The cut locus is a complex gene whose function is necessary for specification of a number of cell types, including the external sensory organs. The cut wing class of mutations of the cut locus are homozygous viable and lack tissue from the wing margin, which is normally composed of external sensory organs and noninnervated bristles. Expression of cut was examined in the developing wings of wild-type and mutant pupae using an antiserum against Cut protein. Cut is expressed in all of the external sensory organs of the wing and the noninnervated bristles of the posterior margin. The cut wing class of mutations prevents Cut expression specifically in the wing margin mechanoreceptors and noninnervated bristles, apparently preventing neural differentiation. The transformed cells die soon after differentiation would have occurred. We identify an enhancer, located about 80 kb upstream of the cut gene promoter, that confers expression in the cells of the mechanoreceptors and noninnervated bristles from a heterologous promoter. The 27 gypsy retrotransposon insertions that prevent expression in these margin cells, all occur between this enhancer and the promoter. These, gypsy insertions probably interfere with the interaction between the enhancer and the cut gene promoter.

The cut gene acts during differentiation to specify the cell types of at least two different tissues, the external sensory organs and the Malpighian tubules. Mutations of the transcribed region of the gene transform the external sensory organs of both adults and larvae into internal chordotonal organs (Bodmer et al. 1987), and the Malpighian tubules into gut (Liu et al. 1991). The gene product is a homeobox-containing protein that is expressed exclusively in the tissues affected by the mutations (Blochlinger et al. 1988). Therefore, the most likely function of the gene is to carry out a program of tissue-specific gene regulation leading to the proper morphological differentiation of the external sensory organs and the Malpighian tubules. The cut gene may also function in differentiation of the other tissues in which it is expressed, but its function in those tissues has not been characterized.

Mutations upstream of the transcribed region affect tissue-specific regulation of cut, altering expression in subsets of the tissues affected by mutations in the transcribed portion. Most upstream mutations are DNA rearrangements – deletions, inversions or insertions of transposable elements. The tissues affected by the various mutations have been described in detail (Blochlinger et al. 1990; Jack, 1985; Johnson and Judd, 1979; Liu et al. 1991). In this work we concentrate on the type of mutation for which the cut locus was named, the cut wing class of mutations. Flies homozygous for cut wing mutations suffer loss of tissue from the entire wing margin. The anterior margin to the tip is normally populated entirely by sensory cells, and the posterior margin is composed of noninnervated bristle-forming cells. Because cut is required for the specification of cell type for other external sensory organs, we have explored the possibility that the loss of wing margin cells in cut mutants is caused by failure of proper differentiation in external sensory organs of the margin.

The role of cut in the differentiation of the wing margin sense organs differs somewhat from its function in other external sensory organs. Lack of cut function in the wing margin eventually results in cell death rather than transformation to chordotonal organs. To understand the role of the cut gene in the wing margin, we have studied its expression and the development of the wing in wild-type and cut mutant flies during late larval life through the first two days of pupation. We have also examined the regulatory elements that control cut expression in the wing margin. We find that cut wing mutations prevent the function of an enhancer that controls expression in the wing margin mechanoreceptors and noninnervated bristles. In these mutants, the cells do not differentiate into sense organs and ultimately die, producing cut wings. The effect of gypsy transposon insertions on the enhancer element are discussed.

Drosophila stocks

All cut mutations used have been described in Jack (1985) and Blochlinger et al. (1988). sc10–1 is described in Lindsley and Grell (1968). Larvae and pupae for dissection were raised at 25° on standard cornmeal-molasses medium in uncrowded conditions.

Immunohistochemistry

Larval imaginai discs and pupal wings were dissected in PBS and fixed for 15–30 min in 3% formalin in PBS. They were then washed for 30min in three changes of PBS, 0.2 % BSA, 0.1% Triton X-100 (PBT). The tissue was incubated with primary antibody for one hour to overnight in PBT+2% normal goat serum (PBTN). Dilutions of antisera were: clp2 1:2000, anti-β-galactosidase (Cappel) 1:4000. Tissues were then washed for 30 min to overnight in PBS and incubated for one hour with horseradish-peroxidase-conjugated goat antirabbit IgG diluted 1:100 in PBTN. After washing 45 min in three changes of PBS, the wings were placed in 2 ml of PBT. 2 ml of 1 mg ml-1 diaminobenzidine in 0.12 M Tris, pH 7.6 were added, the wings were incubated for 10min. Hydrogen peroxide was added to a concentration of 0.0015 %, and the staining was allowed to develop. The reaction was stopped by washing two times in PBT. The wings were then washed two more times in PBS and mounted under a coverslip in 90 % glycerol, 0.1M Tris, pH 7.6.

Construction of ctwHZ transformant lines

The 2.7 kb EcoRI–BamHI restriction fragment from about coordinate −5 on the cut locus map was blunted with Klenow fragment of DNA polymerase and cloned into the T4 DNA polymerase-blunted Kpn\ site of the HZ50 P element transformation vector (Hiromi and Gehring, 1987). Germline transformation of ry506 was performed as described previously (Dorsett et al. 1989).

In Drosophila larvae, cells that will eventually form the adult are sequestered in imaginai discs. Each of the two wing discs gives rise to one wing and half of the dorsal mesothorax. A simplified version of the third instar wing disc fate map (Bryant, 1975) is shown (Fig. 1A), along with a diagram of an adult wing for comparison (Fig. 1B). At the end of the third instar, the larva pupariates and metamorphosis begins. During the first three hours after pupariation (AP), the discs evaginate. In the case of the wing disc, the central part of the disc projects outward to form the wing blade, and the rest of the disc forms dorsal mesothorax, notum and pleural plates. The wing continues to lengthen and, by about 4 h AP, it flattens into two sheets of cells, one dorsal and one ventral, each a single cell layer thick. At around 42–48 h AP, wing cells begin to secrete the adult cuticle, which forms the structure of the adult wing after the nonneuronal cells of the wing die.

Fig. 1.

Diagram of wing disc fate map and adult wing. (A) Fate map of the wing imaginai disc adapted from Bryant (1975). (B) Adult dorsal wing. Abbreviations: AC, Axillary cord; ACV, Campaniform sensillum of the anterior cross vein; AL, Alula; DCo, Distal costa; dHCV, Dorsal campaniform sensillum of the humeral cross vein; DR, Double bristle row; GSR, Giant sensillum of the radius; LI,2,3,4,5, Veins; L3-l,2,3, Campaniform sensillae of L3; MCo, Medial costa; PCo, Proximal costa; PR, Posterior bristle row; Teg, Tegula; TR, Triple bristle row; TSM, Twin sensilla of the margin.

Fig. 1.

Diagram of wing disc fate map and adult wing. (A) Fate map of the wing imaginai disc adapted from Bryant (1975). (B) Adult dorsal wing. Abbreviations: AC, Axillary cord; ACV, Campaniform sensillum of the anterior cross vein; AL, Alula; DCo, Distal costa; dHCV, Dorsal campaniform sensillum of the humeral cross vein; DR, Double bristle row; GSR, Giant sensillum of the radius; LI,2,3,4,5, Veins; L3-l,2,3, Campaniform sensillae of L3; MCo, Medial costa; PCo, Proximal costa; PR, Posterior bristle row; Teg, Tegula; TR, Triple bristle row; TSM, Twin sensilla of the margin.

