During animal development a complex network of highly specialised cells is generated in perfect coordination with other body structures, and consolidated with an outstanding degree of specificity in what we know as the nervous system. Finding how this intricate cellular network is generated and assembled is the aim of the current research in neurogenesis.

Understanding how the nervous system is built requires us to address questions concerning the mechanisms of pattern formation and of cell differentiation, signalling and recognition. In this review, I shall consider exclusively those aspects of neurogenesis regarding the orderly production of neuronal cell diversity in the fruit fly Drosophila melanogaster.

The advantage of Drosophila as an experimental system lies in the excellences of its genetic technology. The analysis of morphogenetic mutants aims at elucidating what genes participate in a given process and what their normal role is in it. The dissection of any biological process by genetic means relies exclusively on the assessment of mutant phenotypes and, therefore, the power of resolution of the scoring process is of paramount importance. Naturally, this resolution increases in proportion with our knowledge of the normal ontogeny of the process. The balanced consideration of these data normally leads to the conception of hypothetical models that describe cellular ontogeny in terms of gene expression and function. It is hoped that the critical testing of these proposals by modern techniques of gene manipulation will ultimately result in the formulation of a consistent theory of neural development.

Most of what we know about the ontogeny of Drosophila’s nervous system comes from comparisons with work first carried out in the grasshopper embryo. Due to its larger embryonic and cellular size, accessibility of some neurons and relatively simple ganglia, the grasshopper has been the system of choice for studies concerning the cellular basis of neural development. These studies stem from the long standing realisation that neurons can be uniquely identified by their place and date of birth and characteristic mor-phology, pattern of connections, physiology and biochemistry. This set of properties led to the construction of maps of neurons which, in turn, revealed the constancy in number and other pattern characteristics of these cells. It had also been known since the times of Wheeler (1891) and Bauer (1904) that neurons arise from precursor neuroblasts by repeated asymmetric division into ganglion mother cells and subsequent division of the latter into pairs of neurons. More recently it was Bate (1976), in an effort to understand the relationship between neuroblasts and their neuronal progeny, investigated the number and organisation of these precursor cells. The maps that he obtained revealed that the neuroblasts are organised in a bilaterally symmetrical, metameric pattern, as reliable in their number and arrange-ment as the adult cells that they produce.

These findings posed two questions: how do the neu-roblasts arise in an orderly pattern and how are neurons produced from these precursors? These two aspects were first addressed in some elegant experiments with the grasshopper embryo (Taghert et al., 1984; Taghert and Goodman, 1984; Doe and Goodman, 1985b).

The neuroblasts that give rise to the central nervous system originate from the neuroectoderm, an anteroposte-rior strip of ectoderm that includes and extends bilaterally from the ventral midline. The decision to become a neuro-blast is taken at random amongst small groups of neuroec -to dermal cells by a process involving cell interactions. Indeed, ablation of a neuroblast with a laser beam triggers another neuroectodermal cell of the group to replace it. This regulative capacity is interpreted as the result of a local inhibitory signal exerted by the neuroblast over adjacent neuroectodermal cells, thus preventing them from similarly becoming neuroblasts. The rest of the cells of the group, from which the neuroblast segregates, then develop as either glia or, epidermal non-neuronal support cells, or they die (Doe and Goodman, 1985a).

As a result of the rapid division cycle of the neuroblast, clones of ganglion mother cells form a columnar array on top of each neuroblast. Subsequent division of the ganglion mother cells originates a neuronal progeny that, therefore, is related by lineage to the underlying neuroblast. The lineage of the neuroblast appears to be invariant and cell autonomous. Invariant because at each division of the neuroblast a ganglion mother cell is produced with a restricted binary neuronal fate and this fate is specified by the ganglion mother cell mitotic ancestry. The autonomy of the neuroblast lineage (i.e. the unfolding of a lineage independently of its environment) is suggested by the results of two experiments. Firstly, if all the cells of one of those small ectodermal groups (distinguished by its late segregation timing and therefore surrounded by differentiated epidermal cells) are killed, the corresponding neuronal progeny is not produced, showing that (1) only neuroectodermal cells can give rise to neuroblasts and (2) neighbouring lineages do not change to compensate for the missing neuro-blast and its progeny. Secondly, ablation of a neuroblast after its first division leads to the persistence and differentiation of the ganglion mother cell produced and, at the same time, segregation of a new neuroblast, which will originate another identical ganglion mother cell. This leads to delayed duplication of the neuronal progeny produced by the first ganglion mother cell, showing that even in a changed environment the lineage develops normally.

