The first step in generating cellular diversity in the Drosophila central nervous system is the formation of a segmentally reiterated array of neural precursor cells, called neuroblasts. Subsequently, each neuroblast goes through an invariant cell lineage to generate neurons and/or glia. Using molecular lineage markers, I show that (1) each neuroblast forms at a stereotyped time and position; (2) the neuroblast pattern is indistinguishable between thoracic and abdominal segments; (3) the development of individual neuroblasts can be followed throughout early neurogenesis; (4) gene expression in a neuroblast can be reproducibly modulated during its cell lineage; (5) identified ganglion mother cells form at stereotyped times and positions; and (6) the cell lineage of four well-characterized neurons can be traced back to two identified neuroblasts. These results set the stage for investigating neuroblast specification and the mechanisms controlling neuroblast cell lineages.

Understanding the mechanisms that generate cellular diversity during development requires a combination of descriptive, experimental and molecular genetic analyses. The power of this approach is exemplified by studies on cell determination in C. elegans and neuronal determination in the Drosophila eye, where elegant descriptive studies (e.g. Sulston et al., 1983; Ready et al., 1976) paved the way for comprehensive molecular genetic analyses (e.g. Sternberg and Horvitz, 1991; Rubin, 1991). In this and the following paper (Cui and Doe, 1992), we combine a descriptive analysis of early neurogenesis with the molecular genetic characterization of a newly identified gene that controls cell specification in the Drosophila central nervous system (CNS).

The Drosophila CNS develops from a monolayer of ventral ectodermal cells called the neurogenic region (Hartenstein and Campos-Ortega, 1984; Campos-Ortega and Hartenstein, 1985). The neurogenic region is bilaterally symmetrical, with the left and right halves separated by a narrow strip of specialized midline cells that generate both neurons and glia (Klambt et al., 1991). Individual cells within the neurogenic region delaminate into the embryo and enlarge, forming a subepidermal two-dimensional array of neural precursors, called neuroblasts (NBs; Wheeler, 1891, 1893). NBs divide asymmetrically to ‘bud off’ smaller ganglion mother cells (GMCs) into the embryo (Bauer, 1904). Each GMC divides once to produce a pair of post-mitotic neurons. Most CNS precursors are NBs, but in every hemisegment of the neurogenic region there is at least one glial precursor (GP), which divides symmetrically to generate CNS longitudinal glia (Jacobs et al., 1989), and one midline precursor (MP2), which divides only once to produce the identified dMP2 and vMP2 neurons (Doe et al., 1988a; Thomas et al., 1984). In this paper, the term NB will be used as a general term for all CNS precursors, although several cells (e.g. MP2 and GP) clearly do not divide asymmetrically and some ‘NBs’ may generate both neurons and glia (Ferdieu and Mahowald, 1989).

NBs were first identified in grasshopper embryos (Wheeler, 1891). Much later it was shown that individual NBs can be uniquely identified, each forms at a characteristic time, and the final pattern of NBs in each segment is invariant (Bate, 1976; Doe and Goodman, 1985a). The grasshopper NB map has led to several important experimental results. Laser ablations of identified NBs show that positional cues within the neurogenic region control NB identity and that cell interactions determine which cells of the neurogenic region will form NBs (Doe and Goodman, 1985b). In addition, the lineage of several identified NBs have been traced, showing that each goes through an invariant cell lineage to generate a specific set of neurons or glia (Bate and Grunewald, 1981; Goodman et al., 1982; Raper et al., 1983; Goodman et al., 1984; Taghert and Goodman, 1984; Bastiani and Goodman, 1986).

To identify the genes controlling each of these developmental events, attention has turned to the Drosophila embryo. Mutations have been identified that increase the number of NBs (Lehmann et al., 1983), decrease NB number (Jimenez and Campos-Ortega, 1979, 1990) or change the fate of NB progeny (Doe et al., 1988a,b; Duffy et al., 1991; Doe et al., 1991; Vaessin et al., 1991; Cui and Doe, 1992). Previous studies show that Drosophila has segmentally reiterated patterns of NBs (Hartenstein et al., 1987; Doe et al., 1988a; Jimenez and Campos-Ortega, 1990); however, the paucity of NB-specific markers and the inability to follow individual NBs throughout development has limited the usefulness of these maps. In addition, the relationship between NBs at each stage has not been determined and only a few NBs have been uniquely identified (Doe et al., 1988a; Martin-Bermudo et al., 1991; Skeath and Carroll, 1992).

To provide the necessary background for investigating cell specification in the Drosophila CNS, I have used antibody and enhancer trap markers to show that individual NBs and GMCs form at a stereotyped time and position, and that each NB can be uniquely identified throughout most of its embryonic cell lineage. In addition, even-skipped (eve) CNS expression is used to determine the complete cell lineage of the well characterized aCC, pCC and RP2 neurons.

Developmental stages and morphological landmarks are described in Campos-Ortega and Hartenstein (1985) and all developmental times are measured at 25°C.

Expression of the following genes were used to identify NBs and/or GMCs: engrailed (en), even-skipped (eve), wingless (wg), achaete (ac), the promoter fusion fushi tarazu-lacZ (ftz-lacZ; insertions on either the second or third chromosome). In addition, the following enhancer trap lines were used: H162 (insertion at 87B4-6, detects the seven-up [svp] pattern), 1530 (insertion at 83C1-2, detects the ming pattern), 1912 (insertion at 45C) and 5953 (insertion at 82B3).

Staged embryos were fixed and stained with antibodies to visualize subsets of NBs or their progeny. Standard methods were used to rear flies and collect embryos (Roberts, 1986); all procedures were done at room temperature except where noted. Embryos were dechorionated in 50% bleach (sodium hypochlorite; Chlorox) for 2-3 minutes and then rinsed under flowing tap water for 1 minute. Embryos were fixed in a 1:1 mix of heptane and fix (PEMFA; 100 mM Pipes, 2 mM EGTA, 1 mM MgSO4, 3.7% formaldehyde, pH 6.9). Fixation was for 12-15 minutes when staining for eve, for 35-45 minutes when staining for en and 20-25 minutes when staining for ac or the enhancer trap marker β-galactosidase (βgal). The fixative was removed and embryos were cracked out of their vitelline membranes in a 1:1 heptane:methanol mix with rapid shaking for 20 seconds. Embryos were washed twice in methanol, twice in phosphate-buffered saline (PBS pH 7.0) and then for 30 minutes in PBT (PBS, 2% BSA, 0.1% Triton X-100, pH 7.0). Primary antibodies (see below) were diluted in PBT and incubated overnight at 4°C with gentle rocking. Embryos were washed for 2 hours in PBT with the solution changed every 15-20 minutes. PBT + 2% normal goat serum was used to block non-specific antibody binding for 30 minutes, and the secondary antibodies were added directly and incubated for 1.5 hours with gentle rocking. Embryos were washed for 1-2 hours in PBT with the solution changed every 15-20 minutes. Fluorescently stained embryos were mounted in 80% glycerol containing 2% n-propyl gallate (Sigma) and the ventral surface visualized on a BioRad 600 confocal microscope. Detection of HRP-conjugated antibodies was in 0.5 mg/ml diaminobenzidine (Sigma), 0.003% H2O2, in PBS (occasionally 0.03% CoCl2 was included to intensify the reaction product). Embryos were dehydrated in a ethanol series, mounted in methyl salicylate, visualized on a Zeiss Axioplan microscope with differential interference contrast optics and photographed on Kodachrome ASA40 film.

