The ventral epidermis of Drosophila melanogaster is derived from longitudinal rows of ectodermal precursor cells that divide and expand to form the ventral embryonic surface. The spitz class genes are required for the proper formation of the larval ventral cuticle. Using a group of enhancer trap lines that stain subsets of epidermal cells, it is shown here that spitz class gene function is necessary for ventral epidermal development and gene expression. Analysis of single-minded mutant embryos implies that ventral epidermal cell fate is influenced by the CNS midline cells.

How tissue and cell identity are established is a fundamental problem of animal development. The molecular and cellular basis of how cellular identity is established has been productively studied in the Drosophila embryo (Lawrence, 1992). One active area of research has investigated how embryonic cells attain their identities along both the anterior/posterior (A/P) and dorsal/ventral (D/V) axes, and at the termini. Genetic and molecular studies have shown that identity along the A/P axis requires a genetic hierarchy that includes maternal, gap, pair-rule, segment polarity and homeotic genes. A number of genes have also been identified that play a role in establishing cell identity along the D/V axis. These maternal and zygotic genes subdivide the embryo into domains including the ectoderm, which gives rise to both epidermis and nervous system. Currently, there is little known about how the epidermis is subdivided into cells with distinct identities along the D/V axis. The subject of this paper concerns how the ventral epidermal cells acquire their fate.

The epidermis can be subdivided into three regions along the D/V axis (Fig. 1; Ferguson and Anderson, 1991). The ventral epidermis (VE) lies at the ventral midline of the embryo and secretes the centralmost section of the ventral denticles. The lateral epidermis (LE) gives rise to the rest of the ventral denticles. The ectodermal precursors of the LE, VE and mesectodermal (MEC) cells are referred to as the ‘neurogenic ectoderm’ since they give rise to the central nervous system (CNS) as well as the epidermis. The MEC cells should be distinguished from the other neurogenic ectodermal cells because they give rise only to CNS midline cells and not epidermis (Thomas et al., 1988; Nambu et al., 1990). The dorsal epidermis (DE) gives rise to the hairs that form on the dorsal cuticle. The VE can be distinguished from the LE and DE in a nearly mature stage 16 embryo due to the large, stretched appearance of its cells and nuclei (Fig. 2A). The genetic work of Nüsslein-Volhard, Wieschaus and co workers identified a large number of genes involved in formation of the larval cuticle and corresponding epidermis (Nüsslein-Volhard et al., 1984; Jürgens et al., 1984; Wieschaus et al., 1984). Included in this collection were a group of mutants missing the centralmost ventral denticles and presumed to have VE defects (Mayer and Nüsslein Volhard, 1988). These genes included pointed (pnt), sichel (sic), single-minded (sim), spitz (spi), Star (S) and veinlet (ve; previously called rhomboid), and were referred to as the ‘spitz class’ of genes. The ocelliless (oc; previously called ortho denticle) gene can also be included in this group based on its ventral midline cuticular defect (Wieschaus et al., 1984, 1992), as can the Epidermal growth factor receptor (Egfr, previously called DER), gene (Clifford and Schüpbach, 1992). More recently, cellular and genetic analysis of salivary gland placode formation has confirmed that the spitz class mutants are defective in formation of the ventralmost ectoderm (Panzer et al., 1992). In this paper, we describe specific cytological markers for VE cells that allow cellular analysis of mutations that influence VE cell fate. It is shown that spitz class gene function is required for gene expression and formation of VE precursors. Based on genetic studies and the expression pattern of the sim gene, it is proposed that cell fate in the VE is influenced by the CNS midline cells.

Fig. 1.

Embryonic and larval sub-divisions. (A) This schematic cross-section represents the cellular blastoderm fate map. Dorsal is at the top of the drawing. AS, amnioserosa; DE, dorsal epidermis; LE, lateral epidermis; VE, ventral epidermis; MEC, mesectoderm; MES, mesoderm. The neurogenic ectoderm (N-ECT) gives rise to both the ventral-lateral epidermis and CNS. It includes the blastoderm precursors that give rise to the LE, VE and MEC. The MEC gives rise to the midline cells of the CNS. (B) This schematic shows one halfsegment of a mature larval cuticle splayed out illustrating the correspondence between epidermal sub-divisions and cuticular structures. The ventral midline of the larva is on the left side of the figure and the dorsal midline is to the right.

Fig. 1.

Embryonic and larval sub-divisions. (A) This schematic cross-section represents the cellular blastoderm fate map. Dorsal is at the top of the drawing. AS, amnioserosa; DE, dorsal epidermis; LE, lateral epidermis; VE, ventral epidermis; MEC, mesectoderm; MES, mesoderm. The neurogenic ectoderm (N-ECT) gives rise to both the ventral-lateral epidermis and CNS. It includes the blastoderm precursors that give rise to the LE, VE and MEC. The MEC gives rise to the midline cells of the CNS. (B) This schematic shows one halfsegment of a mature larval cuticle splayed out illustrating the correspondence between epidermal sub-divisions and cuticular structures. The ventral midline of the larva is on the left side of the figure and the dorsal midline is to the right.

Fig. 2.

Development of VE cells in normal development. All embryos are wild-type and stained with a monoclonal antibody against β galactosidase as visualized with HRP histochemistry. Embryonic stages use the convention of Campos-Ortega and Hartenstein (1985). All embryos show anterior to the left except (F) which has anterior pointing to the top of the page. (A) Stage 15 embryo containing the BL97 enhancer trap insertion, which stains all epidermal cell nuclei. The embryo is tilted so that the lateral and ventral surfaces are shown. The bracket shows the limits of the VE, which is easily visualized at this stage of development, and position of the star indicates the ventral midline. Notice the stretched appearance of the VE cell nuclei compared to the LE and DE cell nuclei. (B) Ventral view of a stage 10 BP28 embryo showing initial expression of β-galactosidase along the ventral midline. There are additional stained cells beneath the focal plane shown in this panel that are probably neuroblasts. This photograph covers about three parasegments. (C) Ventral view of a stage 11 BP28 embryo showing β-galactosidase expression along the ventral midline. The cells that extend laterally (white arrow) may correspond to those derived from the δ1425 mitotic domain and the cells arranged in longitudinal rows (black arrow) are likely derived from the δ14M mitotic domain (Foe, 1989). This photograph covers about three parasegments. (D) Ventral view of a stage 13 BP28 embryo as the VE cells begin their lateral spread. Four segments are shown. (E) Ventral view of a stage 16 BP28 embryo after dorsal closure showing the extended VE cell nuclei. These cells correspond to the bracketed cells in panel A. Four segments are shown. (F) Ventral view of a stage 17 BP28 embryo showing the relationship between the VE cells and the ventral denticles of the cuticle. There are about 25 BP28-positive cells that give rise to the centralmost denticles. Segments A6 and A7 are shown. Scale bar, (A) 50m, (B-F) 16 μm.

