Cell diversity in the Drosophila central nervous system (CNS) is primarily generated by the invariant lineage of neural precursors called neuroblasts. We used an enhancer trap screen to identify the ming gene, which is transiently expressed in a subset of neuroblasts at reproducible points in their cell lineage (i.e. in neuro- blast ‘sublineages’), suggesting that neuroblast identity can be altered during its cell lineage. ming encodes a predicted zinc finger protein and loss of ming function results in precise alterations in CNS gene expression, defects in axonogenesis and embryonic lethality. We propose that ming controls cell fate within neuroblast cell lineages.

The mechanisms controlling the generation of cellular diversity in the central nervous system (CNS) remain a fun- damental question in neurobiology. In Drosophila, neuronal and glial diversity in the CNS is generated in two distinct steps. First, individual cells within the ventral ectoderm delaminate into the embryo and enlarge to form a subepi- dermal, two-dimensional array of neural precursors, most of which are neuroblasts (NBs; Wheeler, 1891; Poulson, 1950). Each NB can be uniquely identified by its time of formation, position of enlargement, pattern of gene expression and/or the neuronal progeny that it generates (reviewed in Doe, 1992a).

The second step of neurogenesis generates most of the cellular diversity in the CNS. NBs repeatedly undergo an oriented, asymmetric division, generating a smaller gan- glion mother cell (GMC; Bauer, 1904), which is displaced into the embryo, and a new NB, which remains at the ven- tral cortex of the CNS. Each GMC divides to produce a characteristic pair of postmitotic neurons (Thomas et al., 1984; Goodman et al., 1984; Taghert and Goodman, 1984; Raper et al., 1983; Doe et al., 1988; Patel et al., 1989). Experimental evidence shows that NBs can produce a normal first GMC despite altered environmental cues in vivo (Doe and Goodman, 1985) or when isolated in vitro (Huff et al., 1989).

The fact that early NB lineages appear invariant in vivo and in vitro suggests that the information required to spec- ify GMC and neuronal identity is contained within the NB lineage. It has been proposed that a class of genes, called ‘NB identity’ genes, are expressed in NBs and their prog- eny and control the cell lineage of each NB (Doe et al., 1991; Doe, 1992a). Expression of these genes could be modulated qualitatively or quantitatively during the NB lin- eage, such that each GMC inherits a different amount of gene product or unique combination of gene products. Loss- of-function mutations would switch NB or GMC identity to a different or novel fate. Currently only a few NB iden- tity genes have been described. The runt gene is expressed in a subset of NBs, GMCs and neurons; it is required for the specification of several identified GMCs and neurons (Duffy et al., 1991). polyhomeotic is expressed in NBs and is required for correct gene expression in GMCs or neu- rons (Smouse et al., 1988; Decamillis et al., 1992). The prospero gene is transcribed in NBs and GMCs, but pro- tein can only be detected in GMCs, where it is required to control GMC gene expression and their subsequent differ- entiation (Doe et al., 1991; Vaessin et al., 1991; Matsuzaki et al., 1992). Thus prospero is best considered a ‘GMC identity’ gene (Doe, 1992a).

We have used the enhancer trap technique (O’Kane and Gehring, 1987) to identify additional NB identity genes; in particular, we looked for genes expressed in qualitatively or quantitatively different patterns during identified NB cell lineages. Here we describe the ming gene (‘fate’ in Chi- nese), which is transcribed in a subset of NBs and GMCs, but not in neurons. Certain neuroblasts express ming from the time of their formation, whereas other identified neu- roblasts express ming only after they have generated a stereotyped number of GMCs. ming encodes a predicted zinc finger protein and loss of ming function leads to precise alterations in CNS gene expression followed by defects in axonogenesis and embryonic lethality.

Genetic analysis

The P element construct of Bier et al. (1989) was used to generate over 2,000 independent enhancer trap lines (C. Q. D., L. Rost, K. Schuske and M. P. Scott, unpublished results), including the embryonic lethal ming1530and ming2092alleles. We used the trans- posase-producing D2-3 chromosome (Robertson et al., 1988) to mobilize the 2092 P element insertion with the goal of creating deletions of the ming gene. We scored for the loss of the w+ marker on the P element. Among the 82 w flies examined, 60 were homozygous lethal and 22 were homozygous viable. 34 of the lethal lines were examined by Southern blots and three chromo- somes (ming24, ming39 and ming78) had deletions of genomic DNA at the insertion site. Embryos homozygous for any of these three alleles do not produce ming transcript (about 25% of the embryos from ming/TM3 parents had no detectable ming RNA expression by whole-mount hybridization) and do not complement the ming1530 or ming2092 insertion alleles. To identify accurately homozygous ming embryos for phenotypic analysis, we balanced the ming24 allele over TM3 containing the ftzlacZ construct (Hiromi and Gehring, 1987); homozygous ming24 embryos can be easily distinguished due to lack of the ftzlacZ expression.

Molecular analysis

The genomic DNA fragments flanking the 1530 and 2092 inser- tions were cloned by the plasmid rescue technique (Bier et al.,1989) and the flanking DNA was localized to cytological posi- tion 83C in polytene chromosomes by the method of Engels et al. (1986). The 4.9 kb fragment of genomic DNA between the 1530 insert and the KspI site (Fig. 1A) was used to isolate 20 cDNA clones from a Drosophila 0-24 hour embryonic lZAPII cDNA library (Stratagene). A deletion series was made across the longest cDNA, P1 (Henikoff, 1984; Promega Erase-a-Base Kit), and both strands were sequenced. Deduced amino acid sequences were compared to the Genbank Database (Pearson and Lipman, 1988). Southern blots were done using standard protocols (Sambrook et al., 1989) to assay the genomic DNA of P element revertant lines.

Northern blots were done using total RNA from 0-12 hour embryos. Embryos were homogenized in lysis buffer containing 7 M Urea, 2% SDS, 0.35 M NaCl, 1 mM EDTA, 10 mM Tris pH 8.0. Following phenol-chloroform extraction, RNA was precipi- tated in the presence of 3 M sodium acetate at pH 5.0. 40 μg of total RNA was run on a 1% agarose-formaldehyde gel, transferred onto a nitrocellulose membrane and hybridized with 32P-labeled P1 cDNA. Hybridization was at 68°C for 1.5 hour in QuickHyb solution (Stratagene). The filters were washed twice in 1× SSC/0.1% SDS at room temperature and once in 0.1× SSC/0.1% SDS at 62°C before autoradiography.