Primary antibodies were mouse antien used at 1:1 (N. Patel and C. Goodman); mouse anti-eve used at 1:1 (N. Patel and C. Goodman); mouse anti-ac used at 1:5 (J. Skeath and S. Carroll); mouse anti-βgal used at 1:1500-1:2000 (Promega); rabbit anti-wg used at 1:1500 (M. van den Heuvel and R. Nusse); rabbit anti-en used at 1:100 (S. DiNardo and P. O’Farrell); and rabbit anti-βgal used at 1:1000 (Cappel/Organon-Teknika). The βgal antibodies were pre-absorbed at 1:10 in PBS against fixed embryos for 1 hour prior to use, which reduced background considerably. Secondary antibodies were HRP-conjugated goat anti-mouse used at 1:400 (BioRad); HRP-conjugated goat anti-rabbit used at 1:300 (Jackson Immunoresearch); lissamine rhodamine-conjugated goat anti-mouse used at 1:200 (Jackson Immunoresearch); and FITCconjugated goat anti-rabbit used at 1:500 (Jackson Immunoresearch). The secondary antibodies were often preabsorbed at 1:10 in PBS against fixed embryos for 1 hour prior to use to reduce background.

Criteria for identifying NBs were (1) located in the subectodermal cell layer, (2) located in the neurogenic region (between the midline CNS and the tracheal pits), (3) relatively large size compared to superficial cells and (4) greater cytoplasm:nuclear ratio compared to GMCs or ectodermal cells. Criteria for identifying GMCs was simply eve, hb or βgal immunoreactivity in the cells adjacent to the dorsal surface of the NBs. Overlap between identified GMCs and NBs was based on camera-lucida tracings of embryos viewed from the ventral surface (en and wg expression were used as landmarks for identifying NBs).

Morphological development of neuroblasts

NBs form by the delamination and enlargement of individual ectodermal cells within the neurogenic region. To describe this process in detail, Nomarski microscopy was used to examine the development of a single NB (4-2). The site of NB 4-2 formation is always between the easily identified NB 3-2 and NB 5-3 (described in detail below). Immediately following formation of NB 3-2, ectodermal cells at the 4-2 position have a columnar morphology, with the nuclei located at ventral cell surface (Fig. 1A). After 3-2 formation, nuclei in one to four cells at the 4-2 position move dorsally (Fig. 1B). Soon after, one cell begins enlarging (Fig. 1C) and ultimately delaminates from the ventral ectoderm to form NB 4-2. Occasionally two cells will begin enlarging at the 4-2 position; one of these cells may die or revert back to an ectodermal fate, since embryos fixed at later stages always show a single NB 4-2. (While it is uncommon to observe dual enlarging cells at one NB position, it can occur at all stages, from S1 to S5 NBs; data not shown.) NB 4-2 divides asymmetrically, budding off a smaller GMC from its dorsal surface (Fig. 1D). Although the first GMC from NB 4-2 is displaced dorsally (inwards), other NB divisions show different, reproducible orientations (discussed in detail below). The dorsal processes of the ectodermal cells are withdrawn and re-establish a columnar morphology following GMC 4-2a formation.

Fig. 1.

Morphological differentiation of neuroblasts. The development of NB 4-2 is drawn based on camera-lucida tracings of embryos viewed with Nomarski optics. (A) Ectodermal cells at the 4-2 position have a uniform columnar morphology, with the nuclei located at the ventral surface of the cells. (B) The nuclei of one to four cells at the 4-2 position move towards the dorsal cell surface. (C) One cell begins to delaminate into the embryo; it shifts both cytoplasm and the nucleus dorsally relative to adjacent cells. (D) Delamination is complete and the new NB 4-2 divides asymmetrically, budding off a smaller ganglion mother cell (top cell) at its dorsal surface. Subsequently, the adjacent ectodermal cells withdraw their dorsal processes and regain a columnar morphology.

Fig. 1.

Morphological differentiation of neuroblasts. The development of NB 4-2 is drawn based on camera-lucida tracings of embryos viewed with Nomarski optics. (A) Ectodermal cells at the 4-2 position have a uniform columnar morphology, with the nuclei located at the ventral surface of the cells. (B) The nuclei of one to four cells at the 4-2 position move towards the dorsal cell surface. (C) One cell begins to delaminate into the embryo; it shifts both cytoplasm and the nucleus dorsally relative to adjacent cells. (D) Delamination is complete and the new NB 4-2 divides asymmetrically, budding off a smaller ganglion mother cell (top cell) at its dorsal surface. Subsequently, the adjacent ectodermal cells withdraw their dorsal processes and regain a columnar morphology.

Neuroblasts form in a stereotyped array

Camera-lucida tracings were made of NBs in fixed, staged embryos (Fig. 2; for the criteria used to define a NB, see Materials and methods). Only the bilateral neurogenic region is addressed in this paper; development and lineage of the midline CNS has already been elegantly described (Klambt et al., 1991). NB names have been chosen to maximize homology to grasshopper NBs — there are many similarities in timing of formation, pattern of gene expression and cell lineage of specific neural precursors (see Discussion). However, the nomenclature is not meant to imply that every Drosophila NB has a definitive grasshopper homolog. NB formation is divided into six stages: early S1 (eS1), S1, S2, S3, S4 and S5. Phases S1-S3 are similar to previously described S1-S3 NB maps (Hartenstein and CamposOrtega, 1984; Hartenstein et al., 1987). It is important to note that these stages are ‘snapshots’ of the developing NB pattern and intermediate patterns can be observed (this is particularly true between the S3, S4 and S5 stages).

Fig. 2.

Development of the embryonic neuroblast pattern. Camera-lucida tracings the NBs in one segment and the adjacent anterior en+ neuroblasts (rows 6, 7 and NB 1-1). Each panel is drawn from a single embryo of a different age. The newest NBs of each stage are shown in grey on the left hemisegment, highlighting the ‘columnar’ organization of NB formation. Ventral midline is indicated by the vertical line. See text for discussion of individual NBs at each stage. (A) The early S1 pattern; late stage 8. (B) The S1 pattern; early stage 9. (C) The S2 pattern; stage 9. (D) The S3 pattern; stage 10. NB 6-4 is not visible in the left hemisegment. (E) The S4 pattern; stage 11. (F) The S5 pattern; late stage 11. NB 5-5 is not visible in this embryo.

Fig. 2.

Development of the embryonic neuroblast pattern. Camera-lucida tracings the NBs in one segment and the adjacent anterior en+ neuroblasts (rows 6, 7 and NB 1-1). Each panel is drawn from a single embryo of a different age. The newest NBs of each stage are shown in grey on the left hemisegment, highlighting the ‘columnar’ organization of NB formation. Ventral midline is indicated by the vertical line. See text for discussion of individual NBs at each stage. (A) The early S1 pattern; late stage 8. (B) The S1 pattern; early stage 9. (C) The S2 pattern; stage 9. (D) The S3 pattern; stage 10. NB 6-4 is not visible in the left hemisegment. (E) The S4 pattern; stage 11. (F) The S5 pattern; late stage 11. NB 5-5 is not visible in this embryo.

The eS1 NBs develop at late stage 8 (staging according to Campos-Ortega and Hartenstein, 1985). There are nine NBs per hemisegment, in an incomplete orthogonal array, with four NBs in the medial column, one NB (5-3) in the intermediate column and four NBs in the lateral column (Fig. 2A). Development of NB 3-2 at early stage 9 completes the S1 array (Fig. 2B); NBs 3-2 and 5-3 are usually closer together in the same segment than between segments, allowing all S1 NBs to be identified by position alone.

The S2 NBs enlarge at stage 9, primarily in the intermediate column (Fig. 2C). Each S2 NB can be uniquely identified by position, with orientation provided by a slight ectodermal groove between NB rows 5 and 6 (the parasegmental groove), as well as the more medial position of NB 2-5 relative to the other three NBs in the lateral column (3-5, 5-6 and 7-4). The S3 NBs develop at stage 10, in columns at the medial and lateral edges of the NB array (Fig. 2D). One of the S3 ‘NBs’ is actually a Glial Precursor (GP; Doe et al., 1988a; Jacobs et al., 1989). The mediolateral position, size and temporal profile of the GP is matched by a nearby neural precursor, called ‘X.’ The fate of this cell is unknown; its similarity to GP suggests it may also generate glia. Although it is usually possible to identify each S3 NB based on its position relative to the GP and X cells, unambiguous NB identification at this stage requires the use of molecular markers (see below).