Fig. 2.

Development of VE cells in normal development. All embryos are wild-type and stained with a monoclonal antibody against β galactosidase as visualized with HRP histochemistry. Embryonic stages use the convention of Campos-Ortega and Hartenstein (1985). All embryos show anterior to the left except (F) which has anterior pointing to the top of the page. (A) Stage 15 embryo containing the BL97 enhancer trap insertion, which stains all epidermal cell nuclei. The embryo is tilted so that the lateral and ventral surfaces are shown. The bracket shows the limits of the VE, which is easily visualized at this stage of development, and position of the star indicates the ventral midline. Notice the stretched appearance of the VE cell nuclei compared to the LE and DE cell nuclei. (B) Ventral view of a stage 10 BP28 embryo showing initial expression of β-galactosidase along the ventral midline. There are additional stained cells beneath the focal plane shown in this panel that are probably neuroblasts. This photograph covers about three parasegments. (C) Ventral view of a stage 11 BP28 embryo showing β-galactosidase expression along the ventral midline. The cells that extend laterally (white arrow) may correspond to those derived from the δ1425 mitotic domain and the cells arranged in longitudinal rows (black arrow) are likely derived from the δ14M mitotic domain (Foe, 1989). This photograph covers about three parasegments. (D) Ventral view of a stage 13 BP28 embryo as the VE cells begin their lateral spread. Four segments are shown. (E) Ventral view of a stage 16 BP28 embryo after dorsal closure showing the extended VE cell nuclei. These cells correspond to the bracketed cells in panel A. Four segments are shown. (F) Ventral view of a stage 17 BP28 embryo showing the relationship between the VE cells and the ventral denticles of the cuticle. There are about 25 BP28-positive cells that give rise to the centralmost denticles. Segments A6 and A7 are shown. Scale bar, (A) 50m, (B-F) 16 μm.

Drosophila strains

The Drosophila strains used in this work were generally the most severe alleles available unless noted otherwise and are identified in Lindsley and Zimm (1992). The sim strains used were sim2 (EMS), sim5 (EMS) and Df(3R)ry619 (X-ray). They are protein nulls and exhibit amorphic CNS phenotypes (Nambu et al., 1991). Alleles of the other genes are as follows: Egfrf6 (severe), Egfrf7 (weak), oc2, pnt1, ve4,spi1 and SIIN. The oc allele was provided by R. Finkelstein and N. Perrimon, and the other mutant strains were obtained from the Bowling Green Stock Center.

Epidermal cell markers

The enhancer trap lines AA69, BL97, BP28 and BP48 were identified in our laboratory during a large-scale screen (Crews et al., 1992). All lines localize β-galactosidase to cell nuclei.

AA69

The P-element transposon inserted adjacent to the ve gene at 62A on the third chromosome (Nambu et al., 1990). Expression of β galactosidase accurately reflects the transcription of the ve gene (Bier et al., 1990). After germband retraction, AA69 is expressed in one to two rows of epidermal cells that lie in the posterior of the segment (Table 1). β-galactosidase is also expressed in the MEC cells from stages 8-17 and the ve gene is expressed in the ventral ectoderm before gastrulation. This likely includes the precursors to both the VE and LE.

Table 1.

Epidermal cell markers

Epidermal cell markers
Epidermal cell markers

BL97

This insertion lies at 32E on the second chromosome and accurately reflects expression of an adjacent gene (S. Hu and S. T. Crews, unpublished data). During stage 10, all epidermal cells express β-galactosidase and continue to do so through stage 17. There is BL97 β-galactosidase expression in all of the MEC cells from stage 9 to stage 17. BL97 is also expressed in the midgut and stomagastric nervous system.

BP28

The BP28 insertion lies at 24DE and accurately reflects expression of an adjacent gene (S. H. Kim and S. T. Crews, unpublished data). It is expressed prominently in the VE beginning at stage 10. The CNS midline cells also show prominent β-galactosidase expression beginning at stage 12. Additional expression is observed in the proventriculus, a small set of lateral CNS cells, PNS, stomagastric nervous system and antennal-maxillary complex.

BP48

The BP48 insertion is on the second chromosome and reflects the expression of an adjacent gene (S. H. Kim and S. T. Crews, unpublished data). This gene is expressed in a subset of VE cells that form the ventral denticles and in a corresponding subset of LE cells (Table 1). BP48 is expressed in a subset of CNS midline cells including the midline glia late in embryogenesis and in numerous other cell types including the hindgut and midgut.

oc

Expression of oc was analyzed using a full-length cDNA probe generously provided by R. Finkelstein and N. Perrimon. oc is transcribed in the VE cells beginning at stage 9 and continues to be expressed in these cells through stage 15. It is also expressed in the CNS midline precursors beginning at stage 8 and is restricted to a subset of CNS midline cells after stage 12 (Finkelstein et al., 1990).

Immunohistochemistry and in situ hybridization

Antibody staining and in situ hybridization of embryonic whole mounts and sections were performed and observed as previously described (Nambu et al., 1991). Balancer chromosomes included either a P[ftz/lacZ] or P[elav/lacZ] marker that allowed identification of homozygous mutant embryos by staining with an antibody against β-galactosidase (Nambu et al., 1990). The illustration in Fig. 5E is a computer-assisted representation of stained cell nuclei. Initially, nuclei were traced from a video scan of the embryo. The traced image was entered into a Macintosh computer by digital scanning and the scanned image was used as a template for the stylized Adobe Illustrator representation shown in the figure. Nuclear counts are expressed as the mean ± standard deviation (s.d.).

