg disc, and Keilin's organ (Fig. 8A). Previous models include specification of proximal leg disc at an intermediate dorsoventral position within the early primordium, and their subsequent ventral migration to encircle the cells that give rise to the distal leg (Goto and Hayashi, 1997b). We have not obtained evidence for this ventral migration in our own experiments. Instead, we observe esg expression as a discrete ring in the primordium by stage 14.

thumbnail image

Figure 8. Proposed fate map of the thoracic limb primordium and summary of genetic interactions involved in its subdivision. A: A summary of the fate map of the thoracic limb primordium of the second thoracic segment. The dorsal-most derivative of the primordium is the wing disc (blue). The more ventral portions of the primordium give rise to two structures, the leg disc (green and purple) and the Keilin's organ (red). The Keilin's organ precursors arise centrally and express Dll, Ct, and Cpo but lack the imaginal determinant Esg. The leg disc precursors may be further subdivided into the cells giving rise to the presumptive distal (green) and putative proximal (purple) domains of the leg imaginal disc. Distal leg disc cells are marked by their coexpression of Dll and Esg proteins (green), whereas proximal cells express only Esg (purple). B: The genetic relationships between the various genes involved in specifying the thoracic limb primordium. The wg and dpp pathways serve key roles in specification of the primordium through initiating the expression of determinants such as Dll and esg (Cohen, 1990; Goto and Hayashi, 1997b). Graded levels of these factors probably serve to subdivide these structures. High levels of Dpp activity (shown with the solid arrow) are required for specification of the wing disc and lower levels (shown with the dashed arrow) are required for specification of the leg disc (Raz and Shilo, 1993; Goto and Hayashi, 1997b; Kubota et al., 2000). Dpp dorsally limits primordium formation (Goto and Hayashi, 1997b). After specification, subdomains are patterned through the activities of several genes. The Keilin's organ is patterned by the activities of Dll and the ASC, which also prevent these cells from adopting imaginal identity. Members of the ASC are under Dll control within the Keilin's organ, and it is unclear whether Dll plays a direct role in repressing imaginal development or works only by means of the ASC (question mark over the arrow between Dll and esg). Imaginal components are patterned by esg (proximal and distal leg) and esg, vg, and sna (wing disc; Hayashi et al., 1993; Fuse et al., 1996; and this work). esg also is involved in preventing imaginal cells from adopting neural identity. Dll patterns the distal part of the leg disc but is not required for specification of imaginal identity. At later stages, ct is expressed in the Keilin's organ precursors under the control of Dll and the ASC. ct is required for establishment of external sensory structures (Wieschaus et al., 1984; Blochlinger et al., 1988; Blochlinger et al., 1990; and this work), and cpo is required for neural function (Bellen et al., 1992).

Download figure to PowerPoint

Lineage tracing experiments of Dll-expressing cells suggest that not all of the cells of the leg imaginal disc have expressed Dll (Campbell and Tomlinson, 1998; Weigmann and Cohen, 1999; Andrews and Boekhoff-Falk, unpublished observations). We also have noted the presence of a small cluster of cells that express esg but not Dll on the ventral side of the primordium. As Dll is not required for formation of proximal leg structures or leg disc-derived body wall (Cohen et al., 1993; Campbell and Tomlinson, 1998), we propose that these cells that express esg and hth but not Dll will give rise to proximal leg disc-derived structures, whereas those that express esg, hth, and Dll give rise to the distal-most portions of the leg disc, with medial domains being specified during larval development. This explanation is consistent with the timing of activation of the medial patterning gene dachshund (dac) during larval development (Mardon et al., 1994; Lecuit and Cohen, 1997).

A major difference between our proposed fate map and those put forth previously is the placement of presumptive distal leg cells within the esg-expressing population (Fig. 8A). We think this strategy is justified for two reasons. First, esg has been implicated in the maintenance of diploidy of imaginal cells (Hayashi et al., 1993). Thus, cells that lack esg expression are unlikely to be imaginal. Second, the central cells that express Dll but lack Esg express ASC. In ASC mutants, these central cells die (this work). Yet mutating the ASC yields the full complement of distal leg segments (Garcia-Bellido and Santamaria, 1978). Thus, the central ASC-expressing cells are unlikely to contribute to the leg imaginal disc. Instead, they are probably Keilin's organ precursors. Consistent with this view, these central cells express Dll and ct as well as ASC genes, and the Keilin's organ is lost in animals mutant for any of these (Wieschaus et al., 1984; Bodmer et al., 1987; Dambly-Chaudiere and Ghysen, 1987; Cohen and Jurgens, 1989; and this work).

