In the developing Drosophila compound eye, a wave of pattern formation and cell-type determination sweeps across the presumptive eye epithelium. This ‘morpho-genetic furrow’ coordinates the epithelial cells’ division cycle, shape and gene expression to produce evenly spaced neural cell clusters that will eventually form the adult ommatidia. As these clusters develop, they rotate inwards to face the eye’s equator and establish tissue polarity. We have found that wingless is strongly expressed in the dorsal margin of the presumptive eye field, ahead of the morpho-genetic furrow. We have shown that inactivation of Wingless results in the induction of an ectopic furrow that proceeds ventrally from the dorsal margin. This ectopic furrow is normal in most respects, however the clusters formed by it fail to rotate, and we propose a two-vector model to account for normal rotation and tissue polarity in the retina. A second consequence of this inactivation of Wingless is that the dorsal head is largely deleted. We have also found that patched loss-of-function mosaic clones induce circular ectopic morphogenetic furrows (consistent with the observations of other workers with the hedgehog, and PKA genes). We use such patched induced furrows to test the two-vector model for cluster rotation and tissue polarity.

The morphogenetic furrow, and its regulation

In Drosophila, the adult retina develops from a monolayer columnar epithelium in the larval eye-antennal imaginal disc (Krafka, 1924; Chen, 1929). Initially the presumptive eye epithelium is unpatterned and undifferentiated. Early in the third larval instar a wave of progressive pattern formation and cell-type differentiation (the morphogenetic furrow) begins to pass across the epithelium from posterior to anterior (Melamed and Trujillo-Cenóz, 1975; Ready et al., 1976). This furrow is a vertical groove in the epithelium and is associated with changes in cell shape, synchronization of the cell’s division cycle, abrupt changes in gene expression and pattern formation (Ready et al., 1976; Tomlinson, 1985, 1988; Wolff and Ready, 1991). In the furrow, subsets of cells are specified to form five cell preclusters that will form the core of the nascent ommatidium, and these remain in G1 arrest (Thomas et al., 1994). The continued progression of the morphogenetic furrow depends on posterior induction by the protein product of the hedgehog (hh) gene (Heberlein et al., 1993; Ma et al., 1993). Ectopic Hh expression anterior to the furrow is sufficient to induce a circular ectopic furrow (Heberlein et al., 1995). The patched (ptc) gene encodes a trans-membrane protein that may regulate the Hh signaling pathway in the receiving cell (Hooper and Scott, 1989; Nakano et al., 1989). ptc mutants appear to have an opposite phenotype to hh, and ptc is epistatic to hh, in the embryo and eye, consistent with a role in the receiving cell (Martinez-Arias et al., 1988; Hidalgo, 1991; Ingham et al., 1991; Bejsovec and Wieschaus, 1993; Martinez Arias, 1994; Schuske et al., 1994). New evidence suggests that the cAMP-dependent protein kinase (PKA) may act in a regulatory pathway with Ptc (Jiang and Struhl, 1995; Lepage et al., 1995; Li et al., 1995), and loss of PKA from retinal mosaic clones has the same effect as gain of Hh (Pan and Rubin, 1995; Strutt et al., 1995).

However, the Hh signal is not required for furrow initiation on the posterior margin (Heberlein et al., 1993; Ma et al., 1993). The decapentaplegic (dpp) gene is expressed in the furrow, and in two domains in the disc margin (like the initials on a capital ‘I’, Blackman et al., 1991). When hh function is removed (with a temperature-sensitive mutation), dpp expression in the furrow is abolished, but the marginal expression domains remain, and continue to progress, as if a ghost furrow remained in between (Ma et al., 1993). That furrow initiation and marginal dpp expression are hh indepen-dent suggests that a second genetic system is responsible for furrow initiation on the posterior margin and its continuous reinitiation along the dorsal and ventral disc margins. In order to study this independent pathway of furrow regulation, we turned to the other major embryonic segment-polarity signaling molecule: Wingless.

The wingless gene (wg) is one of sixteen in the segment-polarity class that are responsible for the establishment of the first compartment boundaries and segmentation in the Drosophila embryo (reviewed in Peifer and Bejsovec, 1992 and van den Heuvel et al., 1993). It encodes a secreted, diffusible product (Wg), which is a homolog of the murine onco-protein Wnt-1 (Rijsewijk et al., 1987; Nusse and Varmus, 1992). The Wg signal acts locally, and may also act at a greater distance as a graded morphogen (Noordermeer et al., 1992; Peifer et al., 1993). In the embryo, an anterior wingless signal is opposed by an posterior hedgehog (hh) signal (Peifer and Bejsovec, 1992). The Wg signal acts via a pathway that may include Notch (Couso and Martinez Arias, 1994) or Patched (Capdevila et al., 1994), and regulates the phosphorylation and localization of the β-catenin homolog Armadillo (Peifer et al., 1994). Wg also acts to pattern the developing Drosophila neu-roblasts (Chu-LaGraff and Doe, 1993) and brain (Kaphingst and Kunes, 1994), as does Wnt-1 in the mouse (reviewed in Nusse and Varmus, 1992). In the leg and wing imaginal discs, Wg is expressed in a ventral domain (Baker, 1988a) and, together with a dorsal Hh signal, acts via the TGF-β homolog Decapentaplegic (Dpp) to pattern these appendages (Cohen and Di Nardo, 1993). Indeed, the site of ectopically expressed Wg can act as an organizer in these discs (Struhl and Basler, 1993), as can a homolog in Xenopus embryos (reviewed in Nusse and Varmus, 1992). Thus we chose to examine wg expression and function in the developing eye (see Results below).

Cluster rotation and tissue polarity in the retina

Once the precluster has resolved, the five cells are initially arranged with the future R8 cell furthest from the furrow (Fig. 1A). Over the next six to eight columns (12 to 16 hours), the clusters rotate 90° so that the final photoreceptor cell (R7) comes to lie closest to the equator (Fig. 1B, and Ready et al., 1976). Thus in the right eye, the clusters dorsal to the equator rotate 90° counterclockwise and those ventral to the equator rotate 90° clockwise. At the same time, the developing R4 cell moves away from the center of the cluster, breaking symmetry (Tomlinson, 1985). During pupal life retinal development continues so that in the adult the eight photoreceptor cells line the center of each ommatidium, and project light gathering microvillar rhabdomeres into a central space (Dietrich, 1909; Waddington and Perry, 1960). These eight rhabdomeres lie in a chiral trapezoidal pattern (Fig. 1C), with six large outer rhabdomeres (from photoreceptor cells R1 through R6), and two smaller inner ones (from R7 and R8, with the R8 rhabdomere below that of R7). The long face of this trapezoid faces towards the anterior and is formed by the rhabdomeres of R1, 2 and 3. The two eyes have ommatidial trapezoids that are mirror-image reflections of each other. There is a second plane of reflection of retinal tissue polarity: the equator at the horizontal center of each eye. The ommatidia in the upper and lower hemispheres of each eye are reflections of each other, in that the R7 cell always lies towards the equator.

Fig. 1.

Development of the arrangement and orientation of the ommatidial photoreceptor cells. The top three panels show stages in the development of an ommatidium in the upper half of a right-hand Drosophila compound eye (after Dietrich, 1909; Waddington and Perry, 1960; Ready et al., 1976; Tomlinson, 1985; Tomlinson and Ready, 1987). Ommatidia in the lower half of the right eye are mirror-image reflections of these forms in their horizontal axes, those in the upper half of the left eye are reflections of these forms in their vertical axes and those in the lower half of the left eye are reflected in both axes. In each panel anterior is to the right and the equator is down. (A) Diagram of the early stages of the development of the ommatidial cluster, just posterior to (left of) the morphogenetic furrow. The proto-R8 cell is the first neuron (heavy circle). R2 and R5 are next and R3 and R4 follow them. These five cells are of the same cell generation and form the ‘precluster’. They are later joined by the R1/6 pair, and finally R7. Initially the cluster’s axis of bilateral symmetry is perpendicular to the furrow, as shown. As these recruitment events proceed the cluster rotates 90° counterclockwise and the R4 cell moves slightly out, so that the arrangement of the cluster is as shown in B. The clusters in the lower half of the eye rotate 90° clockwise. In later development, the R4 cell moves back into the cluster, and the R3 cell moves out, so that in the adult retina the eight photoreceptor cells’ rhabdomeres are in the chiral trapezoidal configuration shown in C. The R7 rhabdomere lies above that of R8, in the center/bottom, as shown. (D-F) Symbols that represent the clusters shown in A-C.