The ventral epidermis

Study of VE formation in wild-type and mutant embryos has been made possible by the identification and use of several enhancer trap lines (O’Kane and Gehring, 1987) that express β-galactosidase within the epidermis. These lines and the epidermal cells they stain are listed in Table 1 The definitive line is BP28, which prominently stains the VE cell nuclei but not the adjacent epidermis; it also is expressed in the MEC. In this paper, the VE comprises those cells that show BP28 lacZ expression. BP28 is a useful lineage marker for the development of these cells from germband extension (embryonic stage 10) to formation of first instar larva.

BP28 staining during embryonic development is shown in Fig. 2B-F. Initial expression begins during the germbandextended stage 10 of embryogenesis in adjacent columns of cells that lie at the ventral midline of the presumptive epidermis (Fig. 2B). Many of the cells remain on the surface and will form the VE, others delaminate internally from the ventral ectoderm and resemble neuroblasts. This observation is similar to those made previously by Foe (1989) who observed cell division and internal movement of ventral ectodermal cells. Stage 11 embryos also show narrow rows of cell nuclei along the ventral midline with segmentally repeated groups of cells extending laterally (Fig. 2C, note arrows). During stages 12-13 the VE cells spread out as dorsal closure begins (Fig. 2D).

Dorsal closure is complete by stage 16 and the VE cells are well spread out (Fig. 2E). Nuclei of the VE cells are larger and have an elongated morphology in comparison to the smaller, more rounded morphology of more lateral epidermal cells (Fig. 2A). There are approximately 50 nuclei/segment at stage 16 that stain with BP28. Fig. 2F shows a stage 17 embryo that has secreted cuticle and shows the relationship between the VE cells and the ventral denticles. At this stage, the nuclei are more rounded and closer together than at stage 16. There are approximately 25 BP28-staining nuclei that lie beneath the central 30% of the denticles.

spitz class gene function is required for early ventral ectodermal gene expression

Previous experiments by Mayer and Nüsslein-Volhard (1988) had shown that mutations in the spitz class genes had ventral cuticle defects consistent with alterations in VE cell fate. These genes include pnt, sic, sim, spi, S and ve. We also include the oc and Egfr genes. These mutants all appear to be missing a number of ventral denticles, although other cuticular pattern defects are often present.

We have directly examined the fate of the VE cells in spitz class mutants (n.b. the maternal-effect sic gene was not analyzed) by using genes and markers expressed specifically in these cells.

We first examined the requirement for spitz class gene function on the expression of BP28 and the oc gene. The BP28 enhancer trap gene was crossed into mutant strains and embryos were stained with a monoclonal antibody against β-galactosidase. Fig. 3 shows the results of staining embryos at stages 12-13; similar results are observed throughout the time period of BP28 expression (stages 10-17). There is strong VE β-galactosidase expression in wild type (Fig. 3A). However, in Egfr, oc, pnt, sim, spi, S and ve, there is a complete or partial absence of BP28 VE expression depending on the mutant strain (Fig. 3B-G). There were less than 10% of the number of wild-type BP28-positive VE cells in Egfr, sim, spi, S and ve mutant embryos. The number of BP28-positive cells in both oc and pnt was reduced (10-50% of wild type), but this effect was less dramatic than in the other spitz class mutants. These results indicate that spitz class gene function is required for BP28 VE expression.

Fig. 3.

spitz class gene function is required for BP28 VE expression. Embryos all contained the BP28 P[lacZ] gene and were stained for β-galactosidase activity. All panels show ventral views of stage 12-13 embryos; results are similar for embryos at other stages of development. Generally, there is a complete absence or loss of >90% of BP28-positive VE cells in all spitz class mutants except oc and pnt, which are often less severe. This may reflect a lesser role of oc in VE formation since the oc2 allele is thought to be null (Finklestein et al., 1990). The few stained cells in the sim mutant embryo are due to non-midline BP28 lacZ-expressing cells. (A)wild-type (+), (B) Egfr, (C) oc, (D) pnt, (E) sim, (F) spi, (G) S, (H) ve. Scale bar, 25 μm.

Fig. 3.

spitz class gene function is required for BP28 VE expression. Embryos all contained the BP28 P[lacZ] gene and were stained for β-galactosidase activity. All panels show ventral views of stage 12-13 embryos; results are similar for embryos at other stages of development. Generally, there is a complete absence or loss of >90% of BP28-positive VE cells in all spitz class mutants except oc and pnt, which are often less severe. This may reflect a lesser role of oc in VE formation since the oc2 allele is thought to be null (Finklestein et al., 1990). The few stained cells in the sim mutant embryo are due to non-midline BP28 lacZ-expressing cells. (A)wild-type (+), (B) Egfr, (C) oc, (D) pnt, (E) sim, (F) spi, (G) S, (H) ve. Scale bar, 25 μm.

The oc gene is expressed in most or all of the VE cells beginning around stage 9 of embryogenesis (Fig. 4A) and continuing to stage 15 or later. Its VE expression is similar to BP28. Figure 4C shows that oc expression is greatly reduced along the ventral midline in a sim mutant embryo. Similar results were obtained for Egfr (Fig. 4B), pnt, spi, S and ve (data not shown), although small numbers of staining cells were sometimes observed. These results are similar to those observed for BP28 expression in spitz class mutant strains. No oc expression was observed in spitz class mutants even in stage 9 embryos when the gene is first expressed in the presumptive VE.

Fig. 4.

VE expression of oc requires spitz class gene function. (A)Ventral view of a stage 10 wild-type (+) embryo hybridized in situ to an oc cDNA probe showing prominent VE oc expression. (B)Ventral view of a stage 10 Egfr embryo hybridized to an oc probe showing absence of expression. Similar results were also observed in pnt, spi, S and ve mutants. (C) Ventral view of a stage 10 sim embryo hybridized to an oc probe showing absence of expression. Scale bar: 25 μm.