Fig. 1.

Molecular features of the ming gene. (A) Restriction map of genomic DNA flanking the 1530 and 2092 P element insertions (triangles). B, BamI; E, EcoRI; G, BglII; H, HindIII; K, KspI; P, PstI and X, XbaI. (B) Part of the P1 cDNA is encoded by genomic DNA in the hatched region; no other embryonic transcript is encoded by the genomic DNA shown. (C) Southern blots reveal that the ming24 deletion removes 850 bp of DNA within the indicated region (open bar). (D) Northern blot using total RNA from 0-12 hours embryos shows that the ming gene encodes a transcript of about 3.3 kb. (E) DNA and predicted amino acid sequence of the P1 cDNA. Amino acid sequence is numbered 1-417 from the first methionine of the open reading frame. The first two methionines are double underlined; the second is associated with a consensus start sequence (Cavener, 1987). An opa repeat (Wharton et al., 1985) is observed between nucleotides 1093 and 1122. EMBL accession number is Z15091. (F) ming encodes a predicted zinc finger protein. Four consecutive repeats between amino acids 118 to 354 can be aligned with the consensus zinc finger motifs of TFIIIA class. Conserved metal-binding amino acids are highlighted in black; other conserved amino acids are highlighted in gray. y, hydrophobic amino acid; arrowheads, other possible metal-binding amino acids.

Fig. 1.

Molecular features of the ming gene. (A) Restriction map of genomic DNA flanking the 1530 and 2092 P element insertions (triangles). B, BamI; E, EcoRI; G, BglII; H, HindIII; K, KspI; P, PstI and X, XbaI. (B) Part of the P1 cDNA is encoded by genomic DNA in the hatched region; no other embryonic transcript is encoded by the genomic DNA shown. (C) Southern blots reveal that the ming24 deletion removes 850 bp of DNA within the indicated region (open bar). (D) Northern blot using total RNA from 0-12 hours embryos shows that the ming gene encodes a transcript of about 3.3 kb. (E) DNA and predicted amino acid sequence of the P1 cDNA. Amino acid sequence is numbered 1-417 from the first methionine of the open reading frame. The first two methionines are double underlined; the second is associated with a consensus start sequence (Cavener, 1987). An opa repeat (Wharton et al., 1985) is observed between nucleotides 1093 and 1122. EMBL accession number is Z15091. (F) ming encodes a predicted zinc finger protein. Four consecutive repeats between amino acids 118 to 354 can be aligned with the consensus zinc finger motifs of TFIIIA class. Conserved metal-binding amino acids are highlighted in black; other conserved amino acids are highlighted in gray. y, hydrophobic amino acid; arrowheads, other possible metal-binding amino acids.

mRNA and protein localization

Localization of ming transcript was examined by in situ hybrid- ization to whole embryos (Tautz and Pfeiffle, 1989). The P1 cDNA and the 4.5 kb genomic fragment from the 1530 insert to the XbaI site were used as nonradioactive probes. Embryos were processed identically to Doe et al. (1991). The following anti- bodies were used for protein localization: mouse anti-β-galactosi- dase monoclonal (1:1,000; Promega), mouse anti-engrailed mon- oclonal (1:1; Patel et al., 1989), mouse BP102 monoclonal (1:10; Klambt et al., 1991) and mouse anti-fasciclin III monoclonal (1:2; Patel et al., 1987). Immunocytochemistry was done by the pro- cedure of Doe et al. (1991), with the exception of fasciclin III staining which was done on nerve cord ‘fillets’ by the method of Thomas et al. (1984).

Isolation and molecular analysis of the ming gene

Two enhancer trap lines, 1530 and 2092, express the marker gene lacZ in an identical subset of NBs. Genomic DNA flanking each insertion was cloned (Fig. 1A) and used to probe whole-mount embryos for RNA expression patterns. The 4.9 kb genomic DNA fragment extending from the 1530 insertion to the KspI site (Fig. 1A) detects a mRNA pattern indistinguishable from the 1530 lacZ pattern (see below) and was used to isolate a 1.3 kb cDNA, called P1 (Fig. 1B). The P1 cDNA hybridizes to mRNA matching the 1530 lacZ pattern in whole-mount embryos (data not shown) and detects a single band of 3.3 kb on northern blots (Fig. 1D). We call this transcription unit the ming gene.

The ming P1 cDNA contains a predicted open reading frame of 417 amino acids that extends to the end of the cDNA (Fig. 1E). Within this open reading frame are four repeated motifs (Fig. 1F), with part of each repeat show- ing homology to the TFIIIA-class zinc finger motif. The other portion of each repeat contains two cysteines and two histidines, which may form a metal-binding domain. The presence of four predicted zinc finger motifs in the ming P1 cDNA suggests that the ming protein could function to regulate gene expression in the CNS.

ming is specifically expressed in a subset of CNS neuroblasts

ming is expressed only in NBs, GMCs and midline of the CNS; it is not detectable in the peripheral nervous system or any other embryonic tissue. The ming transcript and the 1530 enhancer trap lacZ gene are expressed in virtually identical patterns (compare Fig. 2A, B, C, G [lacZ] to Fig. 2D, E, F, H [transcript]), although the 1530 line has a low level of lacZ expression in segmental stripes in the dorsal epidermis (which is a common artifact of this and possibly other enhancer trap constructs; C. Q. D. et al., unpublished results; Ghysen and O’Kane, 1989). ming transcripts are first detected in a subset of the midline CNS cells at stage 9 (data not shown; staging according to Campos-Ortega and Hartenstein, 1985). Slightly later, one NB in every hemiseg- ment (NB 6-1) expresses ming (Fig. 2D). By late stage 11, ming transcripts are expressed in 17 identified NBs in every hemisegment of the CNS (Fig. 2E). By stage 14, ming tran- script levels begin to decrease and, by stage 16, the tran- script is undetectable except in a few cells in the thoracic and terminal abdominal CNS (Fig. 2F). β-galactosidase (β- gal) protein in 1530 embryos persists several hours later than the ming transcript can be detected, presumably due to the greater stability of the βgal protein (Fig. 2C). ming expression is only detectable in NBs and GMCs, not in the neurogenic ectoderm or in neurons (Fig. 2H).