Fig. 1.

Development of the arrangement and orientation of the ommatidial photoreceptor cells. The top three panels show stages in the development of an ommatidium in the upper half of a right-hand Drosophila compound eye (after Dietrich, 1909; Waddington and Perry, 1960; Ready et al., 1976; Tomlinson, 1985; Tomlinson and Ready, 1987). Ommatidia in the lower half of the right eye are mirror-image reflections of these forms in their horizontal axes, those in the upper half of the left eye are reflections of these forms in their vertical axes and those in the lower half of the left eye are reflected in both axes. In each panel anterior is to the right and the equator is down. (A) Diagram of the early stages of the development of the ommatidial cluster, just posterior to (left of) the morphogenetic furrow. The proto-R8 cell is the first neuron (heavy circle). R2 and R5 are next and R3 and R4 follow them. These five cells are of the same cell generation and form the ‘precluster’. They are later joined by the R1/6 pair, and finally R7. Initially the cluster’s axis of bilateral symmetry is perpendicular to the furrow, as shown. As these recruitment events proceed the cluster rotates 90° counterclockwise and the R4 cell moves slightly out, so that the arrangement of the cluster is as shown in B. The clusters in the lower half of the eye rotate 90° clockwise. In later development, the R4 cell moves back into the cluster, and the R3 cell moves out, so that in the adult retina the eight photoreceptor cells’ rhabdomeres are in the chiral trapezoidal configuration shown in C. The R7 rhabdomere lies above that of R8, in the center/bottom, as shown. (D-F) Symbols that represent the clusters shown in A-C.

These rotations require the wild-type function of the nemo gene (Choi and Benzer, 1994). In nemo mutants, the ommatidial clusters rotate the first 45°, but fail to complete their rotation. Mutations in a number of other genes can affect tissue polarity in the eye (Gubb, 1993) and other tissues (Theisen et al., 1994), but the molecular mechanisms underlying tissue polarity have remained obscure. It may be that ommatidial precluster polarity follows from an initial cell polarity established in the founding R8 photoreceptor cells. In EgfrE homozygotes, the ommatidial clusters develop much further apart than normal (Baker and Rubin, 1989), yet they rotate normally (Baker and Rubin, 1992). Therefore the rotating clusters do not require near neighbors to do so, which implies that there may be a global field of retinal polarity. Genetic manipulations that perturb the furrow can cause reversals of retinal polarity (Heberlein et al., 1995; Strutt et al., 1995). We have observed these and other changes in retinal polarity when the furrow is reoriented by means of wg or ptc loss-of-function (see results below), and we propose a model for normal cluster chirality and rotation that can account for all these data.

Drosophila stocks

The wild-type stock used was Canton-S. The wg enhancer trap line was wgen-11, a gift of Alfonso Martinez Arias. dpp expression was visualized with the dpp disc enhancer β-galac-tosidase expression line H1-1, is an insertion of the BS3.0 construct (Blackman et al., 1991) and is a gift of Ron Blackman. The wg temperature-sensitive mutation is wgl-12 (= wgIL, Lindsley and Zimm, 1992, gift of Amy Bejsovec and Alfonso Martinez Arias). wg1 was a gift from Roel Nusse. The ptc alleles used in all data shown here were ptc7M (= tuf5 in Lindsley and Zimm, 1992) and ptc6P (= tuf4). Other alleles tested were: ptcIN (= tuf9) and ptcIIB (= tuf10); all four alleles are gifts of Joan Hooper. The ptc mosaic clones were made by somatic recombination in the first larval instar, using an X-linked heat-shock promoter/FLP recombinase insertion (Golic and Lindquist, 1989), and the second chromosome FRT insertion at band 43D (P[ry+; hs-neo; FRT]43D, Xu and Rubin, 1993). The clones were marked with white by the use of the P[(w, ry)A]1-1 (Hazelrigg et al., 1984) at polytene band 47A on the ptc+ chromosome arm.

In situ hybridization and immunohistochemistry and electron microscopy

In situ hybridization was done as described by Tautz and Pfeifle (1989), as modified by Ma et al. (1993). The probe was a genomic fragment pwg-12 (gift of Nick Baker, Baker, 1987), labeled by nick translation with digox-igenin-dUTP. Immunohistochemistry was done as described by Tomlinson and Ready (Tomlinson and Ready, 1987). The anti-Wg antiserum (van den Heuvel et al., 1993) was a rabbit polyclonal (gift of Roel Nusse). Anti-Elav was Elav-9F8A9 (Robinow and White, 1991) from the Developmental Studies Hybridoma Bank, Iowa, anti-Scabrous (Baker et al., 1990; Mlodzik et al., 1990) is a gift of Nick Baker, anti-Hairy (Carroll and Whyte, 1989) is a gift of Brown and Carroll. dpp expression was visualized by means of an insertion of the BS3.0 construct (Blackman et al., 1991, see above) and anti-β-galactosidase (Promega # Z3782). The secondary antibodies were goat anti-mouse IgG-HRP conjugate (Bio-Rad # 170-6516, visualized with diamino-benzidine) and goat anti-rabbit IgG-Biotin conjugate (Vector Labs # BA-1000, visualized with a ‘Vectastain’ ABC kit, Vector Labs PK-6104). Bromodeoxy-uridine (BrdU, Sigma B-0631) staining was done as described by Wolff and Ready (1991), detected with a mouse anti-BrdU monoclonal (Becton Dickinson # 347580). Scanning electron microscopy was done as described in Moses et al. (1989).

wg is a negative regulator of furrow initiation

Our analysis of the function of the Hh signal in the regulation of the morphogenetic furrow (Ma et al., 1993) suggested that some other system acts to regulate furrow initiation and progression in the disc margins. Hh and Wg signals interact in embryonic segmentation (Ingham and Martinez-Arias, 1992; Peifer and Bejsovec, 1992; Perrimon, 1994) and the development of the leg and wing imaginal discs (Bryant, 1993; Vincent and Lawrence, 1994; Blair, 1995). Therefore, we studied wg expression in the developing eye by RNA in situ hybridization (Fig. 2A) and immunohistochemistry (Fig. 2B). In the antennal disc, Wg is expressed in a dorsal domain (asterisk in Fig. 2), unlike the leg discs. It may be that the eye-antennal imaginal disc is inverted in the dorsoventral direction relative to the leg discs. This would be consistent with the phenotype of the ectopic legs induced by the ectopic expression of Antennapedia (Schneuwly et al., 1987; Gibson and Gehring, 1988). In the eye field, there is a narrow dorsal expression domain, anterior to the furrow (large arrows in Fig. 2), a dot near the site of the future ocelli (small arrows in Fig. 2) and a weak line of expression in the posterior margin (‘V’s in Fig. 2). In some cases, we also observe a weaker, ventral domain of wg expression.

Fig. 2.

wg expression in the developing eye. (A) RNA in situ hybridization to whole-mount disc. (B) Wg antigen. In each panel: the arrowhead shows the position of the furrow, the asterisk shows expression in the antenna, the large arrow shows expression in the dorsal margin, the small arrow shows expression near the future ocelli and the ‘v’ shows the line of expression in the posterior margin. Both panels are at the same magnification, the scale bar is 100 μm, dorsal is up and anterior is to the right.