Fig. 4.

VE expression of oc requires spitz class gene function. (A)Ventral view of a stage 10 wild-type (+) embryo hybridized in situ to an oc cDNA probe showing prominent VE oc expression. (B)Ventral view of a stage 10 Egfr embryo hybridized to an oc probe showing absence of expression. Similar results were also observed in pnt, spi, S and ve mutants. (C) Ventral view of a stage 10 sim embryo hybridized to an oc probe showing absence of expression. Scale bar: 25 μm.

Ventral epidermal cells are absent in older spitz class mutant embryos

The results with BP28 and oc indicate that spitz class mutants can influence gene expression within the presump tive VE. To investigate further the role of the spitz class genes on VE development, we examined the expression of two other genes that are normally expressed within these cells in mutant strains.

The ve gene is expressed in one to two circumferential single-cell rows in the posterior of each segment at the time of germband retraction and later (Fig. 5A,E; Bier et al., 1990). ve expression (as detected using an enhancer trap insertion at the locus, AA69) includes the VE cells in the posterior region of the segment as well as the LE and DE. It is an excellent marker to determine whether the VE cells are absent in spitz class mutant embryos or present but lacking in VE-specific gene expression. There are 16±1 (±s. d.) (n=5) VE cells and 43±2 (n=5) LE and DE cells that express AA69 in A2 segments of wild-type stage 15-16 embryos based on position and morphology. The expression of AA69 is clearly affected in spitz class mutant embryos. All or nearly all of the VE-like cells are absent in sim mutant embryos (Fig. 5C,E). This is indicated by nuclear counts that indicate fewer nuclei in the mutants. There are only 44±3 (n=6) stained cells in sim mutant embryos as opposed to the 59±1 in wild-type, consistent with the loss of the 16 VE cells. Figure 5E provides an interpretative drawing showing that the cells along the ventral midline in sim mutant embryos are most likely derived from the LE and DE that are now joined together.
Fig. 5.

VE gene expression and cell formation requires spitz class gene function. All panels show ventral views of stage 15-16 embryos. Schematic interpretation of A-D is illustrated in panel E. (A) Wild-type embryo (+) showing VE AA69 (ve) expression in one to two rows of epidermal cells. Five segments are shown. (B) Wild-type (+) embryo showing BP48 expression in rows of VE. Five segments are shown. (C)AA69 expression in a sim homozygous mutant embryo (Df(3R)ry619) showing absence of VE cells and joining together of LE and DE cells at the ventral midline. The pnt, spi and S mutant embryos also showed a reduction of VE-like cells although spi and S usually had one to four VE-like cells still remaining and pnt 5-6. (D) BP48 expression in a sim homozygous mutant embryo showing loss of VE cells and joining together of LE at ventral midline. Similar results were obtained for pnt, spi and S mutant embryos although a few VE cells remained. (E) Schematic interpretation of panels A-D. VE cells are shaded; LE and DE are not. Each section of cells is derived from a computer-assisted representation of a stage 16 A6 segment. In the middle of the drawing are the VE cells represented by BP28 β-galactosidase staining. To the left is shown AA69-stained sim mutant and wild-type segments showing the loss of VE and ventral displacement of LE and DE. Overall, the wild-type (+) embryos had 16±1 stained VE cell nuclei and 43±2 LE and DE nuclei. In the sim mutant embryos there were a total of 44±3 total cell nuclei stained, consistent with the loss of VE. To the right, BP48 wild-type (+) and sim mutant embryonic segments are shown. The wild-type segment has 23±2 VE cell nuclei and 49±3 LE nuclei stained. The sim mutant segment has 43±3 cell nuclei stained consistent with loss of VE. Scale bar, 25 μm.

Fig. 5.

VE gene expression and cell formation requires spitz class gene function. All panels show ventral views of stage 15-16 embryos. Schematic interpretation of A-D is illustrated in panel E. (A) Wild-type embryo (+) showing VE AA69 (ve) expression in one to two rows of epidermal cells. Five segments are shown. (B) Wild-type (+) embryo showing BP48 expression in rows of VE. Five segments are shown. (C)AA69 expression in a sim homozygous mutant embryo (Df(3R)ry619) showing absence of VE cells and joining together of LE and DE cells at the ventral midline. The pnt, spi and S mutant embryos also showed a reduction of VE-like cells although spi and S usually had one to four VE-like cells still remaining and pnt 5-6. (D) BP48 expression in a sim homozygous mutant embryo showing loss of VE cells and joining together of LE at ventral midline. Similar results were obtained for pnt, spi and S mutant embryos although a few VE cells remained. (E) Schematic interpretation of panels A-D. VE cells are shaded; LE and DE are not. Each section of cells is derived from a computer-assisted representation of a stage 16 A6 segment. In the middle of the drawing are the VE cells represented by BP28 β-galactosidase staining. To the left is shown AA69-stained sim mutant and wild-type segments showing the loss of VE and ventral displacement of LE and DE. Overall, the wild-type (+) embryos had 16±1 stained VE cell nuclei and 43±2 LE and DE nuclei. In the sim mutant embryos there were a total of 44±3 total cell nuclei stained, consistent with the loss of VE. To the right, BP48 wild-type (+) and sim mutant embryonic segments are shown. The wild-type segment has 23±2 VE cell nuclei and 49±3 LE nuclei stained. The sim mutant segment has 43±3 cell nuclei stained consistent with loss of VE. Scale bar, 25 μm.

Similar results were observed in homozygous mutants of Egfr, spi and S. The weak Egfr allele, Egfr7, was examined since epidermis of the severe allele, Egfr6, was too defective for meaningful interpretation. Mutants of pnt were less extreme than the other spitz class mutants with about half of the wild-type number of VE-like cells present.