Fig. 2.

ming is specifically expressed in a subset of embryonic NBs and GMCs. Detection of βgal protein in the 1530 enhancer trap line (A-C, G) and ming mRNA (D-F, H). (A and D) Early stage 10 embryos (early S3 NB array). Only NB 6-1 (arrowhead) and a few midline cells (arrow) express ming. (B and E) Late stage 11 embryos (S5 NB array). The expression of βgal in the 1530 line and ming mRNA is virtually identical; for example the indicated row 3 NBs (3-4, 3-3, 3-2 and 3-1, small arrowheads from left to right) and NB 7-4 (arrow). NB 5-6 expresses mRNA at this stage (E; large arrowhead), whereas the βgal protein can be detected a short time later in the 1530 line (see Fig. 3). (C and F) Stage 15 embryos. (C) βgal expression persists in most NBs in the 1530 line, probably due to the stability of the protein. (F) In contrast, ming transcript can only be detected in a few cells of the thoracic (wide arrowhead) and terminal CNS (arrow). (G and H) Stage 11 embryos. ming expression is only in NBs (small arrow) and some GMCs (arrowhead), not in ventral ectodermal cells (large arrow) or neurons (dorsal-most cells). (A-F) Ventral view, anterior is up. White triangle indicates the ventral midline. (G) Cross- section view, ventral side down. (H) Lateral view, anterior is left and ventral side down.

Fig. 2.

ming is specifically expressed in a subset of embryonic NBs and GMCs. Detection of βgal protein in the 1530 enhancer trap line (A-C, G) and ming mRNA (D-F, H). (A and D) Early stage 10 embryos (early S3 NB array). Only NB 6-1 (arrowhead) and a few midline cells (arrow) express ming. (B and E) Late stage 11 embryos (S5 NB array). The expression of βgal in the 1530 line and ming mRNA is virtually identical; for example the indicated row 3 NBs (3-4, 3-3, 3-2 and 3-1, small arrowheads from left to right) and NB 7-4 (arrow). NB 5-6 expresses mRNA at this stage (E; large arrowhead), whereas the βgal protein can be detected a short time later in the 1530 line (see Fig. 3). (C and F) Stage 15 embryos. (C) βgal expression persists in most NBs in the 1530 line, probably due to the stability of the protein. (F) In contrast, ming transcript can only be detected in a few cells of the thoracic (wide arrowhead) and terminal CNS (arrow). (G and H) Stage 11 embryos. ming expression is only in NBs (small arrow) and some GMCs (arrowhead), not in ventral ectodermal cells (large arrow) or neurons (dorsal-most cells). (A-F) Ventral view, anterior is up. White triangle indicates the ventral midline. (G) Cross- section view, ventral side down. (H) Lateral view, anterior is left and ventral side down.

ming is expressed at different points during the lineage of identified neuroblasts

The initiation of ming expression in identified NBs was determined using the 1530 line and was confirmed for the ming transcript in several NBs (Figs 2, 3). Some NBs first express ming at the time they enlarge, whereas other NBs express ming only after they have begun their cell lineage. For example, NB 6-1 expresses ming at the time of its for- mation and maintains expression until at least late stage 11 (Fig. 2D,E). In contrast, NB 1-1 forms at stage 9, but ming is not expressed until late stage 10 (Figs 2, 3), after NB 1- 1 has generated at least one GMC (Doe, 1992b). It has been shown that the earliest developing NBs (the S1 NBs) pro- duce their first GMC 10-20 minutes after they enlarge, and that their subsequent cell cycle is 40-50 minutes (Harten-stein et al., 1987). Based on these data, we predict that many of the S1 NBs (e.g. NBs 3-2, 5-2, 5-3, 7-1 and 7-4) only express ming transcript after generating one or more GMCs (Figs 2B,E, 3).

Fig. 3.

Timing of ming expression in identified NBs. Expression of en protein (green) and pgal (ming pattern; red) in 1530 embryos. Co-expression is shown in yellow. (A, B) SI and ming is not expressed, although several NBs have formed that will subsequently express ming (e.g. 1-1 and 7-4). (C) Early S3 neural precursors (early stage I0). ming is first expressed in 6- 1, which also expresses en (see also Fig. 4A). (D) S3 neural precursors (late stage I0). Although no new NBs have formed, ming is now expressed in five NBs. Based on data from Hartenstein et al. (1987) and Doe (1992b), several of these NBs are likely to have generated progeny prior to expressing ming (e.g. I I, 5-2 and 7-4). (E) S4 neural precursors (early stage 11). ming is expressed in 9 NBs, including 5 of the 8 NBs that express en. stage 11). ming is expressed in 9 more NBs (total of 18), including an SI NB (3-2) and an S2 NB (2- 3) which may have generated up to 5 GMCs prior to expressing ming (see text). Ventral view, anterior is up; Dotted line, ventral midline; asterisk, tracheal pit. Nomenclature from Doe (1992b).

Fig. 3.

Timing of ming expression in identified NBs. Expression of en protein (green) and pgal (ming pattern; red) in 1530 embryos. Co-expression is shown in yellow. (A, B) SI and ming is not expressed, although several NBs have formed that will subsequently express ming (e.g. 1-1 and 7-4). (C) Early S3 neural precursors (early stage I0). ming is first expressed in 6- 1, which also expresses en (see also Fig. 4A). (D) S3 neural precursors (late stage I0). Although no new NBs have formed, ming is now expressed in five NBs. Based on data from Hartenstein et al. (1987) and Doe (1992b), several of these NBs are likely to have generated progeny prior to expressing ming (e.g. I I, 5-2 and 7-4). (E) S4 neural precursors (early stage 11). ming is expressed in 9 NBs, including 5 of the 8 NBs that express en. stage 11). ming is expressed in 9 more NBs (total of 18), including an SI NB (3-2) and an S2 NB (2- 3) which may have generated up to 5 GMCs prior to expressing ming (see text). Ventral view, anterior is up; Dotted line, ventral midline; asterisk, tracheal pit. Nomenclature from Doe (1992b).