The S4 and S5 NBs develop at stage 11 and late stage 11, respectively (Fig. 2E,F). Both S4 and S5 NBs form in columns between the existing intermediate and lateral columns of NBs; after these NBs form, it is difficult to discern clear columns or rows of NBs, due to variability in NB positions. Formation of the S5 NBs brings the total number of NBs to 31 in each hemisegment, in addition to the MNB. However, four of these precursors are not present in the S5 array due to reductive divisions to generate neurons or glia (MP2, GP) or unknown events (X, 2-2). The combination of variable NB positions and disappearance of some precursors makes it impossible to identify NBs by their position alone. Accurate identification of S4 and S5 NBs requires molecular markers (described below).

Molecular markers allow individual neuroblasts to be uniquely identified

Eight molecular markers have been used to assay a different, overlapping population of NBs at several stages of neurogenesis; expression of these markers is summarized in Fig. 3 and examples of each are shown in Fig. 4. Antibodies were used to detect the ac (Skeath and Carroll, 1992) (Fig. 4A), wg (van den Heuvel et al., 1989) (Fig. 4C) and en (DiNardo et al., 1985) (Fig. 4D) proteins. The product of the lacZ gene was detected in the promoter fusion construct ftzlacZ (Hiromi et al., 1985) (Fig. 4B). In addition, the lacZ pattern was assayed in four enhancer trap lines: H162, which detects the svp expression pattern (Mlodzik et al., 1990) (Fig. 4F; Fig. 5); 1530, which detects the ming expression pattern (Cui and Doe, 1992) (Fig. 4G); 5953 (C. Q. D., unpublished results) (Fig. 4E); and 1912 (C. Q. D., unpublished results) (Fig. 4H). The expression of lacZ in the enhancer trap or gene fusion lines will be referred to by the name of the line or adjacent gene (e.g. “1912 and svp are expressed…” indicates expression of lacZ in embryos from the 1912 and H162 enhancer trap lines, respectively).

Fig. 3.

Summary of molecular markers expressed in neuroblasts. Summary of the ‘consensus pattern’ of NBs at different developmental stages, with the expression of eight molecular markers indicated in color. ac, en and wg represent the respective protein patterns; ftz, svp, ming, 1912 and 5953 represent βgal patterns; see text for details. (A) Early SI NBs; late stage 8. (B) SI NBs; early stage 9. (C) S2 NBs; stage 9. (D) S3 NBs; stage 10. (E) S4 NBs; stage 11. (F) S5 NBs; late stage 11. Anterior, top; ventral midline, dotted line; large circles, NBs; small black spots, sites of NB formation at the next stage.

Fig. 3.

Summary of molecular markers expressed in neuroblasts. Summary of the ‘consensus pattern’ of NBs at different developmental stages, with the expression of eight molecular markers indicated in color. ac, en and wg represent the respective protein patterns; ftz, svp, ming, 1912 and 5953 represent βgal patterns; see text for details. (A) Early SI NBs; late stage 8. (B) SI NBs; early stage 9. (C) S2 NBs; stage 9. (D) S3 NBs; stage 10. (E) S4 NBs; stage 11. (F) S5 NBs; late stage 11. Anterior, top; ventral midline, dotted line; large circles, NBs; small black spots, sites of NB formation at the next stage.

Fig. 4.

Molecular markers expressed in identified neuroblasts. (A) ac protein at SI NB stage. MP2, left arrow; 3-5, right arrow; 7-1, left arrowhead; 7-4, right arrowhead. (B) βgal protein in ftzlacZ gene fusion embryos at S3 NB stage. Arrow, MP2; large arrowhead, GP. The anterior border of the residual βgal (from bi-segmental blastoderm expression) demarcates the compartment border, with the unstained row 5 NBs just anterior to the stripe; NBs 6-2 and 7-2 containing inherited βgal protein are visible (upper and lower small arrowheads, respectively). (C) wg protein at S3 NB stage. NBs 5-2, 5-3 and 5-6, arrows from right to left, respectively; unstained NB 4-2 (arrowhead) is closely apposed to NB 5-3. This pattern corresponds exactly to the expression of wg in a ‘secretion defective’ mutation (wgIL114), and thus the pattern shown probably represents the cells actually expressing wg. (D) en protein at S4 NB stage. NBs 6-1,7-1, 1-1 and 7-4, arrows from right to left, respectively; MNB, arrowhead. The en+ NBs 6-2 and 7-2 are barely visible between NB 1-1 and NB 7-4 (beneath the en+ ectodermal cells). (E) βgal protein in 5953 embryos at S4 NB stage. Darkly stained NBs 2-1 and 2-3, right and left large arrowheads, respectively; NBs 4-2 and 5-4, top and bottom arrows, respectively; unstained NB 1-1, small arrowhead. Because the embryo is curved, small ectodermal cells lateral to the NBs are in focus; these clusters will give rise to 5953-expressing NBs. (F) βgal protein in Hl62 embryos (svp enhancer trap line) at S4 NB stage. NBs 5-2 and 7-1, top and bottom large arrowheads, respectively (the ‘in between’ unstained NB 6-1 can be seen more clearly in the opposite hemisegment); NBs 3-2 and 1-1. top and bottom small arrowheads, respectively; NBs 7-4 and 2-5, top and bottom arrows, respectively. (G) βgal protein in 1530 embryos (ming enhancer trap line) at S5 NB stage. NBs 5-1, 5-2,5-3, right to left black arrowheads, respectively; NBs 7-2 and 7-4, right and left arrows, respectively; NBs 3-1, 3-2, 3-3, 3-4, right to left white arrowheads, respectively. (H) βgal protein in 1912 embryos at S5 NB stage. NBs 4-1, 5-3, 6-2, 5-5, right to left arrows, respectively. In all panels anterior is up and the ventral midline is indicated (black triangle).

Fig. 4.

Molecular markers expressed in identified neuroblasts. (A) ac protein at SI NB stage. MP2, left arrow; 3-5, right arrow; 7-1, left arrowhead; 7-4, right arrowhead. (B) βgal protein in ftzlacZ gene fusion embryos at S3 NB stage. Arrow, MP2; large arrowhead, GP. The anterior border of the residual βgal (from bi-segmental blastoderm expression) demarcates the compartment border, with the unstained row 5 NBs just anterior to the stripe; NBs 6-2 and 7-2 containing inherited βgal protein are visible (upper and lower small arrowheads, respectively). (C) wg protein at S3 NB stage. NBs 5-2, 5-3 and 5-6, arrows from right to left, respectively; unstained NB 4-2 (arrowhead) is closely apposed to NB 5-3. This pattern corresponds exactly to the expression of wg in a ‘secretion defective’ mutation (wgIL114), and thus the pattern shown probably represents the cells actually expressing wg. (D) en protein at S4 NB stage. NBs 6-1,7-1, 1-1 and 7-4, arrows from right to left, respectively; MNB, arrowhead. The en+ NBs 6-2 and 7-2 are barely visible between NB 1-1 and NB 7-4 (beneath the en+ ectodermal cells). (E) βgal protein in 5953 embryos at S4 NB stage. Darkly stained NBs 2-1 and 2-3, right and left large arrowheads, respectively; NBs 4-2 and 5-4, top and bottom arrows, respectively; unstained NB 1-1, small arrowhead. Because the embryo is curved, small ectodermal cells lateral to the NBs are in focus; these clusters will give rise to 5953-expressing NBs. (F) βgal protein in Hl62 embryos (svp enhancer trap line) at S4 NB stage. NBs 5-2 and 7-1, top and bottom large arrowheads, respectively (the ‘in between’ unstained NB 6-1 can be seen more clearly in the opposite hemisegment); NBs 3-2 and 1-1. top and bottom small arrowheads, respectively; NBs 7-4 and 2-5, top and bottom arrows, respectively. (G) βgal protein in 1530 embryos (ming enhancer trap line) at S5 NB stage. NBs 5-1, 5-2,5-3, right to left black arrowheads, respectively; NBs 7-2 and 7-4, right and left arrows, respectively; NBs 3-1, 3-2, 3-3, 3-4, right to left white arrowheads, respectively. (H) βgal protein in 1912 embryos at S5 NB stage. NBs 4-1, 5-3, 6-2, 5-5, right to left arrows, respectively. In all panels anterior is up and the ventral midline is indicated (black triangle).