Another marker used is an enhancer trap gene (BP48) that corresponds to the subset of VE cells that underlie the denticles (Fig. 5B,E). Expression also extends to a subset of LE cells. Similar to AA69, the BP48 marker provides a convenient way to determine if VE cells are either present and not expressing VE genes in spitz class mutants or if the cells are absent. The experiments examining BP48 staining in spitz class mutant embryos were consistent with those of AA69 (Fig. 5D,E). Homozygous sim mutant embryos showed an absence of BP48-staining cells along the ventral midline from stage 12 until the end of embryogenesis (stage 15 is shown in Fig. 5D,E). Wild-type embryos have 23±2 (n=6) VE cell nuclei and 49±3 (n=6) LE cell nuclei that express β-galactosidase in each segment, whereas the sim mutant had 43±3 (n=4) cell nuclei stained, consistent with an absence of VE cells. Embryos mutant for Egfr7, pnt, spi and S also showed a reduced number of VE cells.

These experiments show that normal VE development does not occur in spitz class mutants as assayed by loss of VE cells and corresponding gene expression. Reduction in the number of VE cells in mutant embryos explains the spitz class cuticle defects that are missing the centralmost ventral denticles. There are several explanations for the loss of VE cells. The VE precursors may fail to carry-out their characteristic cell divisions (Foe, 1989). Alternatively, the precursors divide and then either die or take a non-VE cell identity and fail to behave as ventral epithelium. Since the different genes may not influence VE cell formation in a uniform fashion, the requirement of each spitz class mutant on cell division will have to be directly examined similar to experiments with Egfr mutants, which showed no effect on cycle 14 cell divisions (Clifford and Schüpbach, 1992).

Model for commitment of ventral epidermal cell fate

This paper shows that the spitz class genes influence VE gene expression and development. This expands previous observations that these genes were important for the formation of the medial denticles of the cuticle (Mayer and Nüsslein-Volhard, 1988). Furthermore, it was observed that these genes function early in embryogenesis around stage 9 as indicated by the loss of oc expression in the mutant embryos. Of particular interest is the observation that function of the sim gene is required for VE formation, since sim is not expressed in the VE precursor cells, but its expression is restricted to the adjacent MEC (Thomas et al., 1988; Nambu et al., 1990). Although it was previously suggested that the MEC cells are precursors to both the CNS midline cells and VE (Mayer and Nüsslein-Volhard, 1988), this is not the case; the MEC gives rise only to the CNS midline cells (Nambu et al., 1991). It is concluded that VE cell fate is dependent upon CNS midline cell function. How the CNS midline cells influence the VE is unknown. They may send an inductive signal that establishes VE identity, or be required for the VE precursors to respond to D/V patterning gene activity or a signal emanating from elsewhere. The following model provides a framework for understanding the commitment of VE cell fate (Fig. 6). In the late blastoderm embryo, D/V patterning genes commit a group of cells on either side of the mesoderm to become lateral neurogenic ectoderm (Fig. 6A). This is reflected in the blastoderm expression of the pro-neural achaete-scute complex genes and ve gene in cells of the neurogenic ectoderm (Romani et al., 1987; Kosman et al., 1991; Rao et al., 1991). At approximately the same time, D/V patterning genes act to uniquely commit two strips of cells that lie on either side of the presumptive mesoderm to enter the MEC lineage (Fig. 6B). This coincides with the initial transcription of the sim gene in these cells. As gastrulation takes place, the MEC cells migrate to the ventral midline of the ectoderm (Fig. 6C), sim protein is expressed and the development of the CNS midline lineage begins. The MEC nuclei migrate inwards during germband extension, although they maintain a cytoplasmic connection close to the underlying ectodermal cells (Nambu et al., 1991). During the time period after gastrulation, the MEC cells influence the development of the adjacent VE precursors as reflected in VE-specific gene expression and formation (Fig. 6D). The VE precursor cells then go on to differentiate resulting in the mature VE (Fig. 6E). It is also possible that cell fate of some medial neuroblasts is influenced by the CNS midline and spitz class genes, since they emerge from the same precursor cells as the VE.

Fig. 6.

Model of VE cell commitment. This series of schematic embryonic cross-sections illustrates the formation of the VE. (A) D/V patterning genes function to commit a group of cells to become lateral neurogenic ectoderm (N-ECT). This includes the precursors of the LE, VE and MEC (stippled). These cells give rise to both CNS and epidermis. (B) The MEC lineage (filled) becomes committed around the cellular blastoderm and gastrulation stages coincident with the expression of the sim gene. The VE and LE precursors (stippled) may be equivalent at this time. (C) After gastrulation, the MEC influences the formation of the adjacent VE precursors (arrow). The spitz class genes, including sim, influence VE cell fate during this time of development. (D) The VE precursors become committed to the VE lineage (dashed). At this time, neuroblasts (NB) delaminate from the neurogenic ectoderm and form a neuronal precursor cell layer (stippled and dashed to indicate common origin with LE and VE precursors). The MEC cells undergo nuclear migration inwards, but remain in close cytoplasmic proximity to the adjacent VE. (E) The VE cells spread out during dorsal closure and cover the ventral surface of the mature embryo. The NBs and MEC cells differentiate into the CNS with a distinctive set of midline cells (ML).

Fig. 6.

Model of VE cell commitment. This series of schematic embryonic cross-sections illustrates the formation of the VE. (A) D/V patterning genes function to commit a group of cells to become lateral neurogenic ectoderm (N-ECT). This includes the precursors of the LE, VE and MEC (stippled). These cells give rise to both CNS and epidermis. (B) The MEC lineage (filled) becomes committed around the cellular blastoderm and gastrulation stages coincident with the expression of the sim gene. The VE and LE precursors (stippled) may be equivalent at this time. (C) After gastrulation, the MEC influences the formation of the adjacent VE precursors (arrow). The spitz class genes, including sim, influence VE cell fate during this time of development. (D) The VE precursors become committed to the VE lineage (dashed). At this time, neuroblasts (NB) delaminate from the neurogenic ectoderm and form a neuronal precursor cell layer (stippled and dashed to indicate common origin with LE and VE precursors). The MEC cells undergo nuclear migration inwards, but remain in close cytoplasmic proximity to the adjacent VE. (E) The VE cells spread out during dorsal closure and cover the ventral surface of the mature embryo. The NBs and MEC cells differentiate into the CNS with a distinctive set of midline cells (ML).