Generation of mutations at the ming locus

The enhancer trap insertions 1530 and 2092 are located within 200 bp of each other at cytological position 83C on the third chromosome (Fig. 1A). Both insertions are homozygous embryonic lethal and do not complement each other; neither insertion mutation noticeably affects the dis- tribution or abundance of the embryonic ming transcript.

To obtain mutations that abolish expression of the ming transcript, the P elements in the 1530 and 2092 lines were mobilized using the transposase-producing D2-3 chromo- some (Robertson et al., 1988). We generated both homozy- gous viable and homozygous lethal chromosomes lacking the w+ P element marker; the viable chromosomes show that lethality of the original enhancer trap chromosome is due to the enhancer trap insertion. Southern blots show that three 2092 revertant chromosomes lack some of the genomic DNA flanking the P element insertion (data not shown). These three excision mutations are embryonic lethal, do not complement either insertion mutation and do not express detectable ming transcript by whole-mount hybridization. We refer to these P element excision muta- tions as ming24, ming39 and ming78, and the insertion muta- tions as ming1530 and ming2092. The smallest of the defi- ciencies, ming24, removes 850 bp of DNA at the site of the enhancer trap insertion (Fig. 1C). Because ming24 abolishes expression of the ming transcript without removing coding sequences for other embryonically expressed genes (Fig. 1), we used embryos homozygous for ming24 (called ‘ming embryos’ below) to characterize the loss-of-function ming CNS phenotype.

ming is required for the correct CNS expression of the engrailed gene

The NB expression of ming (as shown by βgal localization in 1530 embryos) has been compared to seven other genes expressed in subsets of NBs (Doe, 1992b), with the hope of identifying potential ‘downstream’ targets of the ming gene. Substantial overlap is observed between expression of ming and engrailed (en) (Figs 3, 4). At early stage 10, ming is expressed in several midline cells and the single NB 6-1; the NB and most of the midline cells also express en (Figs 3C, 4A). By stage 11, ming and en are co-expressed in six NBs and a subset of midline CNS cells (Figs 3F, 4B). There is also overlap between ming and en in a number of GMCs (data not shown), but we have not yet uniquely iden- tified these cells. Due to extensive co-expression of ming and en in NBs and GMCs, we looked in ming embryos for defects in en CNS expression.

Fig. 4.

ming and en are co-expressed in neural precursors. CNS expression of ming (as detected by pgal expression in 1530 embryos; red) and en (as detected with an antibody to en protein; green) using confocal microscopy. (A) Early stage IO embryo showing co-expression in NB 6-1 and a subset of rnidline CNS cells (yellow; arrow). NBs expressing only en (green; small arrowhead) and midline cells expressing only ming (reel: large arrowhead) can also be observed. (B) Late stage 11 embryo showing a subset of the NBs expressing ming and en (other NBs are out of the focal plane). ming and en co-expression can be observed in NB 7-4 (yellow; arrow). en expression in NB 7-3 (green; small arrowhead) and ming expression in NBs 3-3 and 3-4 (reel; large arrowheads) can be observed. Ventral view, anterior is up. White triangle indicates the ventral miclline

Fig. 4.

ming and en are co-expressed in neural precursors. CNS expression of ming (as detected by pgal expression in 1530 embryos; red) and en (as detected with an antibody to en protein; green) using confocal microscopy. (A) Early stage IO embryo showing co-expression in NB 6-1 and a subset of rnidline CNS cells (yellow; arrow). NBs expressing only en (green; small arrowhead) and midline cells expressing only ming (reel: large arrowhead) can also be observed. (B) Late stage 11 embryo showing a subset of the NBs expressing ming and en (other NBs are out of the focal plane). ming and en co-expression can be observed in NB 7-4 (yellow; arrow). en expression in NB 7-3 (green; small arrowhead) and ming expression in NBs 3-3 and 3-4 (reel; large arrowheads) can be observed. Ventral view, anterior is up. White triangle indicates the ventral miclline

Between stage 11 and stage 13, en is expressed in many GMCs and neurons derived from the two rows of en- expressing (en+) NBs (Fig. 3). During stage 14 and 15, the en protein pattern ‘sharpens up’ into several discrete clus- ters of cells, which are shown in Fig. 5 and schematized in Fig. 6. Within each segment of the CNS, en protein is detectable in three large ventral midline cells (VUMs; Patel et al., 1989), a cluster of dorsal medial (DM) cells and two pair of small non-neuronal median support (MS) cells just posterior to the median NB (Fig. 5D,E). In addition, there are four bilateral clusters of en+ cells in each segment: a posterior intermediate (PI) group of 4-6 cells (Fig. 5C), a posterior lateral (PL) group of 8-10 cells (Fig. 5C) and two pair of neurons located near the ventral surface of the ante- rior region of the segment, the NH cells (Fig. 5A,B).

Fig. 5.

ming is required for correct engrailed CNS expression. CNS expression of engrailed protein in stage 15 or 16 wild-type embryos (A-E) and stage 16 homozygous ming24 embryos (F-J). (A and F) Overall pattern of the en expression. Two pairs of NH neurons (arrowheads) and one pair of MS cells (arrow) can be seen in wild-type embryos, but not in ming embryos. (B and G) The en+ NH neurons (arrowhead) are detectable in wild-type embryos, but not in ming embryos. (C and H) The en+ PL group (wide arrowhead) has about twice the number of cells in the ming mutants; the size of each PL cell is reduced in ming embryos. Focal plane in H is slightly more dorsal than in C. (D and E) Wild-type embryos have two pair of en+ MS cells (arrows); (I and J) these cells are not detectable in ming embryos. (A-D and F-I) ventral view, anterior up; (E and J) lateral view, anterior is up and ventral surface is to the left.