Fig. 5.

Molecular marking of individual neuroblasts throughout neurogenesis. Expression of βgal in H 162 embryos (svp enhancer trap line) can be used to follow identified NBs throughout early neurogenesis; NBs 5-2 (arrow) and 7-4 (arrowhead) arc indicated. (A) SI stage. Only NB 5-2 expresses βgal. (B) S2 stage. In addition to NB 5-2 and NB 7-4, other βgal-positive NBs are 3-2 (anterior and lateral to 5-2) and 7-1 (posterior to 5-2). NBs 2-5, 5-3 and 7-2 express βgal but are not visible in this focal plane. (C) S4 stage. In addition to NBs 5-2 and 7-4, βgal can be observed in 5-3 (lateral to 5-2), 7-2 (medial to 7-4) and 2-5 (posterior to 7-4); other NBs express βgal but are not in the plane of focus. (D) S5 stage. NBs 5-2 and 7-4 are indicated; expression of gal in other NBs is summarized in Fig. 3. Ventral midline, triangle; anterior al Lup.

Fig. 5.

Molecular marking of individual neuroblasts throughout neurogenesis. Expression of βgal in H 162 embryos (svp enhancer trap line) can be used to follow identified NBs throughout early neurogenesis; NBs 5-2 (arrow) and 7-4 (arrowhead) arc indicated. (A) SI stage. Only NB 5-2 expresses βgal. (B) S2 stage. In addition to NB 5-2 and NB 7-4, other βgal-positive NBs are 3-2 (anterior and lateral to 5-2) and 7-1 (posterior to 5-2). NBs 2-5, 5-3 and 7-2 express βgal but are not visible in this focal plane. (C) S4 stage. In addition to NBs 5-2 and 7-4, βgal can be observed in 5-3 (lateral to 5-2), 7-2 (medial to 7-4) and 2-5 (posterior to 7-4); other NBs express βgal but are not in the plane of focus. (D) S5 stage. NBs 5-2 and 7-4 are indicated; expression of gal in other NBs is summarized in Fig. 3. Ventral midline, triangle; anterior al Lup.

The combination of position and gene expression is sufficient to uniquely identify all NBs at each stage of neurogenesis. For example, among the medial column of S1 NBs, 2-2 expresses no marker genes, MP2 expresses ac, 5-2 expresses wg and svp, and 7-1 expresses ac and en (Fig. 3B). Among S2 lateral column NBs, 2-5 expresses svp, 3-5 expresses ac, 5-6 expresses wg and 7-4 expresses ac, svp and en (Fig. 3C). The molecular identification of NBs after S4 formation is possible, despite the complexity of the pattern. Markers for a scattered array of NBs (e.g. 1912 and 5953) provide local landmarks that facilitate identification of NBs in most areas of the pattern.

Based on previous NB maps in grasshopper (Bate, 1976; Doe and Goodman, 1985a) and Drosophila (Hartenstein and Campos-Ortega, 1984; Hartenstein et al., 1987), I expected differences in the NB array between thoracic and abdominal segments. In fact, each molecular marker for S5 NBs is expressed identically in thoracic and abdominal segments. However, many thoracic NBs form earlier than their abdominal homologs, such that abdominal segments can be up to one NB stage behind that observed in thoracic segments in a single embryo (data not shown). Segmental differences in the NB array can be found in the extreme termini of the embryo: the first subesophageal (S1) and ninth abdominal (A9) segments have a reduced number of NBs compared to the intervening segments (data not shown).

Individual neuroblasts can be identified throughout neurogenesis

In addition to identifying NBs at each stage of development, individual NBs can be followed from one stage to the next. With this information, it should be possible to determine the ultimate fate of a precursor, the changes in gene expression within a NB cell lineage and the fate of the GMCs produced from a single NB. Most NB markers are expressed in specific subsets of NBs over several hours of development and thus they allow identification of specific, individual NBs at most or all stages of neurogenesis. In addition, several markers have transient coexpression in specific NBs, allowing the NB to be initially identified by one marker and subsequently identified by the second marker.

A good example of a marker that can be used to follow specific NBs throughout neurogenesis is svp (Fig. 5). svp is expressed in NB 5-2 from S1 to S5 stages and NB 7-4 from S2 to S5 stages, allowing these NBs to be followed throughout neurogenesis, where they can be used as markers to identify surrounding NBs (Figs 4F, 5). Other genes are also expressed in the same NBs at multiple stages of development (e.g. en and wg; Fig. 3). To confirm the relative positions of identified NBs at each stage, embryos were double-labeled for different pairs of markers. Fig. 6A shows an embryo labeled for en and ming expression; clearly most en-expressing (en+) NBs co-express ming, although several NBs express only ming. Similar double labels show that most en+ NBs do not express 5953 (Fig. 6B). Double labels have also been done for the following pairs of genes: en/ac, en/wg, en/ftz, en/svp, en/1912, ac/svp and ac/ming (data summarized in Fig. 3).

Fig. 6.

Dual localization of markers for identified neuroblasts. (A) Embryos from the 1530 (ming) enhancer trap line were stained for βgal (red) and en (green) in S5 NB array. Upper arrowhead indicates 3-3 (right) and 3-4 (left); lower arrowhead indicates 5-5. Co-expression can be seen in 6-1 (arrow; yellow). ming expression is an overlay of confocal images from 2 focal planes; the en pattern is from only one of these focal planes. (B) Embryos from the 5953 enhancer trap line were stained for βgal (red) and en (green) in an early S5 NB array (NB 5-5 is not yet present). Expression of 5953 in NBs 2-4, 4-3, 5-4, arrowheads from top to bottom, respectively; expression of en in NB 6-2 (open arrow; other en+ NBs are mostly out of the focal plane). Co expression can be observed in 7-3 (solid arrow; yellow). The embryo is curved, with the focal plane on the neuroecloderm in the top quarter of the photo and on the NB layer in the bottom of the photo; 5953 transiently expresses βgal in clusters of ectodermal cells (small cells at top -of photo) prior to expression in the delaminating NBs. Ventral midline, triangle; anterior at top.

Fig. 6.

Dual localization of markers for identified neuroblasts. (A) Embryos from the 1530 (ming) enhancer trap line were stained for βgal (red) and en (green) in S5 NB array. Upper arrowhead indicates 3-3 (right) and 3-4 (left); lower arrowhead indicates 5-5. Co-expression can be seen in 6-1 (arrow; yellow). ming expression is an overlay of confocal images from 2 focal planes; the en pattern is from only one of these focal planes. (B) Embryos from the 5953 enhancer trap line were stained for βgal (red) and en (green) in an early S5 NB array (NB 5-5 is not yet present). Expression of 5953 in NBs 2-4, 4-3, 5-4, arrowheads from top to bottom, respectively; expression of en in NB 6-2 (open arrow; other en+ NBs are mostly out of the focal plane). Co expression can be observed in 7-3 (solid arrow; yellow). The embryo is curved, with the focal plane on the neuroecloderm in the top quarter of the photo and on the NB layer in the bottom of the photo; 5953 transiently expresses βgal in clusters of ectodermal cells (small cells at top -of photo) prior to expression in the delaminating NBs. Ventral midline, triangle; anterior at top.