A second difference in our proposed fate map is the placement of presumptive proximal leg outside of the Dll expression domain. We believe this positioning is justified based on both Dll phenotypes and lineage analysis and on the requirement for esg in imaginal tissues. Specifically, Dll is not required for the formation of proximal leg disc-derived structures (Cohen et al., 1993) nor has lineage analysis revealed Dll expression in proximal leg disc-derived structures at any stage of development (Campbell and Tomlinson, 1998; Weigmann and Cohen, 1999; Andrews and Boekhoff-Falk, unpublished observation). Thus, the proximal leg disc is likely to arise from cells that express esg but lack Dll. Such cells can be found ventral to the Dll-expressing population.

There are two caveats to our proposed fate map. First, esg lineage data are not available. Thus, we do not know whether all esg-expressing cells of the primordium become imaginal. Nor do we know whether esg expression marks all of the presumptive imaginal cells. Second, although the cell death observed in ASC mutants together with their normal distal leg segmentation suggest that ASC is not required to make the distal leg, it is possible that distal leg normally arises from ASC-expressing cells and that these cells are regenerated in the mutants. Nonetheless, we believe the available data strongly support the fate put forth here.

Fate Determination Within the Thoracic Limb Primordium Requires the Activities of Dll, the ASC, ct, and esg

This study constitutes the first genetic analysis of the roles of Drosophila Dll in a component of the developing embryonic nervous system. Larvae lacking Dll, the ASC or ct are missing Keilin's organ. In embryos null for either Dll or the ASC, the number of cells in the thoracic primordia expressing ct and cpo is significantly reduced and the number of cells expressing esg appears to increase. Therefore, Dll and the ASC function upstream of ct and cpo in specification of the neural portion of the primordium and are required to repress imaginal development in these cells. Nonetheless, Dll and the ASC are not sufficient to convert imaginal precursors to Keilin's organ fate. Although members of the ASC are required for cell survival in late embryonic development, the loss of Keilin's organ cells in ASC nulls appears to be independent of cell death, because Keilin's organ marker expression is reduced well before the onset of cell death.

We note that, although Dll is upstream of the ASC in the Keilin's organ precursors, loss of Dll results in less apoptosis than does loss of the ASC. We hypothesize that this finding could be due to Dll functioning nearer the top of the genetic hierarchy in specification of these cells. In the absence of Dll, cells may more readily adopt a new fate than they do when their fate specification is interrupted later by loss of the ASC. When interrupted later, it is possible that these cells become “confused” as to their fate and become targeted for removal by means of cell death. Consistent with this explanation, cells were observed in the ASC null embryos that expressed both Keilin's organ and imaginal markers. This finding was not observed in Dll null embryos. It may be that targeting of mixed fate cells for death is a common phenomenon in developmental biology. A recent study of the otic placode in zebrafish noted that loss of the later-expressed pax genes resulted in a more severe phenotype than did loss of dlx, even though dlx-3 and dlx-7 are both required for formation of the placode (Solomon and Fritz, 2002) and probably upstream of pax. The cause of the more severe phenotype in the pax mutants was found to be due to cell death, consistent with our model.

Dll is required for both ASC and ct expression. Whether Dll is required simply for ASC expression or also is required later in parallel with the ASC for ct and cpo expression is unclear. The Dll- and ASC-dependent repression of esg appears to be limited to the Keilin's organ precursors because ectopic expression of ASC family members within the thoracic limb primordium does not repress esg, expand the zone of ct and cpo expression, or increase the number of sensory hairs in the differentiated Keilin's organ. Other factors, therefore, are likely to be required. Dll collaboration with ASC could constitute an ancient regulatory mechanism, because the vertebrate Dlx1 and Dlx2 genes cooperate with ASC homologs to regulate formation of γ-aminobutyric acid-ergic neurons in the developing brain (Letinic et al., 2002).

In the absence of ct, the thoracic limb primordium expresses other Keilin's organ markers, including Dll and Cpo. This finding indicates that ct plays a relatively late role in differentiation of the Keilin's organ. ct may specify Keilin's organ precursors as ES class. However, the mechanism appears to be more complicated than a simple choice between ES and chordotonal identity, because ato is not derepressed in ct nulls, nor do the Keilin's organs adopt chordotonal morphologies.