The adult wing margin, which is the boundary of the dorsal arid ventral cell layers as well as the dorsal–ventral compartment boundary, is covered by rows of bristles. The anatomy and development of the bristle rows has been described in detail (Hartenstein and Posakony, 1989; Murray et al. 1984; Palka et al. 1979). Each bristle is composed of a cuticular shaft, a socket and a cap over the tip of the dendrite of a bipolar neuron. The shaft is secreted by a trichogen cell located on the margin, the socket by the tormogen cell located next to the trichogen, one cell away from the margin, and the dendritic cap by a thecogen cell. The fourth cell body from the margin is the neuron (Hartenstein and Posakony, 1989). The anterior margin has three rows of bristles. Of these, the dorsal row is composed of sparsely distributed, multiply innervated, curved chemoreceptive bristles. The medial bristle row is composed exclusively of thick shafted, singly innervated mechanoreceptors, and the ventral row has both multiply innervated chemoreceptors and thin shafted, singly innervated mechanoreceptor bristles with an average of four mechanoreceptors between each chemoreceptor. The chemoreceptors of both rows have recurved hollow bristles and five neurons. One dendrite terminating at the base of the shaft is probably a mechanoreceptor, and the other four dendrites continue up the shaft and probably function as chemoreceptors (Palka et al. 1979). The margin around the wing tip has two identical rows similar in composition to the ventral row of the anterior margin. The posterior margin is composed of two rows of noninnervated hairs, similar in size and density to the thin bristles of the ventral row of the anterior margin. No socket is present on these bristles, but a tormogen-like cell surrounds the trichogen as in the innervated bristles (Hartenstein and Posakony, 1989).

Phenotype of cut wing mutations

Cut wing type mutations of the cut locus produce gaps in. the wing margins in flies homozygous for the mutations. The weakest of the mutations cause only small gaps in the margin at the tip, while the strongest mutations cause the loss of most of the bristles from the entire wing margin as well as some cells from the peripheral part of the wing blade, a loss of about 16 % of the wing cells altogether (Santamaria and Garcia-Bellido, 1975).

Cut protein is expressed early in the differentiation process in all cells of the sense organs on the wing and in the noninnervated bristles of the posterior wing margin. In cut wing mutants, protein is not expressed in the mechanoreceptors or noninnervated bristles of the margin but is still expressed in the other sense organs. The next two sections present a detailed description of the wild-type expression pattern.

Wild-type cut expression in third instar larvae and prepupae

We examined expression of cut gene product in wing discs of late third instar larvae and in wings during the first two days of pupation by histochemical staining using an antibody, clp2, against a Cut peptide (Bloch-linger et al. 1988). Expression of Cut protein is first detected in crawling third instar larvae. Cut staining cells form an omega-shaped arc three to four cells wide in the position of the presumptive wing margin (Fig. 2A). The ends of the arc continue beyond the wing margin, perhaps including the dorsal-ventral compartment boundary of the thoracic portion of the disc. Concurrently, a number of the presumptive sense organs on the wing blade begin to stain with the antibody.

Fig. 2.

Expression of Cut protein in wings of late third instar larvae and prepupae. The anti-Cut antibody was used to stain wing discs and prepupal wings. Right wings and discs are shown. (A) Late third instar larva. (B) White prepupa. Eversion of the disc has begun, moving the margin toward the top of the disc. Precursors of the wing blade campaniform sensillae and the TSM are labelled. (C)White prepupa stained with anti-Cut and mAb 22C10. The points flanking the margin stripe are mAb 22C10 labeling of the chemosensory organ precursors. (D) Prepupa 2h AP. (E) Prepupa 4h AP. The precursors of the campaniform sensillae L3-1 and L3-3 are just beginning to label. (F) Prepupa 6h AP. (G) Anterior margin of the wing at 7 h AP. Mechanoreceptor nuclei are on the margin, and the clusters of nuclei below the margin are the chemoreceptor nuclei. (H) Anterior margin of a 6h wing. The margin has been rolled to the top showing clusters of chemoreceptor nuclei flanking on both sides of the continuous stripe of mechanoreceptors nuclei. (I) Posterior margin of the wing in H. Only the future noninnervated bristles stain. Arrows in B, D, and E mark sense organs of the dorsal wing blade. In B and D, these are the dHCV, GSR, ACV- and L3-2 campaniform sensillae and the TSM bristles. Arrows in E show the L3-1 and L3-3 campaniform sensillae, which are just beginning to express Cut. Bar=50 μm.

Fig. 2.

Expression of Cut protein in wings of late third instar larvae and prepupae. The anti-Cut antibody was used to stain wing discs and prepupal wings. Right wings and discs are shown. (A) Late third instar larva. (B) White prepupa. Eversion of the disc has begun, moving the margin toward the top of the disc. Precursors of the wing blade campaniform sensillae and the TSM are labelled. (C)White prepupa stained with anti-Cut and mAb 22C10. The points flanking the margin stripe are mAb 22C10 labeling of the chemosensory organ precursors. (D) Prepupa 2h AP. (E) Prepupa 4h AP. The precursors of the campaniform sensillae L3-1 and L3-3 are just beginning to label. (F) Prepupa 6h AP. (G) Anterior margin of the wing at 7 h AP. Mechanoreceptor nuclei are on the margin, and the clusters of nuclei below the margin are the chemoreceptor nuclei. (H) Anterior margin of a 6h wing. The margin has been rolled to the top showing clusters of chemoreceptor nuclei flanking on both sides of the continuous stripe of mechanoreceptors nuclei. (I) Posterior margin of the wing in H. Only the future noninnervated bristles stain. Arrows in B, D, and E mark sense organs of the dorsal wing blade. In B and D, these are the dHCV, GSR, ACV- and L3-2 campaniform sensillae and the TSM bristles. Arrows in E show the L3-1 and L3-3 campaniform sensillae, which are just beginning to express Cut. Bar=50 μm.

As the wing begins to elongate from the pouch of the disc at pupariation, the stained margin moves upward with the wing tip. By the beginning of pupariation, the margin has advanced considerably from its position in late third instar larvae, and the earliest sense organs on the wing blade to stain are identifiable by their locations (Fig. 2B) as described in Murray et al. (1984). Staining with the early neural marker mAb 22C10 is also evident at pupariation in two rows along the future anterior margin and tip (Fig. 2C). mAb 22C10 recognizes an antigen of the peripheral nervous system (Canal and Ferrus, 1986; Hartenstein, 1988; Zipursky et al. 1984) and begins to stain early in differentiation of sense organs, before the mother cells divide (Hartenstein and Posakony, 1989). The two rows in the margin are the precursors of the chemosensory organs. Cut protein also begins to be expressed in the chemoreceptor cells soon after pupariation, but their nuclei are not always distinguishable from the other nuclei staining in the margin.