The view put forth in this paper that VE identity is influenced by the MEC parallels recent work indicating that establishment of cell identity along both the A/P axis and the DE requires cell signalling (DiNardo and Heemskerk, 1990; Ferguson and Anderson, 1991). With respect to the DE, it has been proposed that the decapen taplegic gene product, which is related to the secreted vertebrate TGF-β protein (Padgett et al., 1987), may form a gradient along the dorsal regions of the embryo and specify cell identity along the dorsal ectoderm (Ferguson and Anderson, 1991, 1992). Additionally, the genes shortened gastrulation, shrew and tolloid play a role in this process. The tolloid gene product also encodes a secreted protein of the bone morphogenesis protein family (Shimell et al., 1991) as does decapentaplegic. The implication is that much of epidermal cell identity is derived via cellular interactions.

Role of the spitz class genes in VE development

The evidence presented in this paper indicates that the spitz class genes influence VE cell fate and gene expression.

Several issues arise with respect to the function of the spitz class genes in VE development. Are these genes directly involved in the interaction between MEC and VE precursors or in later steps of VE differentiation? If a spitz class gene is involved directly in MEC-VE interactions, which cell type does it function in?

sim

The sim gene encodes a nuclear protein that belongs to the basic-helix-loop-helix (bHLH) family of transcription factors and is likely to be a sequence-specific DNA-binding transcriptional activator (Nambu et al., 1991). It is first expressed as nuclear protein at the completion of gastrulation in the MEC before any overt differentiation of the CNS midline cells. It has been shown that sim gene function is required either directly or indirectly for the transcription of many and probably all genes expressed in the CNS midline cells including the spitz class gene, ve (Nambu et al., 1990). As a consequence, CNS midline precursor cells do not form, divide or differentiate. Similarly, it is likely that sim is required for the transcription of genes in the MEC that influence VE cell fate.

Egfr, spi, S and ve

As shown in this paper, these genes play a role in VE formation and there is also evidence that they may interact with each other in a number of developmental decisions. For example, they all show defects in midline glial formation, and two or more of the genes have been implicated in oogenesis, wing venation, PNS formation and retinal development (Mayer and Nüsslein-Volhard, 1988; Klämbt et al., 1991; Lindsley and Zimm, 1992; Rutledge et al., 1992). The CNS midline precursor cells of these mutants appear normal early in the germband-extended embryo when MEC influence of VE formation is first required. This reinforces the idea that these genes play a direct role in VE formation and their VE defect is not an indirect consequence of their midline glial defects.

Sequence analysis of several spitz class genes suggest they are involved in cell signalling. The Egfr gene encodes the Drosophila homologue of the EGF receptor. During stages 9 and 10, when the earliest VE effects of the spitz class genes are observed, the Egfr gene is expressed in the ventral ectoderm, but not in the MEC (Zak and Shilo, 1990). Inclusion of Egfr in this developmental pathway is more tenuous than spi, S and ve because it has a widespread ventral-lateral ectodermal defect (Clifford and Schüpbach, 1992; Raz and Shilo, 1992). However, weak alleles do show a relatively specific ventral midline cuticular effect (Clifford and Schüpbach, 1992), suggesting that the wide-spread Egfr mutant defect, which results in loss of all denticles may result from more than one developmental event.

The spi gene encodes a protein related to transforming growth factor-α (TGF-α) (Rutledge et al., 1992), which itself is related to epidermal growth factor (EGF) (Marquardt et al., 1983; Massagué, 1983). It has been shown that vertebrate TGF-α binds to and activates the EGF receptor (Todaro et al., 1980), suggesting that spi could encode a ligand for the Egfr gene product (Rutledge et al., 1992). The spi gene is widely expressed, although there is enhanced expression along the ventral midline (Rutledge et al., 1992). The ve gene encodes a transmembrane protein expressed in the blastoderm and germband-extended embryo in the LE and VE precursor cells, and also has strong expression in the MEC cells (Bier et al., 1990). The sequence and expression pattern of S has not been reported.

oc

The oc gene is also clearly required for proper VE formation. It is expressed in the VE precursor cells beginning at stage 9 and continues expression in these cells throughout embryogenesis (Wieschaus et al., 1992). Its function is required for proper CNS midline development (Finkelstein et al., 1990; Klämbt et al., 1991) and is transcribed in the CNS midline cells beginning around stage 8 of development (Finkelstein et al., 1990). Thus, oc is expressed in both the VE and CNS midline precursor cells during the time of VE formation. Since oc is a homeobox-containing protein (Finkelstein et al., 1990), it is likely involved in controlling the transcription of genes involved in VE formation.

pnt

Mutations in pnt generally showed a less severe defect than the other spitz class genes and may indicate a more limited role in VE development since the allele used behaves like a null pointed allele with respect to other embryonic defects (Klämbt, 1993). The pnt gene encodes an ets-like transcription factor and is expressed early in the ventral ectoderm (Klämbt, 1993). It is expressed in a subset of CNS midline cells, but not until germband retraction.

Existing data do not allow placing the spitz group genes in a developmental sequence, although sim functions within the MEC and Egfr and pnt initially in the ventral ectoderm. The other genes could function in either or both types of precursor cells. Additional analysis of the expression patterns of the spitz class genes genes during VE formation, as well as mosaic studies or additional experiments that can direct expression of spitz class genes to either the MEC or VE cells should confirm which of the spitz class genes are directly involved in epidermal cell development and indicate in which cell type they function.

Multiple roles of CNS midline cells in embryogenesis

The mature insect CNS midline cells consist of around 30 functional neurons and glia (Crews et al., 1988; Klämbt et al., 1991). Additional developmental roles for the CNS midline cells have been proposed including interactions with commissural axons (Klämbt et al., 1991; Bastiani et al., 1987), separation of the anterior and posterior commissures (Klämbt et al., 1991), and formation of VE (as shown in this paper). The insect CNS midline cells have a counterpart in the vertebrate embryo at the ventral base of the spinal cord (Jessell et al., 1989). These cells are important for attracting and influencing commissural growth cones (Tessier-Lavigne et al., 1988; Dodd et al., 1988). Additionally, it was shown that the ventral cells are required for proper formation of adjacent neurons (Yamada et al., 1990). This is conceptually similar to the role of the MEC on cell fate commitment of the adjacent ventral neuroectoderm, as described in this paper. It will be of interest to determine if the same genes that are involved in CNS midline formation and VE formation in Drosophila are utilized for neural formation in vertebrates.