Molecular markers not only allow NBs to be followed from one stage to the next, they also can distinguish between the displacement of a NB to a new position and the disappearance of a NB. For example, ming expression can be used to follow the lateral displacement of NB 3-4 between stages S4 and S5 (Fig. 3). In contrast, some precursors disappear from the NB pattern altogether. ftz expression can be used to follow the disappearance of the MP2 and GP precursors from the NB pattern. MP2 delaminates and enlarges like a NB, but it moves dorsally and laterally before dividing symmetrically to produce the dMP2 and vMP2 neurons (Doe et al., 1988a), whereas the GP divides symmetrically and generates longitudinal glia (Doe et al., 1988a; Jacobs et al., 1989). Interestingly, ‘NB’ 2-2 has a similar developmental profile to MP2: it delaminates, enlarges, moves dorsally and disappears from the NB pattern concurrently with MP2 (Fig. 3). It is not known whether 2-2 divides symmetrically, like MP2, or asymmetrically, like a NB.

Neuroblasts reproducibly alter gene expression during their cell lineage

Using the markers described above, it is possible to trace the pattern of gene expression in individual NBs throughout their early cell lineages. It is clear that gene expression can be reproducibly altered during the cell lineage of an identified NB. For example, ac is initially expressed in three NBs and MP2, but is turned off at about the time that the NBs first divide (Figs 3, 4A). In contrast, several genes are initially expressed midway through specific NB lineages. svp is detectable in NB 1-1 only after it has generated its first progeny (Fig. 3E). ming enhancer trap line and transcript are also expressed at different points within identified NB cell lineages (described in detail in Cui and Doe, 1992). These results clearly show that NBs reproducibly switch genes on or off at specific points in their cell lineage. The mechanisms regulating differential NB gene expression are unknown.

Because NBs appear to change identity during the course of their lineage (as reflected by differential gene expression), it is necessary to develop a nomenclature that distinguishes between a NB at each point in its cell lineage. Here I propose a method for uniquely naming each cell in the developing CNS, using NB 1-1 as an example. At the time of formation, the NB is called 1-1; following its first division, the NB is termed 1-1A and the GMC is 1-1a (the same letter indicates sibling cells, with upper case denoting the NB and the lower case indicating the GMC). Subsequent NB divisions produce new NBs (1-1B, 1-1C, etc) and their sibling GMCs (1-1b, 1-1c, etc). The pair of neurons from each GMC are named according to their parental GMC (e.g. 1-1a1 and 1-1a2). This system allows the unique naming of every NB, GMC and neuron throughout neurogenesis.

Several identified GMCs form in a stereotyped position due to the reproducible orientation of neuroblast divisions

Individual GMCs were identified using a variety of molecular markers and camera-lucida tracings were used to establish the spatial relationship between NBs and GMCs (Fig. 7). A close association between GMC and NB is presumably because the NB has recently generated the GMC (i.e. it is the parental NB). Many NBs examined to date (1-1, 2-5, 3-5, 4-2, 5-2, 5-6 and 7-4) have a preferred orientation of NB cell division, ranging from perfectly dorsal (perpendicular to the embryo surface) to nearly lateral (parallel to the embryo surface); other identified GMCs have a more variable position (Fig. 7). The nuclear hunchback (hb) protein can be detected in all NBs and many of the earlyborn GMCs; the position of the first GMCs born from the lateral column of S2 NBs is summarized in Fig. 7A. Each of these NBs usually divides to produce a GMC positioned dorsolaterally from the NB (plane of division about 45° to the embryo surface), although GMC 2-5a tends to be located more dorsally than the other GMCs.

Fig. 7.

Identified GMCs are generated at stereotyped positions. Summary of camera lucida tracings of NBs and GMCs in ventral viewed embryos. Positions of GMCs were overlayed on a consensus map of NBs; the GMC position reflects its exact relationship to the closest NB and its approximate relationship to the other NBs. (A) Position of hb expressing GMCs (light gray) derived from the first division of the lateral S1 NBs (shown at S2 NB stage). The NBs 2-5, 3-5, 5- 6 and 7-4 undergo stereotyped divisions resulting in a lateral GMC position. (B) Position of svp-expressing (dark gray) and eve-expressing (light gray) GMCs relative to S4 stage NBs. The svp-expressing GMCs are probably the second GMC from each NB, whereas the eve-expressing GMCs are probably the first GMC from each NB (see text). The GMCs 5-2b and 7-4b are reproducibly generated from the lateral surface of their parental NB. The eve-expressing GMCs are positioned less precisely but it is still possible to determine NBs 1-1 and 4-2 each generate an eve+ GMC, and NB 6-2 probably generates an eve+ GMC. The eve-expressing GMCs 1-la and putative 6-2a were traced at the S3 stage and GMC 4-2a was traced at the S4 stage; for convenience all GMCs are represented on an S4 NB map. Number ofGMCs traced: 2-5a, 21; 3-5a, 20; 5-6a, 15; 1-la, 28; 4-2a, 15; 5-2b, 24; ‘6-2a:’, 31; 7-4a, 12; 7-4b, 20.

Fig. 7.

Identified GMCs are generated at stereotyped positions. Summary of camera lucida tracings of NBs and GMCs in ventral viewed embryos. Positions of GMCs were overlayed on a consensus map of NBs; the GMC position reflects its exact relationship to the closest NB and its approximate relationship to the other NBs. (A) Position of hb expressing GMCs (light gray) derived from the first division of the lateral S1 NBs (shown at S2 NB stage). The NBs 2-5, 3-5, 5- 6 and 7-4 undergo stereotyped divisions resulting in a lateral GMC position. (B) Position of svp-expressing (dark gray) and eve-expressing (light gray) GMCs relative to S4 stage NBs. The svp-expressing GMCs are probably the second GMC from each NB, whereas the eve-expressing GMCs are probably the first GMC from each NB (see text). The GMCs 5-2b and 7-4b are reproducibly generated from the lateral surface of their parental NB. The eve-expressing GMCs are positioned less precisely but it is still possible to determine NBs 1-1 and 4-2 each generate an eve+ GMC, and NB 6-2 probably generates an eve+ GMC. The eve-expressing GMCs 1-la and putative 6-2a were traced at the S3 stage and GMC 4-2a was traced at the S4 stage; for convenience all GMCs are represented on an S4 NB map. Number ofGMCs traced: 2-5a, 21; 3-5a, 20; 5-6a, 15; 1-la, 28; 4-2a, 15; 5-2b, 24; ‘6-2a:’, 31; 7-4a, 12; 7-4b, 20.

The position of later-born GMCs can be assayed by svp and eve expression (Fig. 7B). Precise GMC positioning occurs following division of NB 5-2, where both svp-expressing GMCs 5-2a and 5-2b are generated from the lateral surface of the NB (only GMC 5-2b is shown in Fig. 7B). GMC 5-2b is distinguished from GMC 5-2a due to its closer position to NB 5-2 at the stage examined. svp can also be used to identify a GMC from NB 7-4, which is probably the second GMC (7-4b), based on its position next to the NB, its time of formation and the presence of an earlier GMC (7-4a) detectable by hb expression at the S2 NB stage (Fig. 7A). Both 7-4a and 7-4b have different, yet stereotyped, positions: 7-4a tends to be produced from the lateral surface of NB 7-4, whereas 7-4b is usually generated from the dorsal surface of the NB (Fig. 7). Expression of eve can also be used to identify uniquely three GMCs; two GMCs are located adjacent to NBs 1-1 and 4-2, whereas the third eve+ GMCs has a more variable position (Fig. 7b). The lineage of the eve+ GMCs is discussed below. Results from using hb, svp and eve expression to identify uniquely GMCs shows that some NB divisions have a characteristic or preferred orientation, which may change with each division.