By 2h AP the presumptive wing margin cells have moved to the actual margin of the pupal wing (Fig. 2D). The wing continues to elongate and, at around 4 h AP, a group of cells that express Cut separate from the continuous stripe of cells directly on the margin (Fig. 2E). These are the precursors of the chemoreceptors. Also at 4 h AP, staining begins to appear in cells on the wing blade that are probably the precursors of the L3-1 and L3-3 campaniform sensillae. The detection of Cut protein in these cells occurs a few hours before the time that they have been reported to stain with antihorseradish peroxidase (anti-HRP) (Murray et al. 1984), another marker for neural differentiation in insects.

By 6h AP, the margin stripe is a continuous row approximately three cells wide running around the entire wing margin, flanked by approximately 22 pairs of cells on each side - dorsal and ventral - along the anterior and tip of the wing (Fig. 2F). The number and positions of cell pairs that flank the margin stripe correspond to the number and positions of the chemosensory organs present in the dorsal and ventral bristle rows of the adult wing (Hartenstein and Posakony, 1989; Palka et al. 1979). At 7–7.5 h AP the labelled nuclei appear as a single row around the entire margin of wings that are mounted flat (Fig. 2G). However, when the margin is rolled to the top and bottom of the wing, as is the case in Fig. 2H and I, a stripe three to four cells wide is apparent on the margin. These cells are the precursors of the mechanoreceptors of the anterior margin and wing tip and of the noninnervated bristles of the posterior margin.

The stained nuclei flanking the margin row increase in number from two nuclei per cluster at 4 h to three or four nuclei at 7–7.5 h, paralleling the proliferation of the neurons of the chemosensory organs, which has been reported to occur from 0 to 8h after puparium formation (Hartenstein and Posakony, 1989). These are clearly chemosensory organ precursors.

The timing of cut staining relative to 22C10 staining is notably different in wing margin mechanoreceptors than in other wing external sensory organs. mAb 22C10 begins to stain the membranes of most sense organ cells before the precursors begin to divide (Hartenstein and Posakony, 1989). For all the wing sense organs except the margin mechanoreceptors, staining with anti-Cut and 22C10 begins at about the same time. Points of 22C10 staining begin to be visible in the premitotic cells of the margin chemosensory organs at Oh AP, and by 6h AP the membranes of 3–4 neurons are stained in each organ (Hartenstein and Posakony, 1989, Fig. 2C). This is similar in time to anti-Cut staining in these cells.

However, the margin mechanoreceptor precursors, which stain with anti-Cut even before pupariation, do not begin to stain with 22C10 until 16–20 h AP (Hartenstein and Posakony, 1989).

Expression in pupal wings 18 to 42 h AP

The cells of the mechanosensory bristles of the wing margin are generated by two final rounds of mitosis occurring between 9 and 14 h AP (Hartenstein and Posakony, 1989). This cell division results in a large increase in the number of wing margin nuclei staining with anti-cut at 18 h AP (Fig. 3A). Along the anterior margin to the wing tip on both the dorsal and ventral cell layers, the nuclei of approximately the first four rows of cells from the margin are stained (Fig. 3B and C). These cells will form the mechanoreceptors of the middle and ventral bristle rows (Hartenstein and Posakony, 1989) and are apparently generated from the three rows of staining cells on the 6–7 h margin by two rounds of division. On the posterior margin, the two rows of cells nearest the margin on the dorsal layer and the two on the ventral layer are stained (Fig. 3D). These correspond in position to the bristle-secreting cells and the tormogen-like cells associated with them (Hartenstein and Posakony, 1989). Also at 18 h, the neurons of the dorsal chemosensory organs form clusters of five cells on the dorsal surface, separate from and slightly nearer the center of the blade than the mechanosensory nuclei (Fig. 3A–C). Each cluster consists of five neurons innervating a single chemosensory bristle. The tormogen and trichogen cell nuclei are stained more darkly and are located above the mechanosensory cells and adjacent to the chemosensory neurons (Fig. 3C). By 24 h AP, the chemosensory nuclei of both dorsal and ventral rows have moved toward the margin and are not always distinguishable from the mechanoreceptor cells in whole-mounted wings (Fig. 3E–G) although the tormogen and trichogen cells are still distinguishable (Fig. 3G). The two rows of cells in the posterior margin are organized in pairs at 24 h suggesting a physical relationship between the two cells.

Fig. 3.

Cut protein expression in wild-type wings. (A–D) 18h AP. (E–H) 24h AP. (I–L) 36h AP. (M–O) 42h AP. B, F, J, and N show the anterior margin with staining about four nuclei side around the margin. The same region is shown in a different plane of focus in C, G, and K. The tormogen and trichogen nuclei (solid arrows) of the dorsal row chemoreceptors can be seen staining darker than the mechanoreceptor nuclei below them. The open arrows show a groups of neurons for individual chemoreceptors. D, H, L, and O show two rows of stained cells on the posterior margin. These pairs of cells correspond to the bristle-forming cells and tormogen-like cells that are associated with them (Hartenstein and Posakony, 1989). In the 42h wings, the bristles are being secreted and have become clearly visible. Bar=50 μm. Panels without bar are the same as B.

Fig. 3.

Cut protein expression in wild-type wings. (A–D) 18h AP. (E–H) 24h AP. (I–L) 36h AP. (M–O) 42h AP. B, F, J, and N show the anterior margin with staining about four nuclei side around the margin. The same region is shown in a different plane of focus in C, G, and K. The tormogen and trichogen nuclei (solid arrows) of the dorsal row chemoreceptors can be seen staining darker than the mechanoreceptor nuclei below them. The open arrows show a groups of neurons for individual chemoreceptors. D, H, L, and O show two rows of stained cells on the posterior margin. These pairs of cells correspond to the bristle-forming cells and tormogen-like cells that are associated with them (Hartenstein and Posakony, 1989). In the 42h wings, the bristles are being secreted and have become clearly visible. Bar=50 μm. Panels without bar are the same as B.

By 36 h AP the cells of the wing margin sense organs have moved into the position in the wing that they will occupy in the adult. On the dorsal surface, the tormogen and trichogen nuclei of the dorsal row chemosensory organs are now visible, separate from the four rows of nuclei that include the cells of the mechanoreceptors and the neurons of the chemoreceptors (Fig. 3I–K). At 42 h AP the bristles are visible. The trichogen cells of the mechanoreceptors are now visible and elongated, and their nuclei clearly express cut (Fig. 3M–O). The nuclei of the trichogen cells are elevated above those of the other cells of the sense organs by the lengthening of the trichogen cells. The tormogen and trichogen cells of the chemoreceptors are no longer clearly distinguishable. After 42h, secretion of the adult cuticle prevents staining with the antiserum.

Effect of cut wing class mutations on Cut antigen expression

The mutation ctL-32 is caused by an insertion of the retrotransposon gypsy and has an extreme cut wing phenotype as well as missing vibrissae. The mutation blocks expression of Cut antigen in the mechanoreceptors and noninnervated bristles of the wing margin, but expression in the chemoreceptors along the margin and in the sensillae on the wing blade is unaffected. The loss of Cut expression in the wing margin results in failure of the mechanoreceptor and noninnervated bristle cells to differentiate, followed by degeneration of the cells, which results in the cut wing phenotype.