What is the function of the Keilin's organ and the reason for its association with the developing leg imaginal disc? The Keilin's organ remains associated with the leg disc throughout larval development, and it is thought that axons from the Keilin's organ to the PNS serve pathfinding functions for later-developing leg neuron axons (Jan et al., 1985; Tix et al., 1989). Enervation of the regenerating limb of the newt precludes distal patterning, and Dlx3 expression is dependent on limb innervation (Mullen et al., 1996). Thus, Dll and Dlx could serve related or analogous roles in axon-induced limb patterning in both flies and vertebrates. As the Keilin's organ cells invaginate before leg disc cells, it is possible that the neural cells assist the migration of the presumptive leg disc to its final position within the larva. The Keilin's organ, thus, may play roles both in development of the leg disc and in differentiation and patterning of the adult leg.

esg is required for the maintenance of diploidy of at least some imaginal tissues (Fuse et al., 1994; Hayashi, 1996). It has been hypothesized that esg may be the critical imaginal determinant (Hayashi et al., 1993). Within the thoracic limb primordium, the expression patterns of esg and the neural markers ct and cpo are mutually exclusive. Thus, either esg is not required to make distal leg, or ct and cpo mark only the Keilin's organ precursors. In the absence of esg, we observed some expansion of ct or cpo domains, indicating that esg may play a role in repressing Keilin's organ fates. However, the Keilin's organ of esg null animals is not increased in size, and TUNEL analysis indicates that there is no extra cell death in esg mutant primordia.

Consistent with the loss of function analysis, ectopic expression of esg is capable of repressing ct and cpo in the presumptive Keilin's organ. The effect of ectopic esg is probably due to repression of ASC genes because the expression of l'sc is severely reduced in the thoracic limb primordium at stage 11 in embryos expressing ectopic esg. esg and other snail family members have the ability to bind to the E-box recognized by bHLH transcription factors such as the ASC proteins (Hayashi et al., 1993; Fuse et al., 1999). Also, the Scutoid mutation in Drosophila, in which there is a loss of sensory bristles from the scutellum, was shown recently to be a gain of function mutation of snail (Fuse et al., 1999). Thus, esg repression of ct and cpo may be mediated by competition with ASC products for E-box binding sites.

There appears to be a developmental switch at play in the developing primordium (Fig. 8B). Cells within the primordium have the potential to adopt either neural or imaginal identity. Within the imaginal precursors, we propose that esg is required for choosing imaginal identity, whereas within the Keilin's organ precursors, repression of esg is essential for neural differentiation to proceed. How, then, are these different potentials established? It has been shown that the imaginal components of the second thoracic primordium, the wing and leg discs, are specified in part due to opposing gradients of Dpp and EGF activity (Kubota et al., 2000). It is possible that high levels of EGF signaling could promote Keilin's organ development. Consistent with this possibility, a high level of rhomboid transcript is accumulated in a cluster of cells in the center of the primordium at stage 11 (Kubota et al., 2000). This accumulation appears to be in approximately the same domain where l'sc is expressed at this stage. Rhomboid is required for the activation of spitz, an activating ligand for the DER receptor (Schweitzer et al., 1995; Wasserman and Freeman, 1998). Thus, it is possible that expression of the ASC in the Keilin's organ precursors is dependent on up-regulation of the EGF pathway in these cells. These precursors, in turn, may maintain neural identity and repress imaginal development. Outside of this domain, cells may adopt either dorsal or ventral imaginal disc identity in response to the relative levels of Dpp and EGF signaling. However, we do not observe marker expression for the three derivatives of the primordium in a pattern directly consistent with this model. Nor does this model of opposing signaling gradients along the dorsoventral axis explain how the Keilin's organ becomes specified as a cluster of cells in the center of the thoracic limb primordium. Clearly, questions still remain with respect to the mechanisms underlying subdivision of the primordium. Examination of the expression patterns of imaginal versus Keilin's organ markers in animals mutant for components of the Dpp, Wg, and EGF signaling pathways could provide additional insights into these intriguing cell fate decisions.

EXPERIMENTAL PROCEDURES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Flies Strains Used

(1) Oregon R flies; (2) DllSA1; Dll(304). The Dll(304) is a P-element carrying the Dll early embryonic enhancer driving β-galactosidase and was used to mark the primordium in the absence of Dll expression (Vachon et al., 1992); (3) DllMP (Sunkel and Whittle, 1987; Cohen et al., 1989); (4) Df(1)sc-B57. This deficiency removes the entire ASC from achaete through asense (Dominguez and Campuzano, 1993); (5) P{ry+t7.2 = PZ}svp07842 (referred to as svp-lacZ hereafter) UAS-l'sc (Hinz et al., 1994); (6) UAS-sc (Chien et al., 1996); (7) ctC145 (Jackson and Blochlinger, 1997; Micchelli et al., 1997); (8) UAS-ct (Ludlow et al., 1996); (9) esgG66B (Whiteley et al., 1992); (10) UAS-esg (Goto and Hayashi, 1999); (11) P{w+mC = GawB}Dllmd23 (referred to as Dll-GAL4 hereafter; Calleja et al., 1996); (12) CyO-wg-lacZ balancer was used to identify esg nulls and for gain of function experiments; and (13) FM7c, P{w+mC = Gal4-Kr.C}DC1, P{w+mC = UAS-GFP.S65T}DC5 and CyO, P{w+mC = Gal4-Kr.C}DC1, P{w+mC = UAS-GFP.S65T}DC7 balancers (Casso et al., 1999) carry Kr-GAL4 and UAS-GFP (green fluorescent protein) insertions were used for identification of null larvae in cuticle preparation experiments. GFP.S65T is an engineered form of GFP that has enhanced fluorescence.