Fig. 2.

wg expression in the developing eye. (A) RNA in situ hybridization to whole-mount disc. (B) Wg antigen. In each panel: the arrowhead shows the position of the furrow, the asterisk shows expression in the antenna, the large arrow shows expression in the dorsal margin, the small arrow shows expression near the future ocelli and the ‘v’ shows the line of expression in the posterior margin. Both panels are at the same magnification, the scale bar is 100 μm, dorsal is up and anterior is to the right.

To determine the function of Wg in the developing eye, we inactivated it with a temperature-sensitive allele (wgl-12, Baker, 1988b; Tearle and Nüsslein-Volhard, 1987). We raised wgl-12 homozygotes at a permissive temperature (16.5°C) until the late second larval instar, then transferred them to a non-permissive temperature (29°C) for 2 days. An ectopic furrow was induced in these wg discs (perpendicular to the endogenous furrow) proceeding down from the dorsal margin (Fig. 3). This ectopic furrow appears normal in most respects: it is associ-ated with normal dpp (Blackman et al., 1991), Scabrous (Mlodzik et al., 1990) and Hairy (Carroll and Whyte, 1989) expression, it produces well-spaced clusters of developing neurons and it confers cell-cycle synchrony (Fig. 3). However, the ectopic furrow is not normal with respect to ommatidial rotation (see below). In some cases, we have observed smaller (i.e. later) ectopic furrows on the ventral side.

Fig. 3.

The late mutant phenotype of temperature-shifted wgl-12 eye-imaginal discs. (A-C) Control normal discs (heterozygous sibs); (D-F) wgl-12 homozygotes. All discs were from animals raised at 16.5°C until the late second instar and then shifted to 29°C for 48 hours. (A,D) Discs stained for the neural nuclear antigen Elav (black) and the expression of dpp (brown). (B,E) Discs stained to show the expression of the Scabrous protein (brown) and the Hairy protein (black). (C,F) Stained to show DNA synthesis at the time of dissection by the incorporation of bromodeoxyuridine (BrdU). Arrowheads show position of the furrow. Arrows in A and B show the position of the future ocelli. All panels are at the same magnification, the scale bar in F is 100 μm, dorsal is up and anterior is to the right.

Fig. 3.

The late mutant phenotype of temperature-shifted wgl-12 eye-imaginal discs. (A-C) Control normal discs (heterozygous sibs); (D-F) wgl-12 homozygotes. All discs were from animals raised at 16.5°C until the late second instar and then shifted to 29°C for 48 hours. (A,D) Discs stained for the neural nuclear antigen Elav (black) and the expression of dpp (brown). (B,E) Discs stained to show the expression of the Scabrous protein (brown) and the Hairy protein (black). (C,F) Stained to show DNA synthesis at the time of dissection by the incorporation of bromodeoxyuridine (BrdU). Arrowheads show position of the furrow. Arrows in A and B show the position of the future ocelli. All panels are at the same magnification, the scale bar in F is 100 μm, dorsal is up and anterior is to the right.

We have examined the earlier development of eye-imaginal discs from which wg function was removed by tem-perature shift for 36 hours (Fig. 4) or 24 hours (data not shown). Both experiments yield indistinguishable results. The first visible effect of removing wg function in very early third instar discs (before furrow initiation) is an increase in dpp expression on the anterior side (compare Fig. 4A and E). However, this loss of wg activity cannot induce furrows in younger (second-instar discs): the example shown in Fig. 4E has not yet initiated the furrow (at early third instar) despite having lacked wg function for 36 hours. Once the furrow does initiate without wg function, it extends abnormally along the dorsal margin (compare Fig. 4B and F), and often to some extent along the ventral margin (consistent with the weaker expression domain seen there, see above). This ectopic furrow does not occupy all of the dpp expression domain. Thus Dpp may be necessary, but is not sufficient for furrow initiation.

Fig. 4.

The early mutant phenotype of temperature-shifted wgl-12 eye-imaginal discs. (A-D) Control normal discs (heterozygous sibs), arranged in a developmental series (youngest on the left); (E-H) wgl-12 homozygotes, arranged in a developmental series (youngest on the left). All discs were from animals raised at 16.5°C until the second instar and then shifted to 29°C for 36 hours (discs from animals shifted for 24 hours at this stage are indistinguishable). All discs were stained for the neural nuclear antigen Elav (black), and the expression of dpp (brown). Note that the first visible effect of removing wg function is to increase the domain of dpp expression towards the anterior (compare A and E). This is soon followed by ectopic furrow activity, mostly on the dorsal margin (compare B and F). Arrowheads indicate the normal limits of dpp expression. All panels are at the same magnification, the scale bar in F is 50 μm, dorsal is up and anterior is to the right.

Fig. 4.

The early mutant phenotype of temperature-shifted wgl-12 eye-imaginal discs. (A-D) Control normal discs (heterozygous sibs), arranged in a developmental series (youngest on the left); (E-H) wgl-12 homozygotes, arranged in a developmental series (youngest on the left). All discs were from animals raised at 16.5°C until the second instar and then shifted to 29°C for 36 hours (discs from animals shifted for 24 hours at this stage are indistinguishable). All discs were stained for the neural nuclear antigen Elav (black), and the expression of dpp (brown). Note that the first visible effect of removing wg function is to increase the domain of dpp expression towards the anterior (compare A and E). This is soon followed by ectopic furrow activity, mostly on the dorsal margin (compare B and F). Arrowheads indicate the normal limits of dpp expression. All panels are at the same magnification, the scale bar in F is 50 μm, dorsal is up and anterior is to the right.

Loss of wg function also deletes that part of the eye-antennal imaginal disc that will give rise to the dorsal head cuticle and ocelli, and there is also some loss of ventral tissue. It is known that wg regulates cell number in the wing imaginal disc (hence the ‘wingless’ phenotype of the original allele, wg1, Baker, 1988a). Indeed, the eyes of wg1 homozygotes are reduced in size, particularly on the ventral side (wg1 is thought cause a partial loss of wg function, Fig. 5C,D). In summary, these data suggest that the normal function of wg expression, anterior to the furrow, in the dorsal margin, is to act as a negative regulator of furrow initiation.

Fig. 5.

Scanning electron micrographs of wild-type and mutant compound eyes. (A,C) Wild-type; (B)wgl-12 homozygous pharate adult, which was raised at 16.5°C until the late second instar, and then shifted to 29°C for 24 hours, then returned to 16.5°C. Note that the compound eyes extend over the dorsal head cuticle, and the ocelli and antennae are deleted; (D) wg1 homozygote. Note that the eye is reduced ventrally. (A,B) Dorsal views at the same magnification, anterior is to the right. (C,D) Lateral views at the same magnification, anterior is to the right and dorsal is up. The scale bars in B and D are both 200 μm.

Fig. 5.

Scanning electron micrographs of wild-type and mutant compound eyes. (A,C) Wild-type; (B)wgl-12 homozygous pharate adult, which was raised at 16.5°C until the late second instar, and then shifted to 29°C for 24 hours, then returned to 16.5°C. Note that the compound eyes extend over the dorsal head cuticle, and the ocelli and antennae are deleted; (D) wg1 homozygote. Note that the eye is reduced ventrally. (A,B) Dorsal views at the same magnification, anterior is to the right. (C,D) Lateral views at the same magnification, anterior is to the right and dorsal is up. The scale bars in B and D are both 200 μm.