In white prepupae (Oh AP), ctL-32 wing discs completely lack staining in the presumptive wing margin but have normal staining in the cells that are precursors of the sense organs on the wing blade (Fig. 4A). At 7h AP in ctL-32, the precursors of chemosensory organs and wing blade sense organs are present and express Cut protein at levels similar to wild type, but no staining is evident in the presumptive mechanoreceptors of the margin (Fig. 4C, compare to Fig. 2F and G). The shape of the wing at this time is the same as wild type, suggesting that the cells are still present. To be certain that the margin cells are present at 0 and 7 h AP, wings of the ctL-32/ctDB7 genotype were stained with the anti-Cut antibody. The ctDB7 allele fails to complement cut wing mutations such as ctL-32. It produces a protein that lacks the homeobox, is not localized to the nucleus, and is presumably nonfunctional (Blochlinger et al. 1988; Liu et al. 1991). At 0 and 7 h AP, the ctL-32 /ctDB7 wings have cytoplasmic staining around the entire margin (Fig. 4B and D), demonstrating that the cells are still present.

Fig. 4.

Cut expression in ctL-32 wings. (A) ctL-32/ctL-32 wing disc Oh AP. No staining is visible on the future margin, but the presumptive wing blade sense organs stain (arrows). (B) ctL-32/ctDB7wing disc at Oh. AP. The staining in the margin is the cytoplasmically localized, nonfunctional product of the ctDB7 allele. The expression of the nonfunctional Cut protein demonstrates that the mechanoreceptor cells are present in mutants at this stage. (C) ctL-32/ctL-32 wing at 7h AP. Cut staining is apparent in the chemoreceptors near the margin and in the wing blade sensillae. However, staining in the presumptive mechanoreceptors on the margin is absent. (D) ctL-32/ctDB7 wing at 7 h AP. Cytoplasmic staining of the nonfunctional Cut protein demonstrates that the mechanoreceptor cells are still present. Bar=50 μm.

Fig. 4.

Cut expression in ctL-32 wings. (A) ctL-32/ctL-32 wing disc Oh AP. No staining is visible on the future margin, but the presumptive wing blade sense organs stain (arrows). (B) ctL-32/ctDB7wing disc at Oh. AP. The staining in the margin is the cytoplasmically localized, nonfunctional product of the ctDB7 allele. The expression of the nonfunctional Cut protein demonstrates that the mechanoreceptor cells are present in mutants at this stage. (C) ctL-32/ctL-32 wing at 7h AP. Cut staining is apparent in the chemoreceptors near the margin and in the wing blade sensillae. However, staining in the presumptive mechanoreceptors on the margin is absent. (D) ctL-32/ctDB7 wing at 7 h AP. Cytoplasmic staining of the nonfunctional Cut protein demonstrates that the mechanoreceptor cells are still present. Bar=50 μm.

At 18 h AP ctL-32 mutants lack staining in the margin cells that constitute the mechanoreceptors and nonin-nervated posterior bristles, while the staining of chemoreceptors is still evident (Fig. 5A and B). At this stage, the shape of the wing is normal. We wère unable to demonstrate that the mechanoreceptor precursors are still present using anti-Cut staining of ctL-32/ctDB7 as described above because 18 h wings of this genotype were difficult to handle. We identified the mutant mechanoreceptor cells at this stage by expression of β-galactosidase driven by a cut wing enhancer. These experiments are described in the final section.

Fig. 5.

Deterioration of the wing margin in ctL-32 /ctL-32 wings. (zX and B) Wing at 18 h AP stained with anti-Cut antibody. The shape of the wing has not begun to change at this stage. The chemoreceptors on the margin and sensillae on the blade are visible, but the other cells on the margin do not stain. Open arrows show the chemoreceptor neurons and solid arrows show a tormogen and trichogen cell. (C) Wing beginning to decay at 24h AP. Similar wings can be seen at 20–24h AP. (D) 36 h wing showing fully deteriorated margin. This shape is attained by 27 h AP. Bar=50 μm.

Fig. 5.

Deterioration of the wing margin in ctL-32 /ctL-32 wings. (zX and B) Wing at 18 h AP stained with anti-Cut antibody. The shape of the wing has not begun to change at this stage. The chemoreceptors on the margin and sensillae on the blade are visible, but the other cells on the margin do not stain. Open arrows show the chemoreceptor neurons and solid arrows show a tormogen and trichogen cell. (C) Wing beginning to decay at 24h AP. Similar wings can be seen at 20–24h AP. (D) 36 h wing showing fully deteriorated margin. This shape is attained by 27 h AP. Bar=50 μm.

Expression of Ag 22C10 in the mechanoreceptors and posterior bristles, normally present by 16 –20 h AP in wild type (Hartenstein and Posakony, 1989), is lost in the ctL-32 mutant flies (Fig. 6A and B), indicating that the cells do not differentiate into sense organs in the absence of cut expression. This loss of 22C10 expression in the absence of cut activity is unique to the wing margin mechanoreceptors. Outside the wing margin, 22C10 is expressed in external sensory organs and chordotonal organs, and also in cells that have been transformed from external sensory to chordotonal organs by loss of function cut mutations (Bodmer et al. 1987). By around 24 h AP the margins of many wings have begun to deteriorate substantially (Fig. 5C), and by 27 h the erosion is almost complete. At 36 h AP, wings of ctL-32 pupae are shaped exactly like wings of adult mutant flies (Fig. 5D). In regions where the margin has eroded well in from the position of the wildtype margin, however, the chemosensory organs are still present and situated on the mutant margin, although they appear somewhat disorganized (Fig. 5D). Accordingly, nearly all of the mechanosensory bristles on the adult ctL-32 wing margin are absent, but many of the chemosensory bristles are present even though the margin is severely disrupted.

Fig. 6.

Loss of expression of a neural antigen in ctL-32/ctL-32 wing margin mechanoreceptors. (A) mAb 22C10 staining of a wild-type wing at 18 h AP. The mechanoreceptors (closed arrow) and chemoreceptors (open arrow) are labeled as well as the axon bundle that will run down the LI vein. (B) 18 h ctL-32/ctL-32 wing stained with mAb 22C10. The chemoreceptors (open arrow) are the only cells in the wing margin that stain. Bar=50 μm.

Fig. 6.

Loss of expression of a neural antigen in ctL-32/ctL-32 wing margin mechanoreceptors. (A) mAb 22C10 staining of a wild-type wing at 18 h AP. The mechanoreceptors (closed arrow) and chemoreceptors (open arrow) are labeled as well as the axon bundle that will run down the LI vein. (B) 18 h ctL-32/ctL-32 wing stained with mAb 22C10. The chemoreceptors (open arrow) are the only cells in the wing margin that stain. Bar=50 μm.

Weak mutations of cut have an effect similar to ctL-32 but less extensive. In ct53d, for example, the margin stripe of Cut labelling is absent only at the tip of the wing. This corresponds well with the adult phenotype of ct53d, in which the wing margin has gaps primarily at the tip. Similar phenotypes are observed for ct461, ctn and ctNS, three weak cut wing mutations. In ct461 the margin stripe is absent only at the tip, but expression appears to be reduced all around the margin. However, adult ct461 wings usually have margin gaps only at the tip. Flies homozygous for the weak lethal mutation ctK survive as adults, and staining is reduced in both the mechanoreceptors and noninnervated bristles of the margin and the chemoreceptors. This is consistent with the fact that ctK is a member of the lethal I class of cut mutations (Jack, 1985), and strong lethal 1 mutations block cut expression throughout the peripheral nervous system (Bodmer et al. 1987).