Embryo Collection and Fixation

Embryos were collected for 6 hr on molasses egg lay caps, then aged for 6 hr in a humidified chamber at 25°C. This process resulted in an average range of late stage 10 to stage 17 (Campos-Ortega and Hartenstein, 1985). Animals to be used for TUNEL analysis were fixed immediately, whereas those for antibody staining were refrigerated for up to 6 hr before processing. Embryos were collected and processed as described in (Langeland, 1999), with some modifications. The concentration of formaldehyde was reduced to 4% (v/v). After two rinses in 90% methanol, 50 mM ethyleneglycoltetraacetic acid; pH 7.0, embryos were rinsed twice with absolute ethanol and rehydrated through a graded series of ethanol:phosphate buffered saline (PBS) mixtures (75%, 50%, and 25% ethanol; 5 min for each wash). They were then washed two times for 5 min each in PT (1× PBS, 0.3% (v/v) Triton X-100; pH 7.0) and used for staining.

Immunohistochemistry

Primary antibodies and dilutions used were as follows: rabbit anti–Distal-less affinity purified antibody (Panganiban et al., 1995), 1:100; mouse anti–Distal-less ascites (Duncan et al., 1998), 1:200; guinea pig anti-Homothorax serum (Abu-Shaar et al., 1999), 1:500; rat anti-Lethal of Scute serum (Martin-Bermudo et al., 1991), 1:200; mouse anti-Achaete hybridoma supernatant (1:2; Skeath and Carroll, 1991); guinea pig anti-Atonal (Hassan et al., 2000), 1:500; mouse anti-Cut hybridoma supernatant (1:20; Blochlinger et al., 1990); rabbit anti-Couch potato affinity purified antibody (Bellen et al., 1992), 1:1000; mouse anti-Escargot ascites (unpublished and generously provided by S. Hayashi), 1:200; mouse anti-β-galactosidase affinity purified monoclonal antibody (Promega), 1:100; rabbit anti–β-galactosidase affinity purified antibody (ICN), 1:2000. Secondary antibodies (affinity purified goat anti-primary host species or biotinylated donkey anti-primary host species) and tertiary antibodies (cyanine dye conjugated donkey anti-goat or streptavidin) were from Jackson Laboratories. Alexa Fluor–conjugated goat anti-primary host species (Molecular Probes) were used in some experiments.

The general protocol for immunostaining was as described in Langeland (1999), except that a high-salt PBT (2× PBS, 4% [w/v] bovine serum albumin [Roche Biochemicals], 0.3% [v/v] Triton X-100, 0.1% [w/v] NaN3, 0.02% [w/v] thimerosol) was used for blocking and primary antibody incubations. Embryos were mounted in Vectashield (Vector Labs) and visualized on a Bio-Rad MRC1024 confocal fitted to a Zeiss Axiophot compound microscope. Image manipulations were performed using Adobe Photoshop and Object Image. Null embryos were identified by the absence of expression of the gene under investigation (ct or Dll), by absence of a wg-lacZ-marked balancer (esg) or by the previously reported abnormalities in the CNS and other parts of the PNS (ASC; Bellen et al., 1992).

For certain antibodies (mouse anti-Dll, rat anti-L'sc, mouse anti-Ac, guinea pig anti-Ato, and mouse anti-Esg), amplification of the signal was required. This amplification was done using the Vector ABC kit, following the manufacturer's instructions with some modifications. Instead of using diaminobenzidine as a substrate, biotinylated tyramide (NEN) was used and allowed to develop for 20 min before washing and detection with cyanine-5–conjugated streptavidin.

In some cases, it was necessary to use two primary antibodies from the same host species in a single experiment. To avoid cross-reactivity, the weaker of the two was applied first, subjected to ABC/tyramide signal amplification, and removed by rinsing briefly three times in stripping buffer (0.2 M glycine–HCl, 0.3% Triton X-100; pH 2.2) and twice for 10 min each in stripping buffer. After neutralization by washing in three changes of PT over 30 min, embryos were reblocked and the other primary antibodies were applied and processed as described above.