Development of identified neurons: cell lineage from neuroblast to neuron

Camera-lucida tracings of eve+ GMCs shows that two GMCs are reproducibly associated with a specific NB (Fig. 7B; Table 1). eve+ GMCs are closely associated with NBs 1-1 and 4-2. The eve+ GMC 1-1a is observed soon after formation of NB 1-1, suggesting that it is the first progeny of this NB, similar to the lineage observed in grasshopper embryos (Goodman et al., 1984). The eve+ GMC near NB 4-2 expresses eve later than 1-la, but it appears to be the only GMC near NB 4-2 and hence it is probably the first born GMC (4-2a). A third eve+ GMC forms between NBs 6-2, 7-1 and 1-1; although it is most closely associated with NB 6-2, it could be derived from any of these NBs. Due to the relatively small number of eve+ cells in the CNS (20 per abdominal hemisegment; Doe et al., 1988a,b; Patel et al., 1989a; C. Q. D., unpublished results) the fate of these eve+ GMC can be followed by examining successively older embryos (data not shown). The eve+ GMC 1-la devel ops into neurons l lal and 1-la2, which are the well described aCC and pCC neurons (Thomas et al., 1984; Doe et al., 1988a,b; Patel et al., 1989a). The eve+ GMC 4-2a develops into the RP2 motoneuron (4-2al; Thomas et al., 1984; Patel et al., 1989a; Sink and Whitington, 1990, 1991; Halpern et al., 1991) and RP2sib (4-2a2), which either undergoes programmed cell death or ceases to express eve soon after it is born (C. Q. D., unpublished results). The eve+ GMC near NB 6-2 develops into a pair of neurons that are part of the 6-cell U/CQ group (Patel et al., 1989a; N.H. Patel, personal communication). These results are the first studies in Drosophila to trace cell lineages from iden tified NBs to identified neurons.

Table 1.

Lineage of the aCC, pCC and RP2 neurons

Lineage of the aCC, pCC and RP2 neurons
Lineage of the aCC, pCC and RP2 neurons

Morphological devel.opment of neuroblasts and GMCs Development of Drosophila

NBs is similar in many ways to that observed in the larger grasshopper embryo. In both organisms, NBs delaminate into the embryo, enlarge and divide asymmetrically. In Drosophila, the adjacent neu roectodennal cells initially extend processes around the NB, but these are withdrawn soon after the NB begins dividing; in grasshopper, these cells differentiate into ‘sheath cells’ which maintain processes that enwrap the family of GMCs and neurons produced from a single NB (Doe and Good man, 1985a). One consequence of the lack of sheath cells in Drosophila is that the progeny of individual NBs are not ‘packaged’ together in a column; instead, GMCs and neurons from adjacent NBs intermingle in the CNS, as was first observed by Hartenstein et al. (1987). This does not mean that all GMC positions are random. Using markers for identified GMCs, it is clear that some (but not all) GMCs are born in a stereotyped position, reflecting a reproducible orientation of the NB division. It is not clear whether a NB divides with the same orientation every time. NB 5-2 generates its first two progeny (GMCs 5-2a and 5-2b) from its lateral surface, whereas NB 7-4 produces two GMCs at slightly different dorsal and dorsolateral positions. The mechanisms regulating oriented NB divisions are unknown.

Development of the neuroblast pattern

The pattern of NBs in Drosophila has been partially characterized (Hartenstein and Campos-Ortega, 1984; CamposOrtega and Hartenstein, 1985; Hartenstein et al., 1987; Doe et al., 1988a; Jimenez and Campos-Ortega, 1990; Skeath and Carroll, 1992). Each study described the pattern of NBs at one or more stages of development, but usually did not determine the relationship between NBs at each of the stages. In contrast, the pattern of NBs has been well characterized in the grasshopper embryo (Bate, 1976; Doe and Goodman, 1985a) and there are many similarities to the pattern of Drosophila NBs described in this paper. Both organisms have about 30 NBs per hemisegment, including a glial precursor (GP) that generates longitudinal glia, and the MP2 precursor that produces the dMP2 and vMP2 neurons. All of the cell types described in Drosophila are present in grasshopper. (The grasshopper has one cell type not present in Drosophila, the cap cell, which is attached to the ventral surface of every NB and is required for their asymmetric divisions [Kawamura and Carlson, 1962].) As first described by Hartenstein and Campos-Ortega (1984), Drosophila NBs form in ‘waves,’ with each new population of NBs arranged roughly in columns. This is true for grasshopper embryos as well (Doe and Goodman, 1985a), and is confirmed for the later-forming S4 and S5 NBs described in this paper.

Drosophila

NBs are named after their approximate homolog in grasshopper, based on primarily on relative position (although time of formation, gene expression and cell lineage data are considered as well). For example, in both grasshopper and Drosophila the en gene is expressed in NBs of rows 6 and 7, and the eve gene is expressed in the first progeny of NBs 1-1, 4-2 and 6-2 (this paper; N. H. Patel, personal communication). Similarly, in both grasshopper and Drosophila, the MP2 and GP appear in the same position, divide similarly and generate the identical neuronal or glial progeny. Thus at least 1-1, 4-2, 6-2, GP, MP2 and the NBs in rows 6 and 7 share some degree of homology between organisms. Nevertheless, identically named NBs in each organism are not necessarily completely homologous. For example, NB 1-1 produces the aCC and pCC neurons in both organisms, yet the en gene is expressed in NB 1-1 in Drosophila but not in grasshopper (this paper; Patel et al., 1989b). The extent of similarity between positionally homologous precursors in grasshopper and Drosophila will require cell lineage analysis and gene expression studies in both organisms.

Grasshopper embryos have clear, but subtle, differences between thoracic and abdominal NB arrays (Bate, 1976; Doe and Goodman, 1985a), yet in Drosophila these differences are not apparent. Segment-specific neurons or glia in Drosophila may be generated by segmental differences in NB lineages or by postembryonic NB formation. Alternatively, segmental differences in embryonic NBs may exist in Drosophila,but go undetected due to lack of markers, the smaller size of the precursors and/or NB formation after the S5 stage.

Neuroblast gene expression

Eight molecular markers have been used to map the array of NBs throughout early neurogenesis. Many NBs at various stages can be uniquely identified based on expression of one or more genes (and all NBs can be identified by their position relative to nearby, marked NBs). Greater precision in NB identification should be possible after the mapping of at least 9 other genes expressed in subsets of NBs (runt [Duffy et al., 1991], gooseberry-early and gooseberry-late [N. H. Patel, personal communication], pox-neuro [Dambly-Chaudiere et al., 1992], dPOU-19 and dPOU-28 [Dick et al., 1991], cut [C. Q. D., unpublished results], scute [Cabrera, 1990] and lethal-of-scute [Martin-Bermudo et al., 1991]). In addition to providing markers for specific NBs, gene expression patterns may reveal co-expression that reflects regulatory interactions occurring in NBs and/or GMCs. For example, the ming and en genes show substantial co-expression in NBs and GMCs, which led to the observation that ming function is required for normal en expression in the CNS (Cui and Doe, 1992).

Many genes are expressed in NBs only during portions of their cell lineage, either only early in the lineage (e.g. ac) or only late in the lineage (e.g. ming, svp, 5953). The change in gene expression is extremely reproducible, but the function of differential gene expression within a NB lineage is not clear. It is likely that differential gene expression within a NB lineage contributes to unique GMC specification. While this remains to be tested in detail, it is known that the ming gene, which is expressed in portions of the NB 7-4 lineage, appears to be required for the correct expression of en within this lineage (Cui and Doe, 1992). The mechanisms controlling differential gene activity in a NB throughout its lineage are unknown.