Fig. 5.

ming is required for correct engrailed CNS expression. CNS expression of engrailed protein in stage 15 or 16 wild-type embryos (A-E) and stage 16 homozygous ming24 embryos (F-J). (A and F) Overall pattern of the en expression. Two pairs of NH neurons (arrowheads) and one pair of MS cells (arrow) can be seen in wild-type embryos, but not in ming embryos. (B and G) The en+ NH neurons (arrowhead) are detectable in wild-type embryos, but not in ming embryos. (C and H) The en+ PL group (wide arrowhead) has about twice the number of cells in the ming mutants; the size of each PL cell is reduced in ming embryos. Focal plane in H is slightly more dorsal than in C. (D and E) Wild-type embryos have two pair of en+ MS cells (arrows); (I and J) these cells are not detectable in ming embryos. (A-D and F-I) ventral view, anterior up; (E and J) lateral view, anterior is up and ventral surface is to the left.

Fig. 6.

Summary of engrailed and fasciclin III expression pattern in wild-type and ming embryos. Schematic drawing based on camera-lucida tracings of engrailed (en) and fasciclin III (fasIII)-stained embryos. ming embryos lack NH and MS cells, and have twice the normal number of PL cells; there is a loss of one fasciclin III-positive fascicle in the posterior commissure. See text for details. Dorsal view, anterior is up; vertical lines, nerve cord boundaries; dotted lines, outline of longitudinal and commissural axons.

Fig. 6.

Summary of engrailed and fasciclin III expression pattern in wild-type and ming embryos. Schematic drawing based on camera-lucida tracings of engrailed (en) and fasciclin III (fasIII)-stained embryos. ming embryos lack NH and MS cells, and have twice the normal number of PL cells; there is a loss of one fasciclin III-positive fascicle in the posterior commissure. See text for details. Dorsal view, anterior is up; vertical lines, nerve cord boundaries; dotted lines, outline of longitudinal and commissural axons.

In ming embryos, epidermal and NB en protein patterns are normal (data not shown). Expression of en in GMCs is roughly normal but, due to the difficulty of identifying indi- vidual GMCs, we may be missing subtle defects in en expression in these cells. By stage 16, however, there are clear alterations in en CNS expression. en protein cannot be detected in the NH or MS cells (Figs 5F,G,I, 6). Because these cells can only be unambiguously detected at stage 16, we do not know when the defects first appear or whether these cells derive from ming-expressing precursors. In con- trast, the number of en+ cells in the PL cluster increases from the 8-10 in wild-type embryos to 20-25 in ming embryos (Fig. 5C,H). Moreover, in ming embryos the size of the PL cells is abnormal, with each cell about half its normal size (Fig. 5H). Three independent ming transcript- negative mutants all show the same altered en expression pattern.

Interestingly, embryos heterozygous for a ming insertion mutation (ming2092) and a ming transcript-negative muta- tion (ming24 or ming39) have a pattern of en expression intermediate in severity between homozygous insertion chromosome (identical to wild type) and homozygous null (described above). The intermediate phenotype of the ming2092/ming24 embryos is a complete loss of en+ MS cells, a reduction in the number of en+ NH cells and a slight increase in the number of en+ PL neurons (data not shown). This shows that the 2092 insertion is a weak ming allele, still homozygous lethal despite having no detectable alter- ations in ming transcript pattern or en CNS expression.

ming is required for the formation of an identified commissural fascicle

Altered GMC or neuronal cell fate is expected to affect later events during CNS development, including elaboration of the axonal scaffold. In fact, ming embryos show subtle, but clear, defects in the pattern of CNS axons. The wild-type CNS has two longitudinal connectives and a segmentally repeated pair of commissures (anterior and posterior), which can be detected with the BP102 monoclonal anti- body (Fig. 7A; Klambt et al., 1991). In ming embryos, the posterior commissure is thinner than normal, with the aver- age diameter decreased almost to half (Fig. 7B). To define further the posterior commissure defects in ming embryos, anti-fasciclin III monoclonal antibody was used to detect a subset of fascicles in the posterior commissure (Patel et al, 1987). The fasciclin III pattern is diagrammed in Fig. 6. In stage-16 wild-type embryos, fasciclin III is expressed in several clusters of cells as well as in three axonal fascicles in the anterior commissure and two fascicles in the pos- terior commissure (Fig. 7C). In ming embryos, fasciclin III expression in the cell bodies and anterior commissure is normal; however, one fasciclin III-positive fascicle is miss- ing from the posterior commissure (Fig. 7D). It is not known whether fasciclin III gene expression is lost in some neurons, whether these neurons express fasciclin III but grow in another fasciclin III-positive axon bundle, or whether the neurons die.

Fig. 7.

ming is required to establish the normal CNS axonal pattern. (A,B) CNS axons detected with the BP102 antibody in stage 16 embryos. (A) In wild- type embryos, the anterior and posterior commissures (ac, pc) have the same diameter; (B) in ming embryos the posterior commissure has a reduced number of axons. (C and D) There are fewer fasciclin III-positive fascicles in ming embryos. Embryos were double labeled with anti-engrailed antibody to identify the ming embryos. (C) Wild-type embryos (note the en+ MS cells; thin arrow) have two fasciclin III-positive fascicles in the posterior commissure (arrowheads); the RP1 neurons (wide arrow) are indicated. (D) ming embryos (note lack of en+ dMS cells; thin arrow) have only one fasciclin III-positive fascicle in the posterior commissure (arrowhead); the RP1 neurons (wide arrow) are indicated. Ventral view, anterior is up.

Fig. 7.

ming is required to establish the normal CNS axonal pattern. (A,B) CNS axons detected with the BP102 antibody in stage 16 embryos. (A) In wild- type embryos, the anterior and posterior commissures (ac, pc) have the same diameter; (B) in ming embryos the posterior commissure has a reduced number of axons. (C and D) There are fewer fasciclin III-positive fascicles in ming embryos. Embryos were double labeled with anti-engrailed antibody to identify the ming embryos. (C) Wild-type embryos (note the en+ MS cells; thin arrow) have two fasciclin III-positive fascicles in the posterior commissure (arrowheads); the RP1 neurons (wide arrow) are indicated. (D) ming embryos (note lack of en+ dMS cells; thin arrow) have only one fasciclin III-positive fascicle in the posterior commissure (arrowhead); the RP1 neurons (wide arrow) are indicated. Ventral view, anterior is up.