Effect of mutations of the achaete-scute complex on Cut expression

The genes of the achaete-scute complex are necessary for the development of most larval and adult external sensory organs (Ghysen and Dambly-Chaudiere, 1988, for review). An exception is the wing margin mechanoreceptors. While the wing blade sensillae and chemoreceptors of the dorsal row have been reported to be absent when achaete. and scute are both mutant, the mechanoreceptors of the margin are unaffected (Gar-cia-Bellido and Santa Maria, 1978; Leyns et al. 1989). We have analyzed Cut protein expression in sc10–1, a double mutation that eliminates the activity of both achaete and scute (Campuzano et al. 1985; Villares and Cabrera, 1987). At 7 h AP in sc10–1 wings, cut protein is absent from all of the blade sensillae and both the dorsal and ventral margin chemosensory organs, but the margin stripe of mechanosensory precursors shows normal cut expression (Fig. 7). This expression pattern is complementary to that in ctL-32. Because we observed that both dorsal and ventral rows of chemosensory organs are absent in sc10–1 pupal wings, we examined wings of sc10–1 adults to determine whether all of the adult structures were missing. As reported previously (Garcia-Bellido and Santa Maria, 1978), the dorsal row is lost and, in addition, the chemosensory organs that are interspersed among the mechanoreceptors of the ventral bristle row are also missing. This confirms our conclusion that the clusters of Cut-expressing cells flanking the margin stripe are precursors of the chemosensory organs.

Fig. 7.

Lack of Cut expression in the chemoreceptors and wing blade sensillae of sc10–1 Bar=50 μm. The sc10–1 mutation blocks Cut expression in the sense organs that cut wing mutants do not affect. A 7h sc10–1 wing was stained with anti-Cut antibody. The mutation has no apparent effect on the mechanoreceptor precursors, but staining is absent in all the other sense organs of the wing. Bar=50 μm.

Fig. 7.

Lack of Cut expression in the chemoreceptors and wing blade sensillae of sc10–1 Bar=50 μm. The sc10–1 mutation blocks Cut expression in the sense organs that cut wing mutants do not affect. A 7h sc10–1 wing was stained with anti-Cut antibody. The mutation has no apparent effect on the mechanoreceptor precursors, but staining is absent in all the other sense organs of the wing. Bar=50 μm.

An enhancer sequence that activates cut expression in the cells affected by the cut wing mutations

Mutations of cut map over a 200 kb region shown in Fig. 8 (Jack, 1985). These mutations fall into groups that disrupt cut expression in different sets of tissues (Liu et al. 1991). However, two deletion mutations and an inversion breakpoint, together define a small region that is required for proper expression of cut in the wing margin but is not necessary in other tissues. The smallest of these deletions ct53d is a 0.5 kb deletion in the interval between −5.0 and −2.5, about 80kb upstream of the first exon. Three transposon insertion alleles have been mapped to the same interval. In addition, the mutations that fall between −2.5 kb on the map and +72, all gypsy insertions upstream of the most 5’ identified exon of the cut transcript, fail to complement the cut wing mutations and, thus, lack cut expression in the wing margin mechanoreceptors and posterior bristles. Many of these gypsy insertion mutations lack expression of cut in other tissues as well (Fig. 8). One explanation for these observations is that a transcription enhancer activating cut expression in the wing margin is located near −5 kb and is at least partially deleted by ct53d. If this is the case, then the gypsy insertions between −2.5 and +78 kb must either act as barriers between the enhancer and promoter, or other sequences between 0 and +72 are also required for wing margin expression and are interrupted by the gypsy insertions.

Fig. 8.

Correlation of the tissue affected with the map positions of cut mutations. The map shows the location of mutations within the gene. The coordinates are in kilobases. The regions in which mutations map are shown as blocks on the map. All of the mutations between 0 and +80 are insertions of the retrotransposon gypsy. The tissues affected or phenotypes of the mutations are shown above the map. Phenotypes: (kinked femur) bent legs, unexpanded wings and small body; (Wing) lack of cut expression in the wing margin mechanoreceptors; (Vibrissae) missing vibrissae; (Larval lethal) undefined larval lethality; (Malpighian tubules, Spiracles, ES organs=external sensory organs, CNS=central nervous system) mutations prevents expression in these tissues in embryos, and the resulting phenotypes are described in the text.

Fig. 8.

Correlation of the tissue affected with the map positions of cut mutations. The map shows the location of mutations within the gene. The coordinates are in kilobases. The regions in which mutations map are shown as blocks on the map. All of the mutations between 0 and +80 are insertions of the retrotransposon gypsy. The tissues affected or phenotypes of the mutations are shown above the map. Phenotypes: (kinked femur) bent legs, unexpanded wings and small body; (Wing) lack of cut expression in the wing margin mechanoreceptors; (Vibrissae) missing vibrissae; (Larval lethal) undefined larval lethality; (Malpighian tubules, Spiracles, ES organs=external sensory organs, CNS=central nervous system) mutations prevents expression in these tissues in embryos, and the resulting phenotypes are described in the text.

To test the hypothesis that ct53d deletes an enhancer, we cloned a wild-type 2.7 kb EcoRI-BamHI fragment spanning the region affected by the ct53d deletion into HZ50 (Hiromi and Gehring, 1987), a P element transformation vector containing a minimal hsp70 promoter upstream of the lacZ gene. This vector allows a cloned enhancer sequence to activate expression of the hsp70-lacZ gene, ry506 flies were transformed with the HZ50 construct containing the cut 2.7 kb EcoRI-BamHI fragment. Two separate transformed lines were obtained. Line ctwHZ-1 had a ry+ insertion into chromosome 3, and ctwHZ-2 had an insertion into chromosome 2. Expression of lacZ was detected in wing discs using anti-β-galactosidase.

In ctwHZ-2, β-galactosidase is expressed in 0h prepupae in the same cells that express cut in the wing margin, although the β-galactosidase is not localized to the nucleus (Fig. 9A, compare to Fig. 2B and Fig. 4B). β-galactosidase is not expressed in the sense organ precursors on the blade, which are not affected by cut wing mutations. This is consistent with expression specifically in the wing margin mechanoreceptors and noninnervated sense organs, although the chemoreceptor cells would not be distinguishable at this stage. However, at 7 h AP, no expression in the wing margin chemoreceptors has become apparent (Fig. 9C). The transformant ctwHZ-1 expressed β-galactosidase in most of the presumptive wing margin but lacked expression in the anterior and posterior wing hinge region (Fig. 9B and D). Nevertheless, since both express j5-galactosidase in the same cells of the wing margin that lose cut expression in cut wing mutants, the 2.7 kb fragment must contain a sequence sufficient to activate expression in these cells. The activity of the sequence is apparently modified by genomic sequences to cause the difference in expression in the wing hinge region. The adjacent genomic sequences must either activate the lacZ gene in the hinge region of ctwHZ-2 or repress it in ctwHZ-1. We believe that the latter is much more likely since activating the gene in the hinge region of the margin would be a much more specific effect.