With the temperature-shift regime described above, wgl-12 homozygous larvae die as early pupae. We moderated this protocol, by raising them as described above, but then returning them to the lower temperature after only 24 hours, and allowing them to form pupae. A few of these animals survive to form pharate adults, and these have grossly distorted heads, with the compound eyes covering most of the dorsal head, consistent with the imaginal disc phenotype (see above), with the ocelli and antennae deleted (Fig. 5A,B). Insects show a great variety of eye and head morphology and it may be that this evolved, in part, through changes in wg expression. We cut sections of the eyes of these wgl-12 temperature-shifted pharate adults, but we found that their internal structure is dis- organized and degenerated (data not shown).

The ectopic wg furrow affects cluster rotation

In normal development, the ommatidial preclusters are initially oriented parallel to the furrow: the first three cells form a chevron, with its axis perpendicular to the furrow. Over the next 12 to 16 hours, these clusters rotate 90° so that their axes point towards the equator (Ready et al., 1976; Tomlinson, 1985; Tomlinson and Ready, 1987). When wg function is removed, the clusters that are formed by the endogenous (vertical) furrow, continue to rotate normally. Thus, wg does not appear to have a direct role in ommatidial rotation. However, the clusters generated by the ectopic (horizontal) furrow fail to rotate (Fig. 6A,B). This ectopic furrow phenotype suggests a simple two-vector model to account for ommatidial rotation in normal development: nascent ommatidial clusters are oriented by two directional vectors, the vector direction to the furrow (white arrow in Fig. 6D), and the vector direction to the equator (black arrow in Fig. 6D). In normal development these two vectors are at right angles, and produce a moment of opposite chirality above and below the equator. However, behind the ectopic furrow generated by the loss of wg function, the two vectors both point in the same direction, this is a nonsense code and the clusters do not rotate.

Fig. 6.

Ommatidial clusters generated by the ectopic furrow generated by the loss of wg function, fail to rotate. (A,C) The same temperature-shifted righthand wgl-12 homozygote imaginal disc as shown in Fig. 3D. Anterior is to the right and dorsal is up. The scale bar in A is 20 μm, and the scale bar in C is 50 μm. (B) Diagrammatic representation of the field shown in A. The dark gray bars show the position of the furrow. The arrowheads show the orientation of the neural cell clusters (as in Fig. 1D,E). The light gray circles are clusters for which orientation was ambiguous. (D) Diagrammatic version of the field shown in C to illustrate the two-vector model for ommatidial rotation. In each area the arrowhead in a circle represents a nascent ommatidial cluster near the furrow which has not yet rotated. The circled arrowheads show the initial orientation of the cluster, which is always perpendicular to the furrow. The white arrows represent the vector direction from the clusters to the furrow. The black arrows represent the vector direction from the clusters to the equator. The curved dotted arrows show the direction in which the clusters will rotate. In area 1 the equator is up and the furrow is to the right. This defines a chiral form that instructs the cluster to rotate clockwise. In area 2 the equator is down and the furrow is to the right. This defines a chiral form (opposite to area 1) that instructs the cluster to rotate counterclockwise. In a normal left-hand eye the areas corresponding to 1 and 2 have the mirror image vectors and rotate appropriately. In area 3 (behind the ectopic furrow) both vectors point in the same direction. This is a meaningless combination and the clusters do not rotate.

Fig. 6.

Ommatidial clusters generated by the ectopic furrow generated by the loss of wg function, fail to rotate. (A,C) The same temperature-shifted righthand wgl-12 homozygote imaginal disc as shown in Fig. 3D. Anterior is to the right and dorsal is up. The scale bar in A is 20 μm, and the scale bar in C is 50 μm. (B) Diagrammatic representation of the field shown in A. The dark gray bars show the position of the furrow. The arrowheads show the orientation of the neural cell clusters (as in Fig. 1D,E). The light gray circles are clusters for which orientation was ambiguous. (D) Diagrammatic version of the field shown in C to illustrate the two-vector model for ommatidial rotation. In each area the arrowhead in a circle represents a nascent ommatidial cluster near the furrow which has not yet rotated. The circled arrowheads show the initial orientation of the cluster, which is always perpendicular to the furrow. The white arrows represent the vector direction from the clusters to the furrow. The black arrows represent the vector direction from the clusters to the equator. The curved dotted arrows show the direction in which the clusters will rotate. In area 1 the equator is up and the furrow is to the right. This defines a chiral form that instructs the cluster to rotate clockwise. In area 2 the equator is down and the furrow is to the right. This defines a chiral form (opposite to area 1) that instructs the cluster to rotate counterclockwise. In a normal left-hand eye the areas corresponding to 1 and 2 have the mirror image vectors and rotate appropriately. In area 3 (behind the ectopic furrow) both vectors point in the same direction. This is a meaningless combination and the clusters do not rotate.

ptc clones induce circular ectopic furrows which reorient the retina

We have previously shown (Ma et al., 1993) that ptc loss-of-function mutations are dominant suppressors of hh loss-of-function in the retina (as they are in the embryo, Peifer and Bejsovec, 1992). The ptc gene encodes a trans-membrane protein (Ptc) and it has been suggested that Ptc may be a receptor for the Hh signal. Therefore we examined ptc function in eye development by making homozygous ptc mutant mosaic clones, with four different alleles (see Materials and Methods), which we detected in the adult retina by negative marking with white. We were able to obtain homozygous mosaic clones with three ptc alleles (ptc7M, ptc6P and ptcIIB, see Materials and Methods). We were unable to obtain homozygous clones with one allele (ptcIN), and it may carry a cell-lethal mutation on the same chromosome arm. In the ptc mutant mosaic clones that we recovered, the ptc mutant tissue is severely disrupted and the mutant clone affects the neighboring wild-type tissue (see below).

We examined larval eye imaginal discs, in which ptc clones were induced, and we observed circular ectopic furrows (Fig. 7A). These ectopic furrows are likely to be induced from the ptc tissue and propagate outwards. It has been shown that ectopic expression of hh can induce such ectopic furrows (Heberlein et al., 1995). As ptc has an opposite phenotype to hh, and is epistatic to it, the phenotypic effect of loss of ptc function in the eye may be equivalent to adding ectopic Hh expression. The PKA gene is also thought to be in this regulatory pathway, perhaps downstream of Ptc, and PKA mutant retinal clones can also induce ectopic furrows (Pan and Rubin, 1995; Strutt et al., 1995). Thus the ptc ectopic furrow is consistent with a simple (linear pathway) interpretation of these data.

Fig. 7.

Ommatidial clusters generated by the circular ectopic furrow induced by a ptc loss-of-function mosaic clone, rotate as predicted by the two-vector model. (A-C) Photomicrographs; (D-F) explanatory diagrams for each. In all panels anterior is to the right, and dorsal is up. All three scale bars are 50 μm. (A) An eye-imaginal disc in which a ptc7M clone was induced, stained to show the Scabrous protein. This reveals the position of the morphogenetic furrows in the eye field (and other structures elsewhere). The endogenous furrow lies between the two arrowheads, and is moving from left to right. A second, circular furrow (arrow) is just anterior to it. (B) Low magnification view of a section of an adult retina, which includes a homozygous ptc7M loss-of-function mosaic clone, marked with white (arrow). (C) High magnification view of the same section, with the white ptc7M clone in the center. (D)Diagram to explain how the two-vector model relates to the larval eye disc shown in A. The blue, zig-zag line is the future equator (shown lower than normal, for clarity). The red lines are the furrows. The circled arrowheads represent developing clusters as shown in Fig 6D. The two-vector model predicts that behind endogenous furrow, the dark blue cluster will rotate clockwise (same as area 1 in Fig. 6D), and the light blue cluster will rotate counterclockwise (same as area 2 in Fig. 6D). The ptc clone (gray) induces an ectopic furrow (red circle). Within this ectopic furrow are shown four clusters: at the eastern, southern, western and northern positions. All of the clusters within the ectopic circle have the equator down, so their black arrow vector points down. However, each has a different vector direction to the furrow that generated them (white arrow). The cluster at the eastern position has the same vector pair as the light blue cluster behind the endogenous furrow, so it is colored light blue, and the model predicts that both will rotate 90° counterclockwise, to form an adult ommatidium identical to that shown in Fig. 1C. The green cluster at the southern position has both vectors pointing down. This is the same as the clusters generated by the wg ectopic furrow in area 3 of Fig. 6D, and thus the model predicts that it will not rotate. However the model does not predict which chiral form will result from the green cluster. The yellow cluster at the western position has the furrow to its left. This corresponds to a mirror-image reflection of the clusters shown in light blue. The two-vector model predicts that this yellow cluster will rotate 90° clockwise, and form the reverse reflection of the chiral form of the light blue clusters. The pink cluster at the northern position has the equatorial vector down and the furrow vector up. This is a nonsense code, not seen in the wg experiments, so the two-vector model makes no prediction of its polarity. (E) Diagram representing the adult section shown in B. The two-vector model predicts that four lobes of similar tissue polarity will form around the ptc clone, as described in D. The colors of the predicted regions of similar polarity correspond to the colors of the clusters in D. The symbols within each area show the polarity and chiral form of the ommatidia that are actually found there (following the convention shown in Fig. 1F). (F) Tracing of the field shown in C. The ptc tissue is shown in gray. Each ommatidium is rendered as the appropriate chiral arrow as defined in Fig. 1F. Ommatidia for which the orientation or chiral form is ambiguous are shown as red circles. The four lobes of tissue polarity that surround the ptc clone (clone no. 1 in Table 1) are color coded the same as D and E.