The authors would like to express their gratitude to Bill Wickner, Owen Witte, Dan Ray, Larry Zipursky, Bill Clark and Fred Eiserling for their support and interest. We would like to thank Utpal Banerjee, Volker Hartenstein, John Nambu and Keith Wharton for comments on the manuscript, Beverley Matthews and Song Hu for isolating the enhancer trap lines from our lab, Josephine Lewis who provided the embryonic sections, and Bob Finkelstein, Song Hu, Alex Kolodkin and Norbert Perrimon for advice and reagents. This work was supported by grants from the Lucille P. Markey Charitable Trust and the NIH (HD25251). S. T.C. is a Lucille P. Markey Scholar.

Bastiani
,
M. J.
,
Harrelson
,
A. L.
,
Snow
,
P. M.
and
Goodman
,
C. S.
(
1987
).
Expression of fasciclin I and II glycoproteins on subsets of axon pathways during neuronal development in the grasshopper
.
Cell
48
,
745
755
.
Bier
,
E.
,
Jan
,
L. Y.
and
Jan
,
Y. N.
(
1990
).
rhomboid, a gene required for dorsoventral axis establishment and peripheral nervous system development in Drosophila melanogaster
.
Genes Dev
.
3
,
190
203
.
Campos-Ortega
,
J. A.
and
Hartenstein
,
V.
(
1985
).
The Embryonic Development of Drosophilamelanogaster
.
New York
:
Springer Verlag
.
Clifford
,
R.
and
Schüpbach
,
T.
(
1992
).
The torpedo (DER) receptor tyrosine kinase is required at multiple times during Drosophila embryogenesis
.
Development
115
,
853
872
.
Crews
,
S.
,
Franks
,
R.
,
Hu
,
S.
,
Matthews
,
B.
and
Nambu
,
J.
(
1992
).
The Drosophila single-minded gene and the molecular genetics of CNS midline development
.
J. Exp. Zoology
261/3
,
234
-
244
..
Crews
,
S. T.
,
Thomas
,
J. B.
and
Goodman
,
C. S.
(
1988
).
The Drosophila single-minded gene encodes a nuclear protein with sequence similarity to the per gene product
.
Cell
52
,
143
151
.
DiNardo
,
S.
and
Heemskerk
,
J.
(
1990
).
Molecular and cellular interactions responsible for intrasegmental patterning during Drosophila embryogenesis
.
Seminars in Cell Biology
1
,
173
183
.
Dodd
,
J.
,
Morton
,
S. B.
,
Karagogeos
,
D.
,
Yamamoto
,
M.
and
Jessell
,
T. M.
(
1988
).
Spatial regulation of axonal glycoprotein expression on subsets of embryonic spinal neurons
.
Neuron
1
,
105
116
.
Ferguson
,
E. L.
and
Anderson
,
K. V.
(
1991
).
Dorsal-ventral pattern formation in the Drosophila embryo: the role of zygotically active genes
.
Curr. Top. Dev. Biol
.
25
,
17
43
.
Ferguson
,
E. L.
and
Anderson
,
K. V.
(
1992
).
Localized enhancement and repression of the activity of the TGF-β family member, decapentaplegic, is necessary for dorsal-ventral pattern formation in the Drosophila embryo
.
Development
114
,
583
597
.
Finkelstein
,
R.
,
Smouse
,
D.
,
Capaci
,
T. M.
,
Spradling
,
A. C.
and
Perrimon
,
N.
(
1990
).
The orthodenticle gene encodes a novel homeo domain protein involved in the development of the Drosophila nervous system and ocellar visual structures
.
Genes Dev
.
4
,
1516
1527
.
Foe
,
V. E.
(
1989
).
Mitotic domains reveal early commitment of cells in Drosophila embryos
.
Development
107
,
1
22
.
Jessell
,
T. M.
,
Bovolenta
,
P.
,
Placzek.
M.
,
Tessier-Lavigne
,
M.
and
Dodd
,
J.
(
1989
).
Polarity and patterning in the neural tube: the origin and function of the floor plate
. In
Cellular Basis of Morphogenesis, Ciba Foundation Symposium
, Vol.
144
, pp.
255
-
280
. Chichester: Wiley.
Jürgens
,
G.
,
Wieschaus
,
E.
,
Nüsslein-Volhard
,
C.
and
Kluding
,
H.
(
1984
).
Mutations affecting the pattern of the larval cuticle in Drosophila melanogaster. II. Zygotic loci on the third chromosome
.
Roux’s Arch. Dev. Biol
.
193
,
283
295
.
Klämbt
,
C.
(
1993
).
The Drosophila gene pointed encodes two ETS-like proteins which are involved in the development of the midline glial cells
.
Development
117
,
163
176
.
Klämbt
,
C.
,
Jacobs
,
J. R.
and
Goodman
,
C. S.
(
1991
).
The midline of the Drosophila central nervous system: a model for the genetic analysis of cell fate, cell migration, and growth cone guidance
.
Cell
64
,
801
815
.
Kosman
,
D.
,
Ip
,
Y. T.
,
Levine
,
M.
and
Arora
,
K.
(
1991
).
Establishment of the mesoderm-neuroectoderm boundary in the Drosophila embryo
.
Science
254
,
118
122
.
Lawrence
,
P. A.
(
1992
).
The Making of a Fly: the Genetics of Animal Design
.
Oxford
:
Blackwell
.
Lindsley
,
D. L.
and
Zimm
,
G. G.
(
1992
).
The Genome of Drosophila melanogaster
.
San Diego
:
Academic Press, Inc
.
Marquardt
,
H.
,
Hunkapillar
,
M. W.
,
Hood
,
L. E.
,
Twardzik
,
D. R.
,
De Larco
,
J. E.
,
Stephenson
,
J. R.
and
Todaro
,
G. J.
(
1983
).