Do neuroblasts change fate with every cell cycle?

NBs of the Drosophila CNS appear positionally and mor- phologically unaltered as they divide to produce GMCs, yet the fact that most early-born GMCs in a lineage exhibit a unique fate suggests that NB identity may change with each cell cycle. This hypothesis is supported by our finding that ming expression is altered at precise points during the lin- eage of identified NBs. We have determined the time of birth for the first GMC from NB 1-1 and NB 7-4 (Doe, 1992b). Combining these data with the timing of ming expression (Fig. 3), we observe that ming transcription is initiated in NBs 1-1 and 7-4 after GMC-1 is born, but prior to the generation of GMC-2. We predict that ming is sim- ilarly expressed at specific, reproducible points during the lineage of many different NBs. For example, the early- forming NB 3-2 always expresses ming about 2.5 hours after its formation, suggesting that it has divided many times prior to triggering ming transcription. Thus ming expression can be reproducibly initiated at different points during the lineage of identified NBs. Two conclusions can be drawn from these results. First, NB identity may change during the course of its cell lineage (as reflected by altered ming expression). Second, GMCs born at different times during a NB lineage may differentially inherit ming gene product.

The change in ming expression within NB lineages is extremely reproducible, but the function of differential ming expression within a NB lineage is not clear. ming is expressed in the latter portion of the NB 7-4 lineage and, when gene expression is abolished throughout the lineage, there are defects in the en-expressing cells that are likely progeny of this NB (they are located dorsal to NB 7-4). Whether ‘ectopic’ expression of ming in the initial portion of the 7-4 lineage (normally ming-negative) can generate a phenotype remains to be seen. Nevertheless, it is attractive to think that differential gene expression in a NB lineage might play a role in specifying GMC identity. The mech- anism controlling differential gene activity in NB lineages is unknown. Timing of gene expression has been proposed to be regulated by the number of rounds of DNA synthe- sis (e.g. Mita-Miyazawa et al., 1985), the nuclear:cytoplas- mic ratio (e.g. Newport and Kirschner, 1982a,b; Edgar et al., 1986), the stability of regulatory factors (discussed in Nasmyth et al., 1990), or a combination of these mecha- nisms (e.g. Edgar and McGhee, 1988). The ming transcript and 1530 line provide excellent markers to address the question of both initial NB specification as well as how NB identity is modified during its cell lineage.

ming encodes a predicted zinc finger protein

Conceptual translation of the ming cDNA reveals a repet- itive zinc finger motif which shares homology with the TFIIIA class proteins. The TFIIIA superfamily proteins reg- ulate gene activity primarily by binding target DNA through the zinc finger domains (Berg, 1990). It also has been shown that some of the TFIIIA class proteins can bind both DNA and RNA (Pelham and Brown, 1980), and some bind exclusively to RNA (Joho et al., 1990). In addition to the TFIIIA consensus sequences, there is another domain in each repeat with two cysteines and two histidines, which may represent a novel metal-binding motif. Although the predicted ming protein has strong homology to TFIIIA class zinc finger domains, it is not clear whether ming protein would be more likely to bind DNA or RNA. In either case, the predicted protein structure is consistent with a role for the ming protein in regulating gene activity.

ming is required for cell specification in the CNS

Loss of ming function results in abnormal en CNS expression: some neurons and glia lose en expression (e.g. the NH and MS cells), whereas elsewhere additional en+ cells are detected (e.g. PL cells). The exact cell lineage of the en+ NH, MS and PL neurons is not known. The clos- est en+ NB to the PL neurons is NB 7-4, which suggests that it is their likely precursor. NB 7-4 expresses ming during the latter portion of its embryonic cell lineage. Loss of ming expression could alter the NB 7-4 lineage in a number of ways. It may be that the GMCs or neurons are transformed to a novel cell fate, with a subsequent change in their pattern of division resulting in more PL cells of a smaller size. Alternatively, the later-born GMCs in the 7-4 lineage normally may not express en, but are transformed towards an earlier (en+) GMC fate in ming embryos. Another possibility is that the NB 7-4 lineage may be extended in ming embryos. Asymmetric NB divisions tend to decrease the volume of the NB (Hartenstein et al., 1987), which may lead to the observed decrease in size of the PL neurons in ming embryos. Finally, a combination of these effects may occur, such as the transformation of all GMCs towards the fate of the first (ming-negative) GMC, which may lead to the extension of the NB lineage (if the NB con- tinues to divide because it never produces the ‘last’ GMC in the lineage). Cell lineage studies of NBs in wild-type and ming embryos will be required to distinguish among these alternatives.

Many cells along the posterior midline of the CNS co- express ming and en (Fig. 3). If the MS cells develop in situ, they almost certainly co-express ming and en; how- ever, we cannot rule out the possibility that the MS cells migrate to their final position, as has been observed for mid- line glia and longitudinal glia (Klambt and Goodman, 1991; Jacobs et al., 1989). The NH cells are located near the ventral side of the CNS, anterior to the en+ NBs and neurons of the same segment. These neurons could develop in situ from en-negative NBs or they may be generated by en+ NBs and migrate anteriorly. Again, cell lineage studies will be required to determine the parental NBs of the NH cells, whether they express ming, and the fate of the NH and MS cells in ming embryos. Regardless of the mechanism, it is clear that ming is required for the normal specification of cell fate in the CNS, as detected by altered en neuronal and glial expression.

The ming gene is not transcribed in neurons, yet we observe defects in the axonal architecture of the CNS. There are fewer axons in the posterior commissure and an iden- tified fasciclin III-positive fascicle is missing. Due to the early expression of ming transcript (in NBs and GMCs, but not neurons) and the nature of the product (predicted zinc finger protein), we feel this phenotype is a secondary effect due to altered NB or GMC fates. A similar phenotype has been observed in prospero mutant embryos (Doe et al., 1991). The nuclear prospero protein can only be detected in GMCs and a subset of glia within the CNS (Vaessin et al., 1991; Matsuzaki et al., 1992), where it regulates gene expression in GMCs (Doe et al., 1991; Vaessin et al., 1991). However, prospero mutants show dramatic alterations in axon morphology, either due to prior defects in GMC or glial specification (Doe et al., 1991) or general defects in neuronal differentiation (Vaessin et al., 1991).