Fig. 7.

Ommatidial clusters generated by the circular ectopic furrow induced by a ptc loss-of-function mosaic clone, rotate as predicted by the two-vector model. (A-C) Photomicrographs; (D-F) explanatory diagrams for each. In all panels anterior is to the right, and dorsal is up. All three scale bars are 50 μm. (A) An eye-imaginal disc in which a ptc7M clone was induced, stained to show the Scabrous protein. This reveals the position of the morphogenetic furrows in the eye field (and other structures elsewhere). The endogenous furrow lies between the two arrowheads, and is moving from left to right. A second, circular furrow (arrow) is just anterior to it. (B) Low magnification view of a section of an adult retina, which includes a homozygous ptc7M loss-of-function mosaic clone, marked with white (arrow). (C) High magnification view of the same section, with the white ptc7M clone in the center. (D)Diagram to explain how the two-vector model relates to the larval eye disc shown in A. The blue, zig-zag line is the future equator (shown lower than normal, for clarity). The red lines are the furrows. The circled arrowheads represent developing clusters as shown in Fig 6D. The two-vector model predicts that behind endogenous furrow, the dark blue cluster will rotate clockwise (same as area 1 in Fig. 6D), and the light blue cluster will rotate counterclockwise (same as area 2 in Fig. 6D). The ptc clone (gray) induces an ectopic furrow (red circle). Within this ectopic furrow are shown four clusters: at the eastern, southern, western and northern positions. All of the clusters within the ectopic circle have the equator down, so their black arrow vector points down. However, each has a different vector direction to the furrow that generated them (white arrow). The cluster at the eastern position has the same vector pair as the light blue cluster behind the endogenous furrow, so it is colored light blue, and the model predicts that both will rotate 90° counterclockwise, to form an adult ommatidium identical to that shown in Fig. 1C. The green cluster at the southern position has both vectors pointing down. This is the same as the clusters generated by the wg ectopic furrow in area 3 of Fig. 6D, and thus the model predicts that it will not rotate. However the model does not predict which chiral form will result from the green cluster. The yellow cluster at the western position has the furrow to its left. This corresponds to a mirror-image reflection of the clusters shown in light blue. The two-vector model predicts that this yellow cluster will rotate 90° clockwise, and form the reverse reflection of the chiral form of the light blue clusters. The pink cluster at the northern position has the equatorial vector down and the furrow vector up. This is a nonsense code, not seen in the wg experiments, so the two-vector model makes no prediction of its polarity. (E) Diagram representing the adult section shown in B. The two-vector model predicts that four lobes of similar tissue polarity will form around the ptc clone, as described in D. The colors of the predicted regions of similar polarity correspond to the colors of the clusters in D. The symbols within each area show the polarity and chiral form of the ommatidia that are actually found there (following the convention shown in Fig. 1F). (F) Tracing of the field shown in C. The ptc tissue is shown in gray. Each ommatidium is rendered as the appropriate chiral arrow as defined in Fig. 1F. Ommatidia for which the orientation or chiral form is ambiguous are shown as red circles. The four lobes of tissue polarity that surround the ptc clone (clone no. 1 in Table 1) are color coded the same as D and E.

The ptc-induced circular ectopic furrow provides a second test of the two-vector model of ommatidial rotation and retinal tissue polarity. Consider nascent clusters, at four positions around the circular furrow (Fig. 7D). The cluster at the Eastern position has the same vector pair (furrow direction and equa-torial direction) as a cluster behind the endogenous furrow (both light blue in Fig. 7D) and should thus rotate normally (light blue arc and lobe in Fig. 7E). The cluster at the southern position (green in Fig. 7D) has the same vector information as the clusters generated by the wg ectopic furrow (see above) and therefore should not rotate. As the R7 cell is formed at the dorsal side of this cluster, it will remain there. However, we can not predict the chiral form of the resulting cluster in the southern lobe (whether the R3 cell face anterior or posterior, see the green lobe, Fig. 7E). The cluster at the western position (yellow in Fig. 7D) has the mirror image reflection of the normal (light blue) cluster, and the two-vector model predicts that it will develop into a reversed ommatidium (yellow lobe, Fig. 7E). The cluster at the northern position (pink in Fig. 7D) has the furrow above and the equator below. This is a nonsense code, but a different code to that seen in the wg ectopic furrow. Thus from the wg data alone, we could not predict its fate.

Table 1.

Numerical data from ptc mosaic clones

Numerical data from ptc mosaic clones
Numerical data from ptc mosaic clones

We scored the orientation of clusters in the adult retina around ptc clones (Figs 7C,F, 8) and, in the best cases, we can detect the four lobes as predicted by the two-vector model (see Table 1). This experiment reveals the meaning of the nonsense code experienced by clusters in the northern lobe (pink in Fig. 7): they rotate normally, however, as they are generated by a horizontal furrow, they eventually come to lie 90° clockwise of the normal position. The position, size and shape of the ptc clones that we obtained were such that we could detect all four lobes in two examples (one shown in Fig. 7C, see Table 1). In a total of 20 clones, we scored the eastern (light blue) lobe in 13 cases, the western (yellow) lobe in 19 cases, the southern (green) lobe in 6 cases and the northern (pink) lobe in 6 cases. In the ideal case, a ptc clone would be a point, or small circle, which would generate a perfect circular furrow, that would propagate outwards in all directions equally. Clones nos. 1 and 2 (Table 1) are close to this ideal (clone no. 1 is shown in Fig. 7). However, many other shapes of clones are actually recovered, and thus the furrow that they generate may not always produce all four lobes as predicted by the two-vector model. Furthermore clones may lie close to the eye margin, which can also eliminate one or more lobes and not all ommatidia can be scored for their orientation: some are morphologically abnormal and some sections are imperfect. A combination of these factors explains the variable numbers of ommatidia scored in each lobe in each clone. An extreme example is clone no. 20, which is a vertical bar across much of the eye, that crosses the equator (Fig. 8; Table 1). In this case, there is no southern lobe as there is no tissue between the clone and the equator. However, polarity within the eastern and western lobes reverses at the equator (Fig. 8). This suggests that the equator preexists the furrow and cannot be altered by the perturbations induced by ptc clones. It should be noted that polarity reversals consistent with these have been reported near ectopic Hh clones (Heberlein et al., 1995) and PKA clones (Strutt et al., 1995). Furthermore, tissue polarity reversals have been detected near ptc mosaic clones in the wing, so a similar system of tissue polarity may operate there (Phillips et al., 1990).