Combined Antibody/TUNEL Staining of Embryos

Primary antibodies were added and washed off as described above. To protect the antibodies, they were post-fixed onto the embryos with 4% formaldehyde in PT for 15 min before proceeding with TUNEL analysis. Embryos were then permeabilized by treatment with proteinase K (19 μg/ml in PT) for 1.5 min. Proteinase treatment was stopped by washing the embryos in three washes over 5 min with 0.2% (w/v) glycine in PT. Embryos were then washed in three changes of PT over 5 min and post-fixed in 4% formaldehyde for 15 min at room temperature then washed again in PT. Embryos were equilibrated into 1× TdT working buffer without enzyme or dUTP added (1× TdT buffer, 1× CoCl2, 0.3% Triton X-100; Roche Biochemicals) for 15 min with gentle rocking. They were then transferred to 70 μl of the full TdT reaction cocktail (same as the equilibration mixture but containing 25 units of TdT and 10 μM of 1:2 mixture of biotinylated-dUTP:dUTP [Roche Biochemicals]). This mixture was allowed to incubate with gentle shaking at 37°C for 3 to 8 hr. The reaction was terminated by washing in six changes of PT over an hour. Secondary and tertiary antibodies were used as previously described to detect the primary antibody, and cyanine-5–conjugated streptavidin was used to detect the TUNEL-positive cells. Images were collected and processed as described under Immunohistochemistry.

Preparation of Larval Cuticles

Fly strains carrying null alleles of the gene in question over the appropriate Kr-GAL4, UAS-GFP balancer were allowed to lay for 1 hr at 25°C, and the embryos were allowed to develop overnight at 25°C in a humidified chamber. Animals were dechorionated in 50% (v/v) Clorox as previously described, and unhatched larvae lacking the green balancer were identified by examining the animals on a Leica Spot HBO 100 system. These animals manually were removed from their vitelline membranes with a tungsten needle and fixed by incubating overnight in a 4:1 mixture of glacial acetic acid:glycerol at 60°C. Cuticles were rinsed several times with PBS and mounted in a 1:1 mixture of Hoyer's:85% lactic acid. After drying for 3 days at 65°C, cuticles were examined on a Zeiss Axioplan 2 microscope fitted with a Zeiss AxioCam. For ectopic expression experiments, Dll-GAL4 over the CyO green balancer was used as driver. Images were processed using Adobe Photoshop.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