Fig. 9.

Wing margin expression of β-galactosidase in a construct containing a cut wing region DNA fused to hsp70–lacZ. The fusion constructs are described in the text. A 2.7 kb EcoRI–BamHI fragment from near −5 kb in Fig. 8 was cloned into the transformation vector HZ50 (Hiromi and Gehring, 1987), upstream of the fusion of the hsp70 minimal promoter fused to lacZ. All wings are stained with anti-β-galactosidase antibody. (A) Oh prepupal wing disc from the ctwHZ-2 transformant line, β-galactosidase is expressed throughout the wing margin including the hinge region. All of the wing margin mechanoreceptors and noninnervated bristles that normally express Cut are labeled with anti-β-kgalactosidase. (B) Oh prepupal wing disc from the transformant line ctwHZ-1. β-galactosidase is expressed in all of the cells of the presumptive wing margin but not in the hinge region. (C) 7h ctwHZ-2 prepupal wing. Anti-β-galactosidase labeling is evident in all of the wing margin mechanoreceptors and noninnervated bristles. (D) 7h ctwHZ-1 prepupal wing. At this stage β-galactosidase expression can be detected along the margin and continuing into the hinge region, but the staining in the hinge region is very light. Bar=50 μm.

Fig. 9.

Wing margin expression of β-galactosidase in a construct containing a cut wing region DNA fused to hsp70–lacZ. The fusion constructs are described in the text. A 2.7 kb EcoRI–BamHI fragment from near −5 kb in Fig. 8 was cloned into the transformation vector HZ50 (Hiromi and Gehring, 1987), upstream of the fusion of the hsp70 minimal promoter fused to lacZ. All wings are stained with anti-β-galactosidase antibody. (A) Oh prepupal wing disc from the ctwHZ-2 transformant line, β-galactosidase is expressed throughout the wing margin including the hinge region. All of the wing margin mechanoreceptors and noninnervated bristles that normally express Cut are labeled with anti-β-kgalactosidase. (B) Oh prepupal wing disc from the transformant line ctwHZ-1. β-galactosidase is expressed in all of the cells of the presumptive wing margin but not in the hinge region. (C) 7h ctwHZ-2 prepupal wing. Anti-β-galactosidase labeling is evident in all of the wing margin mechanoreceptors and noninnervated bristles. (D) 7h ctwHZ-1 prepupal wing. At this stage β-galactosidase expression can be detected along the margin and continuing into the hinge region, but the staining in the hinge region is very light. Bar=50 μm.

In the HZ50 construct, the 2.7 kb fragment is much closer to the promoter than it is in cut, where it is approximately 80 kb from the promoter. The ability to function in a variety of positions and with a heterologous promoter are characteristic of enhancers. The 2.7kb fragment, therefore, contains a wing margin transcription enhancer that activates expression in the wing margin mechanoreceptors and noninnervated bristles.

The expression of lacZ in these cells in ctL-32 wings, activated by ctwHZ-2, shows that at 18 h, just before the shape of the wing changes, the undifferentiated precursors are still present (Fig. 10). Thus, the change in the shape of the wing, which begins around 20 to 24 h AP, must occur by the death of wing margin mechanoreceptors and noninnervated bristles soon after the time they normally differentiate, which is marked by the beginning of Ag 22C10 expression at 16 –20 h AP.

Fig. 10.

ctwHZ-2 β-galactosidase expression in an 18 h ctL-32 wing. At 18 h AP the shape of the ctL-32 wing is normal and the mechanoreceptors and noninnervated bristles are present. The cells are marked with β- galactosidase produced by the ctwHZ-2 construct, demonstrating that, at 18 h the cells that would normally form the mechanoreceptors and noninnervated bristles are present in the wing margin. Bar=50 μm.

Fig. 10.

ctwHZ-2 β-galactosidase expression in an 18 h ctL-32 wing. At 18 h AP the shape of the ctL-32 wing is normal and the mechanoreceptors and noninnervated bristles are present. The cells are marked with β- galactosidase produced by the ctwHZ-2 construct, demonstrating that, at 18 h the cells that would normally form the mechanoreceptors and noninnervated bristles are present in the wing margin. Bar=50 μm.

Function of the cut gene in the wing margin

The cut gene is a complex locus in Drosophila that has regulatory regions defined by mutations that cause defects in specific tissues. Mutations in the cut wing region of the gene cause loss of tissue in the wing margin, which is composed entirely of external sensory organs on the anterior part and tip and of uninnervated bristles on the posterior part (Hartenstein and Posakony, 1989). We show that cut expression is absent in the wing margin mechanoreceptors and noninnervated bristles of cut wing mutants. The absence of expression is due to the inactivation by the mutations of an enhancer 80kb upstream of the 5’ exon.

Protein products of cut are localized to the nuclei of the cells of all external sensory organs in Drosophila (Blochlinger et al. 1988) and are essential for proper differentiation of these cells. Mutations that reduce Cut protein levels or that alter its localization cause most external sensory organs to take on the characteristics of chordotonal organs, internal sense organs morphologically, antigenically and functionally distinguishable from external sensory organs (Blochlinger et al. 1990; Bodmer et al. 1987; Liu et al. 1991). In these cells, the cut gene can be thought of as functioning as a binary switch that is part of a epigenetic code specifying the external sensory organ program of gene expression. The code specifying chordotonal organ development must then be very similar to the code specifying external sensory organ development, with the important difference that cut is off in chordotonal organs.

In wing margin mechanoreceptors, however, loss of cut activity does not cause a clear transformation from one cell type to another. Rather, jn cut wing mutants, these mechanoreceptors fail to differentiate and, instead, die. This is demonstrated by the failure of sense-organ-specific markers to be expressed in the mutant wing margin cells, followed by cell death soon after the time when the markers are normally expressed. The cell death is probably caused by the failure to differentiate. Mutations of the gene zw3 in homozygous patches cause cells all over the wing that are normally epidermal to develop into sensillae (Simpson et al. 1988). In a ct6 background, homozygous zw3 patches on the wing margin are made up of typical margin mechanoreceptors and do not contain margin gaps, which are normally evident around the entire margin of ct6 wings (Ripoll et al. 1988). Apparently, zw3 mutations force the development of the margin mechanoreceptors even without the activity of the cut enhancer for the cells, and differentiation of the mechanoreceptors prevents their death. The different outcome of loss of cut activity in the wing margin external sensory organs could be due to a different combination of regulatory genes specifying wing margin mechanoreceptors than the combination specifying other external sensory organs in larvae and adults. An indication that the wing margin mechanoreceptors are specified by a different set of regulatory genes is that they have no requirement for achaete and scute gene products (Garcia-Bellido and Santa Maria, 1978; Leyns et al. 1989), which are necessary for development of all other external sensory organs.