Fig. 8.

Ommatidial clusters generated by furrows induced by a ptc loss-of-function mosaic clone that crosses the equator, continue to respect the equator. (A) Retinal section, which includes ptc7M clone no. 20 (see Table 1). (B) Tracing of the field shown in A (with some additional information from other sections of the same retina). The ptc tissue is shown in gray. Each ommatidium is rendered as the appropriate chiral arrow as defined in Fig. 1F. Ommatidia for which the orientation or chiral form is ambiguous are shown as red circles. The two lobes of tissue polarity that surround this ptc clone are color coded the same as Fig. 7, except that the western lobe that is ventral to the equator is shaded a darker yellow than that part of the western lobe that is dorsal to the equator. Note that the eastern and western lobes occur as predicted by the two-factor model, and that they are reflected in the equator. Anterior is to the right, and dorsal is up. Scale is as Fig. 7C.

Fig. 8.

Ommatidial clusters generated by furrows induced by a ptc loss-of-function mosaic clone that crosses the equator, continue to respect the equator. (A) Retinal section, which includes ptc7M clone no. 20 (see Table 1). (B) Tracing of the field shown in A (with some additional information from other sections of the same retina). The ptc tissue is shown in gray. Each ommatidium is rendered as the appropriate chiral arrow as defined in Fig. 1F. Ommatidia for which the orientation or chiral form is ambiguous are shown as red circles. The two lobes of tissue polarity that surround this ptc clone are color coded the same as Fig. 7, except that the western lobe that is ventral to the equator is shaded a darker yellow than that part of the western lobe that is dorsal to the equator. Note that the eastern and western lobes occur as predicted by the two-factor model, and that they are reflected in the equator. Anterior is to the right, and dorsal is up. Scale is as Fig. 7C.

We draw two major conclusions from our data. (1) wg and ptc are negative regulators of furrow initiation (wg mainly on the dorsal margin and ptc in the eye field anterior to the endoge-nous furrow). (2) Reorientation of the morphogenetic furrow in wg and ptc mutant conditions can reorient tissue polarity in the eye. These new orientations suggest that ommatidial polarity and chiral form develop in response to induction by two independent pieces of positional information: the vector direction from the cluster to the furrow, and the vector direction from the cluster to the equator.

We have shown that the segment-polarity genes wg and ptc are negative regulators of the morphogenetic furrow in the developing Drosophila compound eye. Wg is strongly expressed on the dorsal margin of the presumptive eye field, anterior to the furrow, and removing its activity (with a temperature-sensitive mutation) causes the induction of an ectopic furrow on the dorsal margin, which moves downwards. Thus Wg acts in the developing eye as a negative regulator of furrow initiation. Previous studies have shown that the other segment polarity signaling molecule (Hh) is necessary (Heberlein et al., 1993; Ma et al., 1993) and sufficient (Heberlein et al., 1995) for furrow induction in the presumptive eye field. However, a second (Hh-independent) system is responsible for furrow initiation on the posterior margin of the eye field, and for its continuous reinitiation along the dorsal and ventral margins (Ma et al., 1993). Recently it has been shown that the dachshund (dac) gene is required for this furrow initiation on the posterior margin (Mardon et al., 1994). It may be that the wg-negative regulation acts through dac. While it is clear that Wg acts as a negative regulator of furrow initiation on the dorsal margin, it is not clear what may do so on the ventral margin. We observe only occasional and weak effects of removing wg function on the ventral side. It may be that other Wnt family proteins act ventrally (two more are known in Drosophila, Russell et al., 1992).

We have also shown that ptc acts as a negative regulator of furrow initiation, anterior to the endogenous furrow. When we remove ptc function, by means of genetic mosaic clones, ectopic circular furrows can be induced. These data are consistent with similar observations with ectopic hh expression (Heberlein et al., 1995), or loss of PKA (Pan and Rubin, 1995; Strutt et al., 1995). In simple terms, it would appear that an ectopic furrow can be induced in the anterior domain of the eye field either by adding Hh, or removing Ptc (or pathway elements downstream of it).

In our wg experiments, we observed that the developing ommatidial clusters generated by the ectopic furrow fail to rotate. This led us to formulate a two-vector model to account for normal rotation, and thus retinal tissue polarity and chirality. We propose that the nascent cluster can sense two vectors: the direction to the furrow and the direction to the equator. Normally these two vectors are perpendicular, and define the two chiral forms, and two orientations found in the right and left eyes. The ectopic wg- and ptc-induced furrows introduce cases of new directional values for the furrow vector, while the equatorial vector is constant. The model can account for the tissue chirality and polarity that we observe in most of these cases. The model is also consistent with the reflected forms seen near ectopic Hh and loss of PKA clones (Heberlein et al., 1995; Strutt et al., 1995). Which cells can sense these vectors? It may be that only the founding (R8) photoreceptor cell receives the two vector signals. Once the founding cell is so patterned, it may direct the development of its neighbors. While we cannot now suggest specific mechanisms by which such cell polarity is maintained in this case, four such systems are known from Drosophila oogenesis alone (reviewed in St. Johnston and Nüsslein-Volhard, 1992).

What are the mechanisms that underlie these two vectors? We can exclude the direct or continuous participation of wg itself, as clusters generated by the endogenous (vertical) furrow, continue to rotate normally when wg function has been removed. However, it is intriguing that some genes in the wg signaling pathway in the embryo, also affect tissue polarity in the eye (Gubb, 1993) and other tissues (Theisen et al., 1994). This information system does not require the close proximity of other clusters, as isolated ommatidia (in Ellipse) rotate normally (Baker and Rubin, 1992). It is not necessary to invoke any long-range influence to account for the furrow vector: as each cluster has had direct contact with the furrow, it may be imprinted directly by the furrow. However, they are at some distance from the equator and a different mechanism must provide the equatorial vector. Gubb (1993) has suggested that the developing clusters may sense the direction of the equator through a propagative mechanism in the furrow. All clusters are not formed in one column simultaneously in the furrow; rather, there is an initial event, at the intersection of the equator and the furrow (the ‘firing center’), and then clusters are formed successively in both directions along the furrow, at approximately 10 minute intervals (Wolff and Ready, 1991). This progression along the furrow is reflected in the progressive expression of the Scabrous protein (Baker et al., 1990; Mlodzik et al., 1990). Thus each cluster experiences progression from the firing center on one side and Gubb (1993) suggests that this could provide information for the direction to the equator. However, our data appear not to support this model for two reasons. First, the clusters generated by the wg ectopic furrow would sense the normal propagation from the firing center, along the ectopic furrow, and would then rotate 90° counterclockwise (but they do not). Second, the ectopic ptc-induced furrows are not connected at all to the equator and thus should have no firing center. Thus the clusters generated by the ptc furrows should not rotate (but they do). Thus we suggest that the clusters respond to a long-range polarity field to sense the equatorial vector. This could be based on a gradient of some substance (as in yeast shmooing, Segall, 1993), or polarity may be established earlier in development and maintained as an imprint (as in the Drosophila oocyte, reviewed by St Johnston and Nüsslein-Volhard, 1992).

We thank Nick Baker, Amy Bejsovec, Ron Blackman, Nadine Brown, Sean Carroll, Joan Hooper, Alfonso Martinez-Arias and Roel Nusse for stocks and materials, Nick Baker, David Gubb, Larry Marsh, Vivian Siegel and Rahul Warrior for their critical discussion, and Leslie Bell, Margaret McFall-Ngai and Rahul Warrior for their comments on this manuscript. This work was supported by a grant to K. M. from the NIH/National Eye Institute.