We have proposed that GMC identity is specified, in part, by the unique combination of gene products inherited from the parental NB (Doe, 1992a). Some genes might be expressed in specific cell types (e.g. the prospero protein is expressed in GMCs only; Fig. 8A). Other genes might be expressed in the entire cell lineage of a subset of NBs (Fig. 8B). Finally, many genes required for generating cell diver- sity in the CNS are likely to be expressed in ‘sublineages’ within a NB cell lineage (Fig. 8C). These genes are likely to contribute to the specification of both NB and GMCs during an individual NB cell lineage. The ming gene belongs in this class; it is reproducibly expressed in specific sublineages of identified NBs. Based on the ming expression pattern (NB sublineages), predicted protein (zinc finger motif) and mutant phenotype (altered CNS gene expression), we suggest that ming is a NB identity gene that controls cell fate during NB cell lineages in the embryonic CNS.

Fig. 8.

Differential gene expression during NB cell lineages. Two NB lineages (1, 2) are shown in each panel. (A) Cell type-specific expression could be specifically in NBs, GMCs, or neurons; here prospero protein expression is shown in GMCs but not NBs or neurons. (B) Lineage-specific expression in the entire lineage of a subset of NBs; this type of expression may control NB-specific aspects of neural development. (C) Sublineage-specific expression in NBs during stereotyped portions of their lineages. ming may be reproducibly expressed at different points in the lineages of NBs 1-1, 2-3, 2-4, 3-1, 3-2, 4-1, 5-2, 5-3, 7-1, 7-2 and 7-4. Large circle, NB; medium circle, GMC; small circle, neuron. Shading indicates mRNA or protein expression.

Fig. 8.

Differential gene expression during NB cell lineages. Two NB lineages (1, 2) are shown in each panel. (A) Cell type-specific expression could be specifically in NBs, GMCs, or neurons; here prospero protein expression is shown in GMCs but not NBs or neurons. (B) Lineage-specific expression in the entire lineage of a subset of NBs; this type of expression may control NB-specific aspects of neural development. (C) Sublineage-specific expression in NBs during stereotyped portions of their lineages. ming may be reproducibly expressed at different points in the lineages of NBs 1-1, 2-3, 2-4, 3-1, 3-2, 4-1, 5-2, 5-3, 7-1, 7-2 and 7-4. Large circle, NB; medium circle, GMC; small circle, neuron. Shading indicates mRNA or protein expression.

We thank N. H. Patel and C. S. Goodman for antibodies, Tim Karr for assistance with confocal microscopy, L. Rost, K. Schuske and M. P. Scott for assistance in generating the 1530 and 2092 lines, and Eric Spana and Ron Blackman for comments on the manuscript. This work was supported by the Searle Scholars Foun- dation, an NSF Presidential Young Investigator Award and the NIH (R01-27056).