Besides the sense organs of the wing, Cut protein is expressed in the noninnervated bristles of the posterior margin, which also die in cut wing mutants. In fact, similarities to the anterior margin mechanoreceptors lead us to believe that they are, in fact, modified mechanoreceptors that lack the neuron and thecogen cells. Similarities between the mechanoreceptors and noninnervated bristles include the following. (1) Their pattern of cut expression in wild type and cut mutant flies is identical. (2) The shafts of noninnervated bristles are similar to those of slender mechanoreceptors of the anterior margin; although they lack sockets, the shaftforming cells of the posterior bristles are surrounded by cells similar to socket-secreting cells (Hartenstein and Posakony, 1989). (3) The shaft-forming cells of the noninnervated bristles stain with the mAb 22C10 (Hartenstein and Posakony, 1989), which stains almost exclusively neural and sensory cells (Canal and Ferrus, 1986; Hartenstein, 1988; Tomlinson and Ready, 1987; Zipursky et al. 1984). (4) zw3 mutations, which in homozygous patches transform epidermal cells into external sensory organs, transform epidermal cells in the posterior part of the wing blade into noninnervated bristles (Simpson et al. 1988).

The similarity between the wing margin mechanoreceptors and the noninnervated bristles and the parallel regulation of cut in the two types of cells suggests that the signals that control the cell type of the two may be very similar. The loss of cut activity through mutation in either cell type causes cell death, possibly by creating a nonsense signal or a signal that codes for programmed cell death.

Interference with enhancer-promoter interaction by gypsy retrotransposon insertions and the Suppressor of Hairy-wing Protein

In addition to rearrangements that delete or invert the cut wing enhancer, insertions of gypsy, a 7.5 kb retrotransposon; also abolish cut expression in wing margin mechanoreceptors and noninnervated bristles. All 27 characterized gypsy retrotransposon insertions fall in the 80 kb interval between the wing margin enhancer and the 5’ exon (Jack, 1985). None fall outside that interval, even though the 70 kb transcription unit would present as large a target if all insertions in the transcribed region caused mutations. Apparently, gypsy insertions can prevent the interaction of the wing margin enhancer with the cut promoter only when located between the enhancer and promoter.

This hypothesis provides a possible explanation for the following observation. As the upstream gypsy insertion sites get closer to the cut promoter, expression is lost in more tissues (Liu et al. 1991, Fig. 8). If multiple cell-type-specific enhancers activate cut expression, and a gypsy insertion only interferes with an enhancer when positioned between that enhancer and the promoter, then more enhancers would be interfered with as the gypsy insertions approach the promoter. A minimum of three more enhancers activating tissue-specific cut expression would be predicted, based on the phenotypes of gypsy insertion mutations (Fig. 8): one at Okb for vibrissae, one between +20 and +40 for an undefined postembryonic function, and one or more between +40 and +71 for external sensory organs, central nervous system, Malpighian tubules and spiracles. Transcription begins at +78 (Fig. 8).

The effect of gypsy on enhancer function has been documented with smaller enhancer-promoter distances at the yellow locus. The y2 allele contains a gypsy insertion that interferes with the activity of two enhancers that it separates from the promoter, but does not interfere with two others (Geyer and Corees, 1987; Geyer et al. 1986; Harrison et al. 1989). Furthermore, the sites of gypsy insertions within the Ultrabithorax locus support the hypothesis that gypsy interferes with an enhancer only when it separates the enhancer from a promoter. The pattern of deletions and gypsy insertions in both the abx/bx and the bxd/pbx regulatory regions is strikingly similar to the pattern of mutations in the cut wing region of cut. Regions defined by deletion mutations, abx and pbx, are far from the Ubx promoter, about 50 kb downstream in the case of abx and between 26 and 35 kb upstream for pbx. The bx gypsy insertions all fall between the abx region and the promoter, while the bxd gypsy insertions are all either within the pbx region or between it and the promoter (Bender et al. 1983; Bender et al. 1985; Peifer and Bender, 1986). The model predicts that abx and pbx deletions remove enhancer elements that activate expression in the tissues affected by the mutations. Indeed, the abx region has been shown to contain such a transcription enhancer (Simon et al. 1990).

Gypsy insertion alleles of many genes are suppressed by loss-of-function mutations of the suppressor of Hairy-wing [su(Hw)] gene (Modolell et al. 1983; Rutledge et al. 1988). This gene encodes a zinc-finger protein (Parkhurst et al. 1988) that binds to a 12 bp DNA sequence repeated several times near the 5’ end of the gypsy transposon (Dorsett, 1990; Spana et al. 1988). Experiments with the hsp70 promoter indicate that the su(Hw) protein-binding sites are the only portion of gypsy required for interference with upstream transcription control elements. Insertion of su(Hw) protein-binding sites between the two upstream heat-shock regulatory elements and the promoter vastly reduces heat-shock transcription (Holdridge and Dorsett, 1991), while insertion just upstream of the regulatory elements has no detectable effect on heatshock transcription. These observations suggest that the su(Hw) protein mediates interference with enhancerpromoter interactions.

We thank Cheng Liu for assistance with germline transformations and Dennis Ballinger for critical reading and thoughtful comments on the manuscript. This work was supported by National Science Foundation grant DMB 8811519 to J.J. and American Cancer Society grant NP-715 to D.D.