Neuroblast cell lineages

Cell lineage analysis in the grasshopper embryo shows that specific NBs go through invariant lineages to generate stereotyped family of neurons (Goodman et al., 1984; Raper et al., 1983; Doe and Goodman, 1985b; Bate and Grunewald, 1981; Goodman et al., 1982; Raper et al., 1983; Taghert and Goodman, 1984). The lineage studies described here suggests that this is also true in Drosophila. The 1-1 and 4-2 NBs are closely associated with an eve+ GMC. This correlation suggests that the 1-1 and 4-2 NBs generate eve+ GMCs, although direct lineage studies will be required to absolutely prove this relationship. These GMCs generate the aCC, pCC, RP2 and RP2sib neurons. The axon morphology, muscle target and certain gene expression profiles are known for the aCC and RP2 motoneurons. The lineage studies described here provide a starting point for linking many previously disparate events of neurogenesis: from formation and specification of an individual NB (e.g. NB 4-2), to lineal determination of NB progeny (e.g. GMC 4-2a), to the morphological differentiation of an identified neuron (e.g. RP2).

Future directions

The markers described here should be useful for investigating several areas of neurogenesis. (1) How is NB formation regulated? The ‘neurogenic’ and ‘proneural’ genes control NB formation (Campos-Ortega and Knust, 1990), yet it is not known if these genes affect all NBs identically. It may be that each mutation affects the formation of different subsets of NBs. (2) How are NBs uniquely specified? Due to the lack of markers for identified NBs, genes positionally expressed along the anterior-posterior or dorsal-ventral axes of the neurogenic region have yet to be tested for a role in NB specification. (3) What mechanisms control NB cell lineages? Genes expressed in subsets of NBs and their progeny, could be examined to determine relative expression patterns; overlaps would highlight potential regulatory interactions. For example, the ming gene is co-expressed with en in several NBs and GMCs, and indeed we find that ming is required for normal en CNS expression (Cui and Doe, 1992). (4) What is the complete cell lineage of identified NBs? Individual NBs could be labeled using a variety of techniques (e.g. transplantation, injection, photoactivation) and their neuronal or glial progeny identified. Lineage studies may reveal functional relationships between lineally related cells, such as neurotransmitter choice (Goodman et al., 1979; Taghert and Goodman, 1984). In addition, lineage studies will provide a foundation for interpreting experimental and genetic perturbations.

Many thanks to L. Rost, K. Schuske and M. P. Scott for assistance in generating the 1530 and 1912 lines, and for providing the 5953 line. Antibodies were generously supplied by N. H. Patel and C. S. Goodman, M. van den Heuvel and R. Nusse, J. Skeath and S. Carroll, and S. DiNardo and P. O’Farrell. A slide of embryos beautifully stained for hb was loaned by S. Macedonio (Zeiss technical representative) who received it from P. MacDonald and G. Struhl several years ago. Thanks to K. Doe for drawing Fig. 1; to Q. Chu-LaGraff for embryos stained for wg; to Steve Paddock and Sean Carroll for assistance with confocal microscopy; and to Volker Hartenstein and Xuan Cui for comments on the manuscript. This work was supported by the Searle Scholars Foundation, an NSF Presidential Young Investigator Award and the NIH (R01-27056).