Baker
,
N. E.
(
1987
).
Molecular cloning of sequences from wingless, a segment polarity gene in Drosophila: the spatial distribution of a transcript in embryos
.
EMBO J
.
6
,
1765
1773
.
Baker
,
N. E.
(
1988a
).
Transcription of the segment-polarity gene wingless in the imaginal discs of Drosophila, and the phenotype of a pupal-lethal wg mutation
.
Development
102
,
489
497
.
Baker
,
N. E.
(
1988b
).
Embryonic and imaginal requirements for wingless, a segment-polarity gene in Drosophila
.
Dev. Biol
.
125
,
96
108
.
Baker
,
N. E.
,
Mlodzik
,
M.
and
Rubin
,
G. M.
(
1990
).
Spacing differentiation in the developing Drosophila eye: a fibrinogen-related lateral inhibitor encoded by scabrous
.
Science
250
,
1370
1377
.
Baker
,
N. E.
and
Rubin
,
G. M.
(
1989
).
Effect on eye development of dominant mutations in Drosophila homologue of the EGF receptor
.
Nature
340
,
150
153
.
Baker
,
N. E.
and
Rubin
,
G. M.
(
1992
).
Ellipse mutations in the Drosophila homologue of the EGF receptor affect pattern formation, cell division, and cell death in eye imaginal discs
.
Dev. Biol
.
150
,
381
396
.
Bejsovec
,
A.
and
Wieschaus
,
E.
(
1993
).
Segment polarity gene interactions modulate epidermal patterning in Drosophila embryos
.
Development
119
,
501
517
.
Blackman
,
R. K.
,
Sanicola
,
M.
,
Raferty
,
L. A.
,
Gillevet
,
T.
and
Gelbart
,
W. M.
(
1991
).
An extensive 3′ cis-regulatory region directs the imaginal disk expression of decapentaplegic, a member of the TGF-β family in Drosophila
.
Development
111
,
657
665
.
Blair
,
S. S.
(
1995
).
Hedgehog digs up an old friend
.
Nature
373
,
656
657
.
Bryant
,
P. J.
(
1993
).
The polar coordinate model goes molecular
.
Science
259
,
471
472
.
Capdevila
,
J.
,
Pariente
,
F.
,
Sampedro
,
J.
,
Alonso
,
J. L.
and
Guerrero
,
I.
(
1994
).
Subcellular localization of the segment polarity protein patched suggests an interaction with the wingless reception complex in Drosophila embryos
.
Development
120
,
987
998
.
Carroll
,
S. B.
and
Whyte
,
J. S.
(
1989
).
The role of the hairy gene during Drosophila morphogenesis: stripes in imaginal discs
.
Genes Dev
.
3
,
905
916
.
Chen
,
T. Y.
(
1929
).
On the development of imaginal buds in normal and mutant Drosophila melanogaster
.
J. Morph. Physiol
.
47
,
135
199
.
Choi
,
K.-W.
and
Benzer
,
S.
(
1994
).
Rotation of photoreceptor clusters in the developing Drosophila eye requires the nemo gene
.
Cell
78
,
125
136
.
Chu-LaGraff
,
Q.
and
Doe
,
C. Q.
(
1993
).
Neuroblast specification and formation regulated by wingless in the Drosophila CNS
.
Science
261
,
1594
1597
.
Cohen
,
S. M.
and
Di Nardo
,
S.
(
1993
).
wingless: from embryo to adult
.
Trends Genet
.
9
,
189
192
.
Couso
,
J. P.
and
Martinez Arias
,
A.
(
1994
).
Notch is required for wingless signaling in the epidermis of Drosophila
.
Cell
79
,
259
272
.
Dietrich
,
W.
(
1909
).
Die facettenaugen der Dipteren
.
Z. Wiss. Zool
.
92
,
465
539
.
Gibson
,
G.
and
Gehring
,
W. J.
(
1988
).
Head and thoracic transformations caused by ectopic expression of Antennapedia during Drosophila development
.
Development
102
,
657
675
.
Golic
,
K. G.
and
Lindquist
,
S.
(
1989
).
The Flp recombinase of yeast catalyzes site-specific recombination in the Drosophila genome
.
Cell
59
,
499
509
.
Gubb
,
D.
(
1993
).
Genes controlling cellular polarity in Drosophila
.
Development
1993
Supplement
,
269
-
277
.
Hazelrigg
,
T.
,
Levis
,
R.
and
Rubin
,
G. M.
(
1984
).
Transformation of white locus DNA in Drosophila: dosage compensation, zeste interaction, and position effects
.
Cell
36
,
469
481
.
Heberlein
,
U.
,
Singh
,
C. M.
,
Luk
,
A. Y.
and
Donohoe
,
T. J.
(
1995
).
Growth and differentiation in the Drosophila eye coordinated by hedgehog
.
Nature
373
,
709
711
.
Heberlein
,
U.
,
Wolff
,
T.
and
Rubin
,
G. M.
(
1993
).
The TGFβ homolog dpp and the segment polarity gene hedgehog are required for propagation of a morphogenetic wave in the Drosophila retina
.
Cell
75
,
913
926
.
Hidalgo
,
A.
(
1991
).
Interactions between segment polarity genes and the generation of the segmental pattern in Drosophila
.
Mech. Dev
.
35
,
77
87
.
Hooper
,
J. E.
and
Scott
,
M. P.
(
1989
).
The Drosophila patched gene encodes a putative membrane protein required for segmental patterning
.
Cell
59
,
751
765
.
Ingham
,
P. W.
and
Martinez-Arias
,
A.
(
1992
).
Boundaries and fields in early embryos
.
Cell
68
,
221
235
.
Ingham
,
P. W.
,
Taylor
,
A. M.
and
Nakano
,
Y.
(
1991
).
Role of the Drosophila patched gene in positional signaling
.
Nature
353
,
184
186
.
Jiang
,
J.
and
Struhl
,
G.
(
1995
).
Protein kinase A and hedgehog signaling in Drosophila limb development
.
Cell
80
,
563
572
.
Kaphingst
,
K.
and
Kunes
,
S.
(
1994
).
Pattern formation in the visual centers of the Drosophila brain: wingless acts via decapentaplegic to specify the dorsoventral axis
.
Cell
78
,
437
448
.
Krafka
,
J. J.
(
1924
).
Development of the compound eye of Drosophila melanogaster and its Bar-eyed mutant
.
Biol. Bull. Mar. Biol. Lab., Woods Hole
44
,
143
149
.
Lepage
,
T.
,
Cohen
,
S. M.
,
Diaz-Benjumea
,
F. J.
and
Parkhurst
,
S. M.
(
1995
).
Signal transduction by cAMP-dependent protein kinase A in Drosophila limb patterning
.
Nature
373
,
711
715
.
Li
,
W.
,
Ohlmeyer
,
J. T.
,
Lane
,
M. E.
and
Kalderon
,
D.
(
1995
).
Function of protein kinase A in Hedgehog signal transduction and Drosophila imaginal disc development
.
Cell
80
,
553
562
.
Lindsley
,
D. L.
and
Zimm
,
G. G.
(
1992
).
The Genome of Drosophila melanogaster
.
San Diego
:
Academic Press
.
Ma
,
C.
,
Zhou
,
Y.
,
Beachy
,
P. A.
and
Moses
,
K.
(
1993
).
The segment polarity gene hedgehog is required for progression of the morphogenetic furrow in the developing Drosophila eye
.
Cell
75
,
927
938
.
Mardon
,
G.
,
Solomon
,
N. M.
and
Rubin
,
G. M.
(
1994
).
dachshund encodes a nuclear protein required for normal eye and leg development in Drosophila
.
Development
120
,
3473
3486
.
Martinez Arias
,
A.
(
1994
).
Pathways of cell communication during development: signaling and epistases