We thank the following individuals for their generous gifts of reagents and input into this work: Dr. Shigeo Hayashi for sharing an unpublished monoclonal anti-Esg antibody as well as esg fly stocks used in this study; Dr. Seth Blair for providing ctC145 and UAS-ct fly stocks; Dr. Fernando Jimenez for providing the rat anti-L'sc antibody; Dr. James Skeath for providing the svp-lacZ, UAS-l'sc and UAS-sc fly stocks, and helpful input into the work; Dr. Hugo Bellen for the guinea pig anti-Ato antibody; Dr. Ian Duncan for the monoclonal anti-Dll antibody; Dr. Sean Carroll for the anti-Sna antibody and the use of his confocal microscope. The authors also are grateful to two anonymous reviewers for insightful comments. The anti-Ct and 22C10 antibodies used in these studies were obtained from the Developmental Studies Hybridoma Bank developed under the auspices of the NICHD and maintained by the University of Iowa, Department of Biological Sciences (Iowa City, IA). R.B. is a trainee of the Molecular Biosciences Training Grant and the recipient of a Cremer fellowship. G.B.-F. is the recipient of a Young Investigator Award in Molecular Studies of Evolution from the Sloan Foundation and the National Science Foundation. G.B.-F was funded by the NIH.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  • Abu-Shaar M, Ryoo HD, Mann RS. 1999. Control of the nuclear localization of Extradenticle by competing nuclear import and export signals. Genes Dev 13: 935945.
  • Bellen HJ, Kooye S, D'Evelyn D, Pearlman J. 1992. The Drosophila Couch Potato protein is expressed in nuclei of peripheral neuronal precursors and shows homology to RNA-binding proteins. Genes Dev 6: 21252136.
  • Blochlinger K, Bodmer R, Jack J, Jan LY, Jan YN. 1988. Primary structure and expression of a product from cut, a locus involved in specifying sensory organ identity in Drosophila. Nature 333: 629635.
  • Blochlinger K, Bodmer R, Jack J, Jan LY, Jan YN. 1990. Patterns of expression of cut, a protein required for external sensory organ development in wild-type and cut mutant Drosophila embryos. Genes Dev 4: 13221331.
  • Blochlinger K, Jan LY, Jan YN. 1991. Transformation of sensory organ identity by ectopic expression of Cut in Drosophila. Genes Dev 5: 11241135.
  • Bodmer R, Barbel S, Sheperd S, Jack JW, Jan LY, Jan YN. 1987. Transformation of sensory organs by mutations of the cut locus of D. melanogaster. Cell 51: 293307.
  • Brand M, Jarman AP, Jan LY, Jan YN. 1993. asense is a Drosophila neural precursor gene and is capable of initiating sense organ formation. Development 119: 117.
  • Brewster R, Hardiman K, Deo M, Khan S, Bodmer R. 2001. The selector gene cut represses a neural cell fate that is specified independently of the Achaete-scute-Complex and atonal. Mech Dev 105: 5768.
  • Calleja M, Moreno E, Pelaz S, Morata G. 1996. Visualization of gene expression in living adult Drosophila. Science 274: 252255.
  • Campbell G, Tomlinson A. 1995. Initiation of the proximodistal axis in insect legs. Development 121: 619628.
  • Campbell G, Tomlinson A. 1998. The roles of the homeobox genes aristaless and Distal-less in patterning the legs and wings of Drosophila. Development 125: 44834493.
  • Campos-Ortega JA, Hartenstein V. 1985. The embryonic development of Drosophila melanogaster. Berlin: Springer-Verlag. p 227.
  • Casares F, Mann RS. 1998. Control of antennal versus leg development in Drosophila. Nature 392: 723726.
  • Casso D, Ramirez-Weber F, Kornberg TB. 1999. GFP-tagged balancer chromosomes for Drosophila melanogaster. Mech Dev 88: 229232.
  • Chien CT, Hsiao CD, Jan LY, Jan YN. 1996. Neuronal type information encoded in the basic-helix-loop-helix domain of proneural genes. Proc Natl Acad Sci U S A 93: 1323913244.
  • Cohen SM. 1990. Specification of limb development in the Drosophila embryo by positional cues from segmentation genes. Nature 343: 173177.
  • Cohen S. 1993. Imaginal disc development. In: BateM, Martinez-AriasA, editors. The development of Drosophila melanogaster. Vol. 2. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press. p 747841.
  • Cohen SM, Jurgens G. 1989. Proximal-distal pattern formation in Drosophila: cell autonomous requirement for Distal-less gene activity in limb development. EMBO J 8: 20452055.
  • Cohen S, Bronner M, Kuttner F, Jurgens G, Jackle H. 1989. Distal-less encodes a homeodomain protein required for limb formation in Drosophila. Nature 338: 432434.
  • Cohen B, Simcox AA, Cohen SM. 1993. Allocation of the thoracic imaginal primordia in the Drosophila embryo. Development 117: 597608.
  • Dambly-Chaudiere C, Ghysen A. 1987. Independent subpatterns of sense organs require independent genes of the achaete-scute complex in Drosophila larvae. Genes Dev 1: 297306.