Bauer
,
V.
(
1904
).
Zur inneren Metamorphose des Centralnervensystems der Insecten
.
Zool. Jahrb. Abt. Anat. Ontog. Tiere
20
,
123
152
.
Berg
,
J.
(
1990
).
Zinc finger domains: hypotheses and current knowledge
.
Annu. Rev. Biophys. Biophys. Chem
.
19
,
405
21
.
Bier
,
E.
,
Vaessin
,
H.
,
Shepherd
,
S.
,
Lee
,
K.
,
McCall
,
K.
,
Barbel
,
S.
,
Ackerman
,
L.
,
Carretto
,
R.
,
Uemura
,
T.
,
Grell
,
E.
,
Jan
,
L. Y.
and
Jan
,
Y. N.
(
1989
).
Searching for pattern and mutation in the Drosophila genome with a P-lacZ vector
.
Genes Dev
.
3
,
1273
1287
.
Campos-Ortega
,
J. A.
and
Hartenstein
,
V.
(
1985
).
The Embryonic Development of Drosophila melanogaster
.
Berlin
:
Springer-Verlag
.
Cavener
,
D. R.
(
1987
).
Comparison of the consensus sequence flanking translational start sites in Drosophila and vertebrates
.
Nucleic Acids Res
.
15
,
1353
1361
.
Decamillis
,
M.
,
Cheng
,
N.
,
Pierre
,
D.
and
Brock
,
H. W.
(
1992
).
The polyhomeotic gene of Drosophila encodes a chromatin protein that shares polytene chromosome-binding sites with Polycomb
.
Genes Dev
.
6
,
223
232
.
Doe
,
C. Q.
and
Goodman
,
C. S.
(
1985
).
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.
,
Smouse
,
D.
and
Goodman
,
C. S.
(
1988
).
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
.
Doe
,
C. Q.
(
1992a
).
The generation of neuronal diversity in the Drosophila embryonic central nervous system
.
In Determinants of Neuronal Identity
. (ed.
M.
Shankland
and
E.
Macagno
). pp.
119
154
.
New York
:
Academic Press
.
Doe
,
C. Q.
(
1992b
).
Molecular markers for identified neuroblasts and ganglion mother cells in the Drosophila central nervous system
.
Development
116
,
855
863
.
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
.
Edgar
,
B. A.
,
Kiehle
,
C. P.
and
Schubiger
,
G.
(
1986
).
Cell cycle control by the nucleo-cytoplasmic ratio in early Drosophila development
.
Cell
44
,
365
372
.
Edgar
,
L. G.
and
McGhee
,
J. D.
(
1988
).
DNA synthesis and the control of embryonic gene expression in C. elegans
.
Cell
53
,
589
599
.
Engels
,
W. R.
,
Preston
,
C. R.
,
Thompson
,
P.
and
Eggleston
,
W. B.
(
1986
).
In situ hybridization to Drosophila salivary chromosomes with biotinylated DNA probes and alkaline phosphatase
.
Focus
8
,
6
8
.
Ghysen
,
A.
and
O’Kane
,
C.
(
1989
).
Neural enhancer-like elements as specific cell markers in Drosophila
.
Development
105
,
35
52
.
Goodman
,
C. S.
,
Bastiani
,
M. J.
,
Doe
,
C. Q.
,
DuLac
,
L. S.
,
Helfand
,
S. L.
,
Kuwada
,
J. Y.
and
Thomas
,
J. B.
(
1984
).
Cell recognition during neuronal development
.
Science
225
,
1271
1279
.
Hartenstein
,
V.
,
Rudloff
,
E.
and
Campos-Ortega
,
J. A.
(
1987
).
The pattern of proliferation of the neuroblasts in the wild-type embryo of Drosophila melanogaster
.
Roux’s Arch. Dev. Biol
.
196
,
473
485
.
Henikoff
,
S.
(
1984
).
Unidirectional digestion with exonuclease III creates targeted breakpoints for DNA sequencing
.
Gene
28
,
351
359
.
Hiromi
,
Y.
and
Gehring
,
W. J.
(
1987
).
Regulation and function of the Drosophila segmentation gene fushi tarazu
.
Cell
50
,
963
74
.
Huff
,
R.
,
Furst
,
A.
and
Mahowald
,
A. P.
(
1989
).
Drosophila embryonic neuroblasts in culture: autonomous differentiation of specific neurotransmitters
.
Dev. Biol
.
134
,
146
157
.
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
.
Joho
,
K. E.
,
Darby
,
M. K.
,
Crawford
,
E. T.
and
Brown
,
D. D.
(
1990
).
A finger protein structurally similar to TFIIIA that binds exclusively to 5S RNA in Xenopus
.
Cell
61
,
293
300
.
Klambt
,
C.
and
Goodman
,
C. S.
(
1991
).
The diversity and pattern of glia during axon pathway formation in the Drosophila embryo
.
Glia
4
,
205
213
.
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
.
Matsuzaki
,
F.
,
Koizumi
,
K.
,
Hama
,
C.
,
Yoshioka
,
T.
and
Nabeshima
,
Y.
(
1992
).
Cloning of the Drosophila prospero gene and its expression in ganglion mother cells
.
Biochem. Biophys. Res. Comm
.
182
,
1326
1332
.
Mita-Miyazawa
,
I.
,
Ikegami
,
S.
and
Satoh
,
N.
(
1985
).
Histospecific acetylcholinesterase development in the presumptive muscle cells isolated from 16-cell-stage ascidian embryos with respect to the number of DNA replications
.
J. Embryol. Exp. Morph
.
87
,
1
12
.
Nasmyth
,
K.
,
Adolf
,
G.
,
Lydall
,
D.
and
Seddon
,
A.
(
1990
).
The identification of a second cell cycle control on the HO promoter in yeast: Cell cycle regulation of SWI5 nuclear entry
.
Cell
62
,
631
647
.
Newport
,
J.
and
Kirschner
,
M.
(
1982a
).
A major developmental transition in early Xenopus embryos: I. Characterization and timing of cellular changes at the midblastula stage
.
Cell
30
,
675
686
.
Newport
,
J.
and
Kirschner
,
M.
(
1982b
).
A major developmental transition in early Xenopus embryos: II. Control of the onset of transcription
.
Cell
30
,
687
696
.
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
.
Patel
,
N. H.
,
Snow
,
P. M.
and
Goodman
,
C. S.
(
1987
).
Characterization and cloning of fasciclin III: a glycoprotein expressed on a subset of neurons and axon pathways in Drosophila
.
Cell
48
,
975
88
.
Patel
,
N. H.
,
Schafer
,
B.
,
Goodman
,
C. S.
and
Holmgren
,
R.
(
1989
).
The role of segment polarity genes during Drosophila neurogenesis
.
Genes Dev
.
3
,
890
904
.
Pearson
,
W. R.
and
Lipman
,
D. J.
(
1988
).
Improved tools for biological sequence comparison
.
Proc. Natl. Acad. Sci. USA
85
,
2444
2448
.
Pelham
,
H. R. B.
and
Brown
,
D. D.
(
1980
).
A specific transcription factor that can bind either the 5S RNA gene or 5S RNA
.
Proc. Natl. Acad. Sci. USA
77
,
4170
4174
.
Poulson
,
D. F.
(
1950
).
Histogenesis, organogenesis, and differentiation in the embryo of Drosophila melanogaster
.
In The Biology of Drosophila
. (ed.
M.
Demerec
). pp.
168
274
.
New York
:
Wiley
.
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
.
Robertson
,
H. M.
,
Preston
,
C. R.
,
Phillis
,
R. W.
,
Johnson-Schlitz
,
D. M.
,
Benz
,
W. K.
and
Engels
,
W. R.
(
1988
).
A stable genomic source of P element transposase in Drosophila melanogaster
.
Genetics
118
,
461
470
.
Sambrook
,
J.
,
Fritsch
,
E. F.
and
Maniatis
,
T.
(
1989
).
Molecular Cloning: A Laboratory Manual
.
Cold Spring Harbor
:
Cold Spring Harbor Press
.
Smouse
,
D.
,
Goodman
,
C. S.
,
Mahowald
,
A. P.
and
Perrimon
,
N.
(
1988
).
polyhomeotic: a gene required for the embryonic development of axon pathways in the central nervous system of Drosophila
.
Genes Dev
.
2
,
830
842
.
Taghert
,
P. H.
and
Goodman
,
C. S.
(
1984
).
Cell determination and differentiation of identified serotonin-immunoreactive neurons in the grasshopper embryo
.
J. Neurosci
.
4
,
989
1000
.
Tautz
,
D.
and
Pfeiffle
,
C.
(
1989
).
A non-radioactive in situ hybridization method for the localization of specific RNAs in Drosophila embryos reveals translational control of the segmentation gene hunchback
.
Chromosoma
98
,
81
85
.
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
.
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
.
Wharton
,
K. A.
,
Yedvobnick
,
B.
,
Finnerty
,
V. G.
and
Artavanis-Tsakonas
,
S.
(
1985
).
opa: a novel family of transcribed repeats shared by the Notch locus and other developmentally regulated loci in D. melanogaster
.
Cell
40
,
55
62
.
Wheeler
,
W. M.
(
1891
).
Neuroblasts in the arthropod embryo
.
J. Morph
.
4
.
337
343
.