.
Trends Genet
.
10
,
219
222
.
Martinez-Arias
,
A.
,
Baker
,
N. E.
and
Ingham
,
P. W.
(
1988
).
Role of segment polarity genes in the definition and maintenance of cell states in the Drosophila embryo
.
Development
103
,
157
170
.
Melamed
,
J.
and
Trujillo-Cenóz
,
O.
(
1975
).
The fine structure of the eye imaginal disks in Muscoid flies
.
J. Ultrastr. Res
.
51
,
79
93
.
Mlodzik
,
M.
,
Baker
,
N. E.
and
Rubin
,
G. M.
(
1990
).
Isolation and expression of scabrous, a gene regulating neurogenesis in Drosophila
.
Genes Dev
.
4
,
1848
1861
.
Moses
,
K.
,
Ellis
,
M. C.
and
Rubin
,
G. M.
(
1989
).
The glass gene encodes a zinc-finger protein required by Drosophila photoreceptor cells
.
Nature
340
,
531
536
.
Nakano
,
Y.
,
Guerrero
,
I.
,
Hidalgo
,
A.
,
Taylor
,
A.
,
Whittle
,
J. R. S.
and
Ingham
,
P. W.
(
1989
).
A protein with several possible membrane-spanning domains encoded by the Drosophila segment polarity gene patched
.
Nature
341
,
508
512
.
Noordermeer
,
J.
,
Johnston
,
P.
,
Rijsewijk
,
F.
,
Nusse
,
R.
and
Lawrence
,
P. A.
(
1992
).
The consequences of ubiquitous expression of the wingless gene in the Drosophila embryo
.
Development
116
,
711
719
.
Nusse
,
R.
and
Varmus
,
H. E.
(
1992
).
Wnt genes
.
Cell
69
,
1073
1087
.
Pan
,
D.
and
Rubin
,
G. M.
(
1995
).
cAMP-dependent protein kinase and hedgehog act antagonistically in regulating decapentaplegic transcription in Drosophila imaginal discs
.
Cell
80
,
543
552
.
Peifer
,
M.
and
Bejsovec
,
A.
(
1992
).
Knowing your neighbors: cell interactions determine intrasegmental patterning in Drosophila
.
Trends Genet
.
8
,
243
249
.
Peifer
,
M.
,
Orsulic
,
S.
,
Pai
,
L.-M.
and
Loureiro
,
J.
(
1993
).
A model system for cell adhesion and signal transduction in Drosophila
.
Development
1993
Supplement
,
163
-
176
.
Peifer
,
M.
,
Pai
,
L.-M.
and
Casey
,
M.
(
1994
).
Phosphorylation of the Drosophila adherens junction protein Armadillo: roles for Wingless signal and Zeste-white 3 kinase
.
Dev. Biol
.
166
,
543
556
.
Perrimon
,
N.
(
1994
).
The genetic basis of patterned baldness in Drosophila
.
Cell
76
,
781
784
.
Phillips
,
R. G.
,
Roberts
,
I. J. H.
,
Ingham
,
P. W.
and
Whittle
,
J. R. S.
(
1990
).
The Drosophila segment polarity gene patched is involved in a position-signaling mechanism in imaginal discs
.
Development
110
,
105
114
.
Ready
,
D. F.
,
Hanson
,
T. E.
and
Benzer
,
S.
(
1976
).
Development of the Drosophila retina, a neurocrystalline lattice
.
Dev. Biol
.
53
,
217
240
.
Rijsewijk
,
F.
,
Schuermann
,
M.
,
Wagenaar
,
E.
,
Parren
,
P.
,
Weigel
,
D.
and
Nusse
,
R.
(
1987
).
The Drosophila homolog of the mouse mammary oncogene int-1 is identical to the segment polarity gene wingless
.
Cell
50
,
649
657
.
Robinow
,
S.
and
White
,
K.
(
1991
).
Characterization and spatial distribution of the ELAV protein during Drosophila melanogaster development
.
J. Neurobiol
.
22
,
443
461
.
Russell
,
J.
,
Gennissen
,
A.
and
Nusse
,
R.
(
1992
).
Isolation and expression of two novel Wnt/wingless gene homologues in Drosophila
.
Development
115
,
475
485
.
Schneuwly
,
S.
,
Klemenz
,
R.
and
Gehring
,
W. J.
(
1987
).
Redesigning the body plan of Drosophila by ectopic expression of the homoeotic gene Antennapedia
.
Nature
325
,
816
818
.
Schuske
,
K.
,
Hooper
,
J. E.
and
Scott
,
M. P.
(
1994
).
patched overexpression causes loss of wingless expression in Drosophila embryos
.
Dev. Biol
.
164
,
300
311
.
Segall
,
J. E.
(
1993
).
Polarization of yeast cells in spatial gradients of α mating factor
.
Proc. Natl. Acad. Sci. USA
90
,
8332
8336
.
St. Johnston
,
D.
and
Nüsslein-Volhard
,
C.
(
1992
).
The origin of pattern and polarity in the Drosophila embryo
.
Cell
68
,
201
219
.
Struhl
,
G.
and
Basler
,
K.
(
1993
).
Organizing activity of wingless protein in Drosophila
.
Cell
72
,
527
540
.
Strutt
,
D. I.
,
Wiersdorff
,
V.
and
Mlodzik
,
M.
(
1995
).
Regulation of furrow progression in the Drosophila eye by cAMP-dependent protein kinase A
.
Nature
373
,
705
709
.
Tautz
,
D.
and
Pfeifle
,
C.
(
1989
).
A non-radioactive in situ hybridization method for the localization of specific RNAs in Drosophila embryos reveals translational control of the segmentation gene hunchback
.
Chromosoma
98
,
81
85
.
Tearle
,
R.
and
Nüsslein-Volhard
(
1987
).
Tübingen mutants and stocklist
.
Dros. Inf. Serv
.
66
,
209
269
.
Theisen
,
H.
,
Purcell
,
J.
,
Bennett
,
M.
,
Kansagara
,
D.
,
Syed
,
A.
and
Marsh
,
J. L.
(
1994
).
disheveled is required during wingless signaling to establish both cell polarity and cell identity
.
Development
120
,
347
360
.
Thomas
,
B. J.
,
Gunning
,
D. A.
,
Cho
,
J.
and
Zipursky
,
S. L.
(
1994
).
Cell cycle progression in the developing Drosophila eye: roughex encodes a novel protein required for the establishment of G1
.
Cell
77
,
1003
1014
.
Tomlinson
,
A.
(
1985
).
The cellular dynamics of pattern formation in the eye of Drosophila
.
J. Embryol. Exp. Morph
.
89
,
313
331
.
Tomlinson
,
A.
(
1988
).
Cellular interactions in the developing Drosophila eye
.
Development
104
,
183
193
.
Tomlinson
,
A.
and
Ready
,
D. F.
(
1987
).
Neuronal differentiation in the Drosophila ommatidium
.
Dev. Biol
.
120
,
366
376
.
van den Heuvel
,
M.
,
Klingensmith
,
J.
,
Perrimon
,
N.
and
Nusse
,
R.
(
1993
).
Cell patterning in the Drosophila segment: engrailed and wingless antigen distributions in segment polarity mutant embryos
.
Development
1993
Supplement
,
105
-
114
.
Vincent
,
J.-P.
and
Lawrence
,
P. A.
(
1994
).
It takes three to distalize
.
Nature
372
,
132
133
.
Waddington
,
C. H.
and
Perry
,
M. M.
(
1960
).
The ultra-structure of the developing eye of Drosophila
.
Proc. Roy. Soc. Lond. B
153
,
155
178
.
Wolff
,
T.
and
Ready
,
D. F.
(
1991
).
The beginning of pattern formation in the Drosophila compound eye: the morphogenetic furrow and the second mitotic wave
.
Development
113
,
841
850
.
Xu
,
T.
and
Rubin
,
G. M.
(
1993
).
Analysis of genetic mosaics in developing and adult Drosophila tissues
.
Development
117
,
1223
1237
.