  • Dominguez M, Campuzano S. 1993. asense, a member of the Drosophila achaete-scute complex, is a proneural and neural differentiation gene. EMBO J 12: 20492060.
  • Dong PD, Chu J, Panganiban G. 2000. Coexpression of the homeobox genes Distal-less and homothorax determines Drosophila antennal identity. Development 127: 209216.
  • Duncan DM, Burgess E, Duncan I. 1998. Control of distal antennal identity and tarsal development in Drosophila by spineless-aristapedia, a homolog of the mammalian dioxin receptor. Genes Dev 12: 12901303.
  • Estella C, Rieckhof G, Calleja M, Morata G. 2003. The role of buttonhead and Sp1 in the development of the ventral imaginal discs of Drosophila. Development 130: 59295941.
  • Fuse N, Hirose S, Hayashi S. 1994. Diploidy of Drosophila imaginal cells is maintained by a transcriptional repressor encoded by escargot. Genes Dev 8: 22702281.
  • Fuse N, Hirose S, Hayashi S. 1996. Determination of wing cell fate by the escargot and snail genes in Drosophila. Development 122: 10591067.
  • Fuse N, Matakatsu H, Taniguchi M, Hayashi S. 1999. Snail-type zinc finger proteins prevent neurogenesis in Scutoid and transgenic animals of Drosophila. Dev Genes Evol 209: 573580.
  • Garcia-Bellido A, Santamaria P. 1978. Developmental analysis of the achaete-scute system of Drosophila melanogaster. Genetics 88: 469486.
  • Gonzalez-Crespo S, Abu-Shaar M, Torres MC, Martinez-Arias A, Mann RS. 1998, Antagonism between extradenticle function and Hedgehog signalling in the developing limb. Nature 394: 196199.
  • Goodman CS, Doe CQ. 1993. Embryonic development of the Drosophila central nervous system. In: BateM, Martinez-AriasA, editors. The development of Drosophila melanogaster. Vol. 2. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press. p 11311206.
  • Goto S, Hayashi S. 1997a. Cell migration within the embryonic limb primordium of Drosophila as revealed by a novel fluorescence method to visualize mRNA and protein. Dev Genes Evol 207: 194198.
  • Goto S, Hayashi S. 1997b. Specification of the embryonic limb primordium by graded activity of Decapentaplegic. Development 124: 125132.
  • Goto S, Hayashi S. 1999. Proximal to distal cell communication in the Drosophila leg provides a basis for an intercalary mechanism of limb patterning. Development 126: 34073413.
  • Hassan B, Bermingham N, He Y, Sun Y, Jan Y, Zoghbi H, Bellen H. 2000. atonal regulates neurite arborization but does not act as a proneural gene in the Drosophila brain. Neuron 25: 549561.
  • Hayashi S. 1996. A Cdc2 dependent checkpoint maintains diploidy in Drosophila. Development 122: 10511058.
  • Hayashi S, Hirose S, Metcalfe T, Shirras AD. 1993. Control of imaginal cell development by the escargot gene of Drosophila. Development 118: 105115.
  • Hemavathy K, Ashraf S, Ip Y. 2000. Snail/Slug family of repressors: slowly going into the fast lane of development and cancer. Gene 257: 112.
  • Hinz U, Giebel B, Campos-Ortega JA. 1994. The basic-helix-loop-helix domain of Drosophila Lethal of scute protein is sufficient for proneural function and activates neurogenic genes. Cell 76: 7787.
  • Jackson SM, Blochlinger K. 1997. cut interacts with Notch and protein kinase A to regulate egg chamber formation and to maintain germline cyst integrity during Drosophila oogenesis. Development 124: 36633672.
  • Jan YN, Ghysen A, Christoph I, Barbel S, Jan LY. 1985. Formation of neuronal pathways in the imaginal discs of Drosophila melanogaster. J Neurosci 5: 24532464.
  • Jarman AP, Ahmed I. 1998. The specificity of proneural genes in determining Drosophila sense organ identity. Mech Dev 76: 117125.
  • Jarman AP, Grau Y, Jan LY, Jan YN. 1993. atonal is a proneural gene that directs chordotonal organ formation in the Drosophila peripheral nervous system. Cell 73: 13071321.
  • Keilin D. 1911. Comptes Rendus Hebdomadaires des Seances de l'Acadamie des Sciences. Vol. 153. Paris: Gauthier-Villars, Imprimeur Libraire. p 977979.
  • Keilin D. 1915. Bulletin Scientifique de la France et de la Belgique. Vol. 49. London: Dulau and Co. p 15198.
  • Kojima T, Sato M, Saigo K. 2000. Formation and specification of distal leg segments in Drosophila by dual Bar homeobox genes, BarH1 and BarH2. Development 127: 769778.
  • Kubota K, Goto S, Eto K, Hayashi S. 2000. EGF receptor attenuates Dpp signaling and helps to distinguish the wing and leg cell fates in Drosophila. Development 127: 37693776.
  • Kubota K, Goto S, Hayashi S. 2003. The role of Wg signaling in the patterning of embryonic leg primordium in Drosophila. Dev Biol 257: 117126.
  • Langeland JA. 1999. In: PaddockSW, editor. Confocal microscopy methods and protocols. Vol. 122. Totowa: Humana Press. p 167172.
  • Lecuit T, Cohen SM. 1997. Proximal-distal axis formation in the Drosophila leg. Nature 388: 139145.
  • Letinic K, Zoncu R, Rakic P. 2002. Origin of GABAergic neurons in the human neocortex. Nature 417: 645649.