Bender
,
W.
,
Akam
,
M.
,
Karch
,
F.
,
Beachy
,
P. A.
,
Peifer
,
M.
,
Spierer
,
P.
,
Lewis
,
E. B.
and
Hogness
,
D. S.
(
1983
).
Molecular genetics of the bithorax complex in Drosophila melanogaster
.
Science
221
,
23
29
.
Bender
,
W.
,
Weiffenbach
,
B.
,
Karch
,
F.
and
Peifer
,
M.
(
1985
).
Domains of cis-interaction in the bithorax complex
.
Cold Spring Harbor Symp. quant. Biol
.
50
,
173
180
.
Blochunger
,
K.
,
Bodmer
,
R.
,
Jack
,
J.
,
Jan
,
L. Y.
and
Jan
,
Y. N.
(
1988
).
Primary structure and expression of a product from cut, a locus involved in specifying sensory organ identity in Drosophila
.
Nature
333
,
629
635
.
Blochunger
,
K.
,
Bodmer
,
R.
,
Jan
,
L. Y.
and
Jan
,
Y. N.
(
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
,
1322
1331
.
Bodmer
,
R.
,
Barbel
,
S.
,
Sheperd
,
S.
,
Jack
,
J. W.
,
Jan
,
L. Y.
and
Jan
,
Y. N.
(
1987
).
Transformation of sensory organs by mutations of the cut locus of Drosophila melanogaster
.
Cell
51
,
293
307
.
Bryant
,
P. J.
(
1975
).
Pattern formation in the imaginai wing disc of Drosophila melanogaster
.
J. exp. Zool
.
193
,
49
78
.
Campuzano
,
S.
,
Carramouno
,
L.
,
Cabrera
,
C. V.
,
Ruiz-Gomez
,
M.
,
Villares
,
R.
,
Boronat
,
A.
and
Modollel
,
J.
(
1985
).
Molecular genetics of the achaete-scute gene complex of D. melanogaster
.
Cell
40
,
327
338
.
Canal
,
I.
and
Ferrus
,
A.
(
1986
).
The pattern of early neuronal differentiation in Drosophila
.
J. Neurogen
.
3
,
293
319
.
Dorsett
,
D.
(
1990
).
Potentiation of a polyadenylylation site by a downstream protein-DNA interaction
.
Proc. natn. Acad. Sci. U.S.A
87
,
4373
4377
.
Dorsett
,
D.
,
Viguanti
,
G. A.
,
Rutledge
,
B. J.
and
Meselson
,
M.
(
1989
).
Alteration of hsp82 gene expression by the gypsy transposon and suppressor genes in Drosophila melanogaster
.
Genes Dev
.
3
,
454
468
.
Garcia-Bellido
,
A.
and
Santa Maria
,
P.
(
1978
).
Developmental analysis of the achaete-scute system of Drosophila melanogaster
.
Genetics
88
,
469
486
.
Geyer
,
P. K.
and
Corces
,
V. G.
(
1987
).
Separate regulatory elements are responsible for the complex pattern of tissuespecific and developmental transcription of the yellow locus on Drosophila melanogaster
.
Genes Dev
.
1
,
996
1004
.
Geyer
,
P. K.
,
Spana
,
C.
and
Corces
,
V. G.
(
1986
).
On the molecular mechanism of gypsy-induced mutations at the yellow locus of Drosophila melanogaster
.
EMBO J
.
5
,
2657
2662
.
Ghysen
,
A
and
Dambly-Chaudiere
,
C.
(
1988
).
From DNA to form: The achaete-scute complex
.
Genes Dev
.
2
,
495
501
.
Harrison
,
D. A.
,
Geyer
,
P. K.
,
Spana
,
C.
and
Corces
,
V. G.
(
1989
).
The gypsy retrotransposon of Drosophila melanogaster: Mechanisms of mutagenesis and interaction with the suppressor of Hairy-wing locus
.
Dev. Genetics
.
10
,
239
248
.
Hartenstein
,
V.
(
1988
).
Development of Drosophila larval sensory organs: spatiotemporal pattern of sensory neurones, peripheral axonal pathways and sensilla differentiation
.
Development
102
,
869
886
.
Hartenstein
,
V.
and
Posakony
,
J. W.
(
1989
).
Development of adult sensilla on the wing and notum of Drosophila melanogaster
.
Development
107
,
389
405
.
Hiromi
,
Y.
and
Gehring
,
W. J.
(
1987
).
Regulation and function of the Drosophila segmentation gene fushi tarazu
.
Cell
50
,
963
974
.
Holdridge
,
C.
and
Dorsett
,
D.
(
1991
).
Repression of hsp70 heat shock gene transcription by the suppressor of Hairy-wing protein of Drosophila melanogaster
.
Molec. cell. Biol
.
11
,
1894
1900
.
Jack
,
J. W.
(
1985
).
Molecular organization of the cut locus of Drosophila melanogaster
.
Cell
42
,
869
876
.
Johnson
,
T. K.
and
Judd
,
B. H.
(
1979
).
Analysis of the cut locus of Drosophila melanogaster
.
Genetics
92
,
485
502
.
Leyns
,
L.
,
Dambly-Chaudiere
,
C.
and
Ghysen
,
A.
(
1989
).
Two different sets of cis elements regulate scute to establish two different sensory patterns
.
Roux’s Arch, devl Biol
.
198
,
227
232
.
Lindsley
,
D. L.
and
Grell
,
E. H.
(
1968
).
Genetic Variations of Drosophila melanogaster. Carnegie Inst. Wash
.
Liu
,
S.
,
Mcleod
,
E.
and
Jack
,
J.
(
1991
).
Four distinct regulatory regions of the cut locus and their effect on cell type specification in Drosophila
.
Genetics
127
,
151
159
.
Modolell
,
J.
,
Bender
,
W.
and
Meselson
,
M.
(
1983
).
Drosophila melanogaster mutations suppressible by suppressor of Hairy-wing are insertions of a 7.3-kilobase mobile element
.
Proc. natn. Acad. Sci. U.S.A
.
80
,
1678
1682
.
Murray
,
M. A.
,
Schubiger
,
M.
and
Palka
,
J.
(
1984
).
Neuron differentiation and axon growth in the developing wing of Drosophila melanogaster
.
Devi Biol
.
104
,
259
273
.
Palka
,
J.
,
Lawrence
,
P. A.
and
Hart
,
H. S.
(
1979
).
Neural projection patterns from homoeotic tissue of Drosophila studied in bithorax mutants and mosaics
.
Devi Biol
.
69
,
549
575
.
Parkhurst
,
S. M.
,
Harrison
,
D. A.
,
Remington
,
M. P.
,
Spana
,
C.
,
Kelley
,
R. L.
,
Coyne
,
R. S.
and
Corces
,
V. G.
(
1988
).
The Drosophila su(Hw) gene, which controls the phenotypic effect of the gypsy transposable element, encodes a putative DNA binding protein
.
Genes Dev
.
2
,
1205
1215
.
Peifer
,
M.
and
Bender
,
W.
(
1986
).
The anterobithorax and bithorax mutations of the bithorax complex
.
EMBO J
.
5
,
2293
2303
.
Ripoll
,
P.
,
El Messal
,
M.
,
Laran
,
E.
and
Simpson
,
P.
(
1988
).
A gradient of affinities for sensory bristles across the wing blade of Drosophila melanogaster
.
Development
103
,
757
767
.
Rutledge
,
B. J.
,
Mortin
,
M. A.
,
Schwartz
,
E.
,
Thierry-Mieg
,
D.
and
Meselson
,
M.
(
1988
).
Genetic interactions of modifier genes and modifiable alleles in Drosophila melanogaster
.
Genetics
119
,
391
397
.
Santamaria
,
P.
and
Garcia-Bellido
,
A.
(
1975
).
Developmental biology of two wing scalloping mutants ct6 and BxJ of Drosophila melanogaster
.
Wilhelm Roux’s Arch, devl Biol
.
178
,
233
245
.
Simon
,
J.
,
Peifer
,
M.
,
Bender
,
W.
and
O’Connor
,
M.
(
1990
).
Regulatory elements of the bithorax complex that control expression along the anterior-posterior axis
.
EMBO J
.
9
,
3945
3956
.
Simpson
,
P.
,
El Messal
,
M.
,
Moscoso Del Prado
,
J.
and
Ripoll
,
P.
(
1988
).
Stripes of positional homologies across the wing blade of Drosophila melanogaster
.
Development
103
,
391
401
.
Spana
,
C.
,
Harrison
,
D. A.
and
Corces
,
V. G.
(
1988
).
The Drosophila melanogaster suppressor of Hairy-wing protein binds to specific sequences of the gypsy retrotransposon
.
Genes Dev
.
2
,
1414
1423
.
Tomlinson
,
A.
and
Ready
,
D. F.
(
1987
).
Neuronal differentiation in the Drosophila ommatidium
.
Devl Biol
.
120
,
366
376
.
Villares
,
R.
and
Cabrera
,
C. V.
(
1987
).
The achaete-scute complex of D. melanogaster: Conserved domains in a subset of genes required for neurogenesis and their homology to myc
.
Cell
50
,
415
424
.
Zipursky
,
S. L.
,
Venkatesh
,
T. R.
,
Teplow
,
D. B.
and
Benzer
,
S.
(
1984
).
Neuronal development in the Drosophila retina: Monoclonal antibodies as molecular probes
.
Cell
36
,
15
26
.