Bastiani
,
M. J.
and
Goodman
,
C. S.
(
1986
).
Guidance of neuronal growth cones in the grasshopper embryo. III. Recognition of specific glial pathways
.
J. Neurosci
.
6
,
3542
3551
.
Bate
,
C. M.
(
1976
).
Embryogenesis of an insect nervous system I. A map of the thoracic and abdominal neuroblasts in Locusta migratoria
.
J. Embryol. Exp. Morph
.
35
,
107
123
.
Bate
,
C. M.
and
Grunewald
,
E. B.
(
1981
).
Embryogenesis of an insect nervous system. II. A second class of neuron precursor cells and the origin of the intersegmental connectives
.
J. Embryol. Exp. Morph
.
61
,
317
330
.
Bauer
,
V.
(
1904
).
Zur inneren Metamorphose des Centralnervensystems der Insecten
.
Zool. Jahrb. Abt. Anat. Ontog. Tiere
20
,
123
152
.
Cabrera
,
C. V.
(
1990
).
Lateral inhibition and cell fate during neurogenesis in Drosophila: the interactions between scute, Notch and Delta
.
Development
109
,
733
742
.
Campos-Ortega
,
J. A.
and
Hartenstein
,
V.
(
1985
).
The Embryonic Development of Drosophila melanogaster
.
Berlin
:
Springer-Verlag
.
Campos-Ortega
,
J. A.
and
Knust
,
E.
(
1990
).
Genetic mechanisms in early neurogenesis of Drosophilamelanogaster
.
Ann. Rev. Genet
.
24
,
387
407
.
Carroll
,
S. B.
and
Scott
,
M. P.
(
1985
).
Localization of the fushi tarazu protein during Drosophila embryogenesis
.
Cell
43
,
47
57
.
Cui
,
X.
and
Doe
,
C. Q.
(
1992
).
ming is expressed in neuroblast sublineages and regulates gene expression in the Drosophila central nervous system
.
Development
116
,
943
952
.
Dambly-Chaudiere
,
C.
,
Jamet
,
E.
,
Burri
,
M.
,
Bopp
,
D.
,
Basler
,
K.
,
Hafen
,
E.
,
Dumont
,
N.
,
Spielmann
,
P.
,
Ghysen
,
A.
and
Noll
,
M.
(
1992
).
The paired box gene pox neuro: a determinant of poly-innervated sense organs in Drosophila
.
Cell
69
,
159
172
.
Dick
,
T.
,
Yang
,
X.
,
Yeo
,
S.
and
Chia
,
W.
(
1991
).
Two closely linked Drosophila POU domain genes are expressed in neuroblasts and sensory elements
.
Proc. Natl. Acad. Sci. USA
88
,
7645
7649
.
DiNardo
,
S.
,
Kuner
,
J. M.
,
Theis
,
J.
and
O’Farrell
,
P. H.
(
1985
).
Development of embryonic pattern in Drosophila melanogaster as revealed by accumulation of the nuclear engrailed protein
.
Cell
43
,
59
69
.
Doe
,
C. Q.
and
Goodman
,
C. S.
(
1985a
).
Early events in insect neurogenesis. I. Development and segmental differences in the pattern of neuronal precursor cells
.
Dev. Biol
.
111
,
193
205
.
Doe
,
C. Q.
and
Goodman
,
C. S.
(
1985b
).
Early events in insect neurogenesis. II. The role of cell interactions and cell lineage in the determination of neuronal precursor cells
.
Dev. Biol
.
111
,
206
219
.
Doe
,
C. Q.
,
Hiromi
,
Y.
,
Gehring
,
W. J.
and
Goodman
,
C. S.
(
1988a
).
Expression and function of the segmentation gene fushi tarazu during Drosophila neurogenesis
.
Science
239
,
170
175
.
Doe
,
C. Q.
,
Smouse
,
D.
and
Goodman
,
C. S.
(
1988b
).
Control of neuronal fate by the Drosophila segmentation gene even-skipped
.
Nature
333
,
376
378
.
Doe
,
C. Q.
,
Chu-LaGraff
,
Q.
,
Wright
,
D. M.
and
Scott
,
M. P.
(
1991
).
The prospero gene specifies cell fates in the Drosophila central nervous system
.
Cell
65
,
451
464
.
Duffy
,
J. B.
,
Kania
,
M. A.
and
Gergen
,
J. P.
(
1991
).
Expression and function of the Drosophila gene runt in early stages of neural development
.
Development
113
,
1223
1230
.
Ferdieu
,
J. R.
and
Mahowald
,
A. P.
(
1989
).
Glial interactions with neurons during Drosophila embryogenesis
.
Development
106
,
749
757
.
Goodman
,
C. S.
,
O’Shea
,
M.
,
McCaman
,
R. E.
and
Spitzer
,
N. C.
(
1979
).
Embryonic development of identified neurons: Temporal pattern of morphological and biochemical differentiation
.
Science
204
,
219
222
.
Goodman
,
C. S.
,
Raper
,
J. A.
,
Ho
,
R. K.
and
Chang
,
S.
(
1982
).
Pathfinding of neuronal growth cones in grasshopper embryos
.
In Developmental Order: Its Origin and Regulation
. (eds.
S.
Subtelny
and
P. B.
Green
). pp.
275
316
.
New York
:
Alan R. Liss
.
Goodman
,
C. S.
,
Bastiani
,
M. J.
,
Doe
,
C. Q.
,
DuLac
,
S.
,
Helfand
,
S. L.
,
Kuwada
,
J. Y.
and
Thomas
,
J. B.
(
1984
).
Cell recognition during neuronal development
.
Science
225
,
1271
1279
.
Halpern
,
M. E.
,
Chiba
,
A.
,
Johansen
,
J.
and
Keshishian
,
H.
(
1991
).
Growth cone behavior underlying the development of stereotypic synaptic connections in Drosophila embryos
.
J. Neurosci
.
11
,
3227
3238
.
Hartenstein
,
V.
and
Campos-Ortega
,
J. A.
(
1984
).
Early neurogenesis in wild-type Drosophila melanogaster
.
Roux’s Arch. Dev. Biol
.
193
,
308
325
.
Hartenstein
,
V.
,
Rudloff
,
E.
and
Campos-Ortega
,
J. A.
(
1987
).
The pattern of proliferation of the neuroblasts in the wild-type embryo of Drosophilamelanogaster
.
Roux’s Arch. Dev. Biol
.
196
,
473
485
.
Hiromi
,
Y.
,
Kuroiwa
,
A.
and
Gehring
,
W. J.
(
1985
).
Control elements of the Drosophila segmentation gene fushi tarazu
.
Cell
43
,
603
613
.
Jacobs
,
J. R.
,
Hiromi
,
Y.
,
Patel
,
N. H.
and
Goodman
,
C. S.
(
1989
).
Lineage, migration and morphogenesis of longitudinal glia in the Drosophila CNS as revealed by a molecular lineage marker
.
Neuron
2
,
1625
1631
.
Jimenez
,
F.
and
Campos-Ortega
,
J. A.
(
1979
).
A region of the Drosophila genome necessary for CNS development
.
Nature
282
,
310
312
.
Jimenez
,
F.
and
Campos-Ortega
,
J. A.
(
1990
).
Defective neuroblast commitment in mutants of the achaete-scute complex and adjacent genes of D. melanogaster
.
Neuron
5
,
81
89
.
Kawamura
,
K.
and
Carlson
,
J. G.
(
1962
).
Studies on cytokinesis in neuroblasts of the grasshopper, Chortophaga viridifasciata (de geer). III. Factors determining the location of the cleavage furrow
.
Exp. Cell Res
.
26
.,
411
423
.
Klambt
,
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
.
Lehmann
,
R.
,
Jimenez
,
R.
,
Dietrich
,
V.
and
Campos-Ortega
,
J. A.
(
1983
).
On the phenotype and development of mutants of early neurogenesis in Drosophila melanogaster
.
Roux’s Arch. Dev. Biol
.
192
,
62
74
.
Martin-Bermudo
,
M. D.
,
Martinez
,
C.
,
Rodriguez
,
A.
and
Jimenez
,
F.
(
1991
).
Distribution and function of the lethal of scute gene product during early neurogenesis in Drosophila
.
Development
113
,
445
454
.
Mlodzik
,
M.
,
Hiromi
,
Y.
,
Weber
,
U.
,
Goodman
,
C. S.
and
Rubin
,
G. M.
(
1990
).
The Drosophilaseven-up gene, a member of the steroid receptor gene superfamily, controls photoreceptor cell fates
.
Cell
60
,
211
224
.
Patel
,
N. H.
,
Schafer
,
B.
,
Goodman
,
C. S.
and
Holmgren
,
R.
(
1989a
).
The role of segment polarity genes during Drosophila neurogenesis
.
Genes and Dev
.
3
,
890
904
.
Patel
,
N. H.
,
Martin-Blanco
,
E.
,
Coleman
,
K. G.
,
Poole
,
S. J.
,
Ellis
,
M. C.
,
Kornberg
,
T. B.
and
Goodman
,
C. S.
(
1989b
).
Expression of engrailed proteins in arthropods, annelids and chordates
.
Cell
58
,
955
968
.
Patel
,
N. H.
,
Ball
,
E. E.
and
Goodman
,
C. S.
(
1992
).
Changing role of even-skipped during the evolution of insect pattern formation
.
Nature
357
,
339
342
.
Raper
,
J. A.
,
Bastiani
,
M. J.
and
Goodman
,
C. S.
(
1983
).
Pathfinding by neuronal growth cones in grasshopper embryos I. Divergent choices made by growth cones of sibling neurons
.
J. Neurosci
.
3
,
20
30
.
Ready
,
D. F.
,
Hanson
,
T. E.
and
Benzer
,
S.
(
1976
).
Development of the Drosophila retina, a neurocrystalline lattice
.
Dev. Biol
.
53
,
217
240
.
Roberts
,
D. B.
(
1986
).
Drosophila, a Practical Approach
.
Eynsham, England
:
IRL Press Limited
.
Rubin
,
G. M.
(
1991
).
Signal transduction and the fate of the R7 photoreceptor in Drosophila
.
Trends Genet
.
7
,
372
377
.
Sink
,
H.
and
Whitington
,
P. M.
(
1990
).
Location and connectivity of abdominal motoneurons in the embryo and larva of Drosophila melanogaster
.
J. Neurobiol
.
12
,
298
311
.
Sink
,
H.
and
Whitington
,
P. M.
(
1991
).
Pathfinding in the central nervous system and periphery by identified embryonic Drosophila motor axons
.
Development
112
,
307
316
.
Skeath
,
J. B.
and
Carroll
S. B.
(
1992
).
Regulation of proneural gene expression and cell fate during neuroblast segregation in the Drosophila embryo
.
Development
114
,
939
946
.
Sternberg
,
P. W.
and
Horvitz
,
H. R.
(
1991
).
Signal transduction during C. elegans vulval induction
.
Trends Genet
.
7
,
366
371
.
Sulston
,
J. E.
,
Schierenberg
,
E.
,
White
,
J. G.
and
Thompson
,
J. N.
(
1983
).
The embryonic cell lineage of the nematode C. elegans
.
Dev. Biol
.
100
,
64
119
.
Taghert
,
P. H.
and
Goodman
,
C. S.
(
1984
).
Cell determination and differentiation of identified serotonin-containing neurons in the grasshopper embryo
.
J. Neurosci
.
4
,
989
1000
.
Thomas
,
J. B.
,
Bastiani
,
M. J.
,
Bate
,
C. M.
and
Goodman
,
C. S.
(
1984
).
From grasshopper to Drosophila: a common plan for neuronal development
.
Nature
310
,
203
207
.
van den Heuvel
,
M.
,
Nusse
,
R.
,
Johnston
,
P.
and
Lawrence
,
P. A.
(
1989
).
Distribution of the wingless gene product in Drosophila embryos: a protein involved in cell-cell communication
.
Cell
59
,
739
749
.
Vaessin
,
H.
,
Grell
,
E.
,
Wolff
,
E.
,
Bier
,
E.
,
Jan
,
L. Y.
and
Jan
,
Y. N.
(
1991
).
prospero is expressed in neuronal precursors and encodes a nuclear protein that is involved in the control of axonal outgrowth in Drosophila
.
Cell
67
,
941
953
.
Wheeler
,
W. M.
(
1891
).
Neuroblasts in the arthropod embryo
.
J. Morph
.
4
,
337
343
.
Wheeler
,
W. M.
(
1893
).
A contribution to insect embryology
.
J. Morph
.
8
,
1
160
.