  • Ludlow C, Choy R, Blochlinger K. 1996. Functional analysis of Drosophila and mammalian cut proteins in flies. Dev Biol 178: 149159.
  • Mardon G, Solomon NM, Rubin GM. 1994. dachsund encodes a nuclear protein required for normal eye and leg development in Drosophila. Development 120: 34733486.
  • Martin-Bermudo MD, Martinez C, Rodriguez A, Jimenez F. 1991. Distribution and function of the lethal of scute gene product during early neurogenesis in Drosophila. Development 113: 445454.
  • Merlo GR, Zerega B, Paleari L, Trombino S, Mantero S, Levi G. 2000. Multiple functions of Dlx genes. Int J Dev Biol 44: 619626.
  • Micchelli C, Rulifson E, Blair S. 1997. The function and regulation of cut expression on the wing margin of Drosophila: Notch, Wingless and a dominant negative role for Delta and Serrate. Development 124: 14851495.
  • Mullen LM, Bryant SV, Torok MA, Blumberg B, Gardiner DM. 1996. Nerve dependency of regeneration: the role of Distal-less and FGF signalling in amphibian limb regeneration. Development 122: 34873497.
  • Panganiban G, Sebring A, Nagy L, Carroll SB. 1995. The development of crustacean limbs and the evolution of arthropods. Science 270: 13631366.
  • Panganiban G, Irvine SM, Lowe C, Roehl H, Corley LS, Sherbon B, Grenier J, Fallon JF, Kimble J, Walker M, Wray GA, Swalla BJ, Martindale MQ, Carroll SB. 1997. The origin and evolution of animal appendages. Proc Natl Acad Sci U S A 94: 51625166.
  • Percival-Smith A, Hayden DJ. 1998. Analysis in Drosophila melanogaster of the interaction between Sex combs reduced and extradenticle activity in the determination of tarsus and arista identity. Genetics 150: 189198.
  • Percival-Smith A, Weber J, Gilfoyle E, Wilson P. 1997. Genetic characterization of the role of the two HOX proteins, Proboscipedia and Sex Combs Reduced, in determination of adult antennal, tarsal, maxillary palp and proboscis identities in Drosophila melanogaster. Development 124: 50495062.
  • Raz E, Shilo B-Z. 1993. Establishment of ventral cell fates in the Drosophila embryonic ectoderm requires DER, the EGF receptor homolog. Genes Dev 7: 19371948.
  • Rieckhof GE, Casares F, Ryoo HD, Abu-Shaar M, Mann RS. 1997. Nuclear translocation of extradenticle requires homothorax, which encodes an extradentical-related homeodomain protein. Cell 91: 171183.
  • Schweitzer R, Shaharabany M, Seger R, Shilo BZ. 1995. Secreted Spitz triggers the DER signaling pathway and is a limiting component in embryonic ventral ectoderm determination. Genes Dev 9: 15181529.
  • Skeath JB. 1999. At the nexus between pattern formation and cell-type specification: the generation of individual neuroblast fates in the Drosophila embryonic central nervous system. Bioessays 21: 922931.
  • Skeath JB, Carroll SB. 1991. Regulation of achaete-scute gene expression and sensory organ pattern formation in the Drosophila wing. Genes Dev 5: 984995.
  • Solomon K, Fritz A. 2002. Concerted action of two dlx paralogs in sensory placode formation. Development 129: 31273136.
  • Struhl G. 1981. A homeotic mutation transforming leg to antenna in Drosophila. Nature 292: 635638.
  • Struhl G. 1982. Genes controlling segmental specification in the Drosophila thorax. Proc Natl Acad Sci U S A 79: 73807384.
  • Sunkel CE, Whittle JRS. 1987. Brista: a gene involved in the specification and differentiation of distal cephalic and thoracic structures in Drosophila melanogaster. Wilhelm Rouxs Arch Dev Biol 196: 124132.
  • Tix S, Bate M, Technau G. 1989. Pre-existing neuronal pathways in the leg imaginal discs of Drosophila. Development 105: 855862.
  • Vachon G, Cohen B, Pfeifle C, McGuffin ME, Botas J, Cohen SM. 1992. Homeotic genes of the Bithorax Complex repress limb development in the abdomen of the Drosophila embryo through the target gene Distal-less. Cell 71: 437450.
  • Wasserman JD, Freeman M. 1998. An autoregulatory cascade of EGF receptor signaling patterns the Drosophila. egg Cell 95: 355364.
  • Weigmann K, Cohen S. 1999. Lineage-tracing cells born in different domains along the PD axis of the developing Drosophila leg. Development 126: 38233830.
  • Whiteley M, Noguchi PD, Sensabaugh SM, Odenwald WF, Kassis JA. 1992. The Drosophila gene escargot encodes a zinc finger motif found in snail-related genes. Mech Dev 36: 117127.
  • Wieschaus E, Nuesslein-Volhard C, Juergens G. 1984. Mutations affecting the pattern of larval cuticle in Drosophila melanogaster III Zygotic loci on the X-chromosome and the fourth chromosome. Rouxs Arch Dev Biol 193: 296307.
  • Wimmer E, Jackle H, Pfeifle C, Cohen S. 1993. A Drosophila homologue of human Sp1 is a head-specific segmentation gene. Nature 366: 690694.
  • Zipursky S, Venkatesh T, Teplow D, Benzer S. 1984. Neuronal development in the Drosophila retina: monoclonal antibodies as molecular probes. Cell 36: 1526.