Transforming growth factors produced by retrovirus-transformed rodent fibroblasts and human melanoma cells: amino acid sequence homology with epidermal growth factor
.
Proc. Natl. Acad. Sci. USA
80
,
4684
4688
.
Massagué
,
J.
(
1983
).
Epidermal growth factor-like transforming growth factor. I. Isolation, chemical characterization and potentiation by other transforming growth factors from feline sarcoma virus-transformed rat cells
.
J. Biol. Chem
.
258
,
13606
13613
.
Mayer
,
U.
and
Nüsslein-Volhard
,
C
. (
1988
).
A group of genes required for pattern formation in the ventral ectoderm of the Drosophila embryo
.
Genes Dev
.
2
,
1496
1511
.
Nambu
,
J. R.
,
Franks
,
R. G.
,
Hu
,
S.
and
Crews
,
S. T.
(
1990
).
The single-minded gene of Drosophila is required for the expression of genes important for the development of CNS midline cells
.
Cell
63
,
63
75
.
Nambu
,
J. R.
,
Lewis
,
J. L.
,
Wharton
,
K. A.
and
Crews
,
S. T.
(
1991
).
The Drosophila single-minded gene encodes a helix-loop-helix protein which acts as a master regulator of CNS midline development
.
Cell
67
,
1157
1167
.
Nüsslein-Volhard
,
C.
,
Wieschaus
,
E.
and
Kluding
,
H.
(
1984
).
Mutations affecting the pattern of the larval cuticle in Drosophila melanogaster. I. Zygotic loci on the second chromosome
.
Roux’s Arch. Dev. Biol
.
193
,
267
282
.
O’Kane
,
C. J.
and
Gehring
,
W. J.
(
1987
).
Detection in situ of genomic regulatory elements in Drosophila
.
Proc. Natl. Acad. Sci. USA
84
,
9123
9127
.
Padgett
,
R. W.
,
St. Johnston
,
R. D.
and
Gelbart
,
W. M.
(
1987
).
A transcript from a Drosophila pattern gene predicts a protein homologous to the transforming growth factor-β family
.
Nature
325
,
81
84
.
Panzer
,
S.
,
Weigel
,
D.
and
Beckendorf
,
S. K.
(
1992
).
Organogenesis in Drosophila melanogaster: embryonic salivary gland determination is controlled by homeotic and dorsoventral patterning genes
.
Development
114
,
49
57
.
Rao
,
Y.
,
Vaessin
,
H.
,
Jan
,
L. Y.
and
Jan
,
Y.-N.
(
1991
).
Neuroectoderm in Drosophila embryos is dependent on the mesoderm for positioning but not for formation
.
Genes Dev
.
5
,
1577
1588
.
Raz
,
E.
and
Shilo
,
B.-Z.
(
1992
).
Dissection of the faint little ball (flb) phenotype: determination of the development of the Drosophila central nervous system by early interactions in the ectoderm
.
Development
114
,
113
123
.
Romani
,
S.
,
Campuzano
,
S.
and
Modolell
,
J.
(
1987
).
The achaete-scute complex is expressed in neurogenic regions of Drosophila embryos
.
EMBO J
.
6
,
2085
2092
.
Rutledge
,
B. J.
,
Zhang
,
K.
,
Bier
,
E.
,
Jan
,
Y. N.
and
Perrimon
,
N.
(
1992
).
The Drosophila spitz gene encodes a putative EGF-like growth factor involved in dorsal-ventral axis formation and neurogenesis
.
Genes Dev
.
6
,
1503
1517
.
Shimell
,
M. J.
,
Ferguson
,
E. L.
,
Childs
,
S. R.
and
O’Connor
,
M. B.
(
1991
).
The Drosophila dorsal-ventral patterning gene tolloid is related to human bone morphogenetic protein 1
.
Cell
67
,
469
481
.
Tessier-Lavigne
,
M.
,
Placzek
,
M.
,
Lumsden
,
A. G. S.
,
Dodd
,
J.
and
Jessell
,
T. M.
(
1988
).
Chemotropic guidance of developing axons in the mammalian central nervous system
.
Nature
336
,
775
778
.
Thomas
,
J. B.
,
Crews
,
S. T.
and
Goodman
,
C. S.
(
1988
).
Molecular genetics of the single-minded locus: a gene involved in the development of the Drosophila nervous system
.
Cell
52
,
133
141
.
Todaro
,
G. J.
,
Fryling
,
C.
and
De Larco
,
J. E.
(
1980
).
Transforming growth factors produced by certain human tumor cells: polypeptides that interact with epidermal growth factor receptors
.
Proc. Natl. Acad. Sci. USA
77
,
5258
5262
.
Wieschaus
,
E.
,
Nüsslein-Volhard
,
C.
and
Jürgens
,
G.
(
1984
).
Mutations affecting the pattern of the larval cuticle in Drosophila melanogaster. III. Zygotic loci on the X-chromosome and fourth chromosome
.
Roux’s Arch. Dev. Biol
.
193
,
296
307
.
Wieschaus
,
E.
,
Perrimon
,
N.
and
Finkelstein
,
R.
(
1992
).
orthodenticle activity is required for the development of medial structures in the larval and adult epidermis of Drosophila
.
Development
115
,
801
811
.
Yamada
,
T.
,
Placzek
,
M.
,
Tanaka
,
H.
,
Dodd
,
J.
and
Jessell
,
T. M.
(
1991
).
Control of cell pattern in the developing nervous system: polarizing activity of the floor plate and notochord
.
Cell
64
,
635
647
.
Zak
,
N. B.
,
Wides
,
R. J.
,
Schejter
,
E. D.
,
Raz
,
E.
and
Shilo
,
B.-Z.
(
1990
).
Localization of the DER/flb protein in embryos: implications on the faint little ball lethal phenotype
.
Development
109
,
865
874
.