The generation of ganglion mother cells, unlike that of neuroblasts, is not subjected to regulation and, in fact, killing of a ganglion mother cell results in the absence of its neuronal progeny (Doe and Goodman, 1985b). However, the binary fate of the ganglion mother cell progeny is not fixed and ablation of one of the neurons results in the remaining one developing into either of the two alternatives (Kuwada and Goodman, 1985). These experiments showed that cell interactions again intervene, this time to regulate the fate of sibling neurons, in the last step of the neuroblast lineage.

In summary, although only a fraction of some neuroblast lineages has been analysed and no lineage tracer experi-ments have been undertaken, the present evidence supports the notion that neuroblasts are born with a set of instruc-tions, or a genetic programme, to develop autonomously an invariant cell lineage.

As identified neurons can be traced back to the positions in the neuroectoderm where their precursor neuroblasts originate, it is important to ask how the fate of each of these precursor cells is specified. Three possibilities have been envisioned (Doe and Goodman, 1985b): (1) determination of individual neuroectodermal cells, (2) determination of groups of neuroectodermal cells that subsequently would resolve into one neuroblast per group by cell interactions and (3) all neuroectodermal cells are equivalent; cell inter-actions mediate the selection of some of these cells as neu-roblasts whose fate is then assigned in correlation with the position where segregation occurs.

The neuroblast ablation experiments described above were used again to discriminate amongst these alternatives, this time seeking the origin of the ectodermal cell that substitutes for the killed neuroblast. The results of three dif-ferent ablation experiments led to the suggestion that the ectodermal cell replacing the ablated neuroblast originates in an adjacent group, which normally would have resolved into a different lineage. This was interpreted as a positional translocation within the neuroectoderm, and it was taken to indicate that neuroectodermal cells are not determined to any specific neural fate, but are equivalent. On these grounds, the third alternative was favoured and consequently it was proposed that the decision to become a neuroblast is independent of the acquisition of its subsequent fate (i.e., neuronal progeny generated).

As revealing as the above experiments are, they are difficult and, as with any other approach to experimental embryology, they are subject to interpretation. An independent evaluation of these results was therefore sought and it was felt that genetic analysis could provide an alternative and complementary approach. Indeed, in such an experimental system, one could ask whether there are mutations that mimic the regulative properties manifested by the ablation of the neuroblast and neurons, or that impair the fate of the neuroblast and whether their phenotype is consistent with the models described. But first of all, how does one go about applying genetic analysis to insect neurogenesis?. Undoubtedly, Drosophila could provide such an experimental system, but would there be sufficient conservation of developmental strategies between hoppers and flies to support such comparison?.

The 300 Myr separating hoppers and flies are reflected in the substantial differences in their ontogeny. For instance, hoppers are hemimetabolous insects and flies holometabolous. This means that the former advance into adulthood directly from postembryonic development, while the latter has an intermediate larval stage followed by meta-morphosis. A second important difference concerns the organisation of the germ band. Grasshoppers are short-germ-band insects whereas Drosophila belongs to the long-germ-band class. This means that, in the former, the metameric germ band is generated by a budding off process from a short subterminal zone of the blastoderm. In contrast, in Drosophila, the metameric germ band arises with-out growth, by partition of the blastodermal space into seg-mental units (Anderson, 1972).

In spite of these differences, the comparative neuronal anatomy of these two organisms manifests an outstanding degree of similarity, if in a different time frame and size scale (Thomas et al., 1984). Indeed, a subset of well-known neurons of the common, ladder-like scaffold of hoppers and flies display identical morphology, growth cone and selective fasciculation patterns. These attributes are thought to reflect the deployment of similar temporal and spatial recognition signals. It is therefore not surprising that the overall features of neural developmental also appear to follow a common strategy. The neuroblasts similarly arise from a subpopulation of an identically positioned neuroectodermal cell layer and attain the characteristic symmetrical, metameric pattern described for the grasshopper (Campos-Ortega and Hartenstein, 1985; Doe et al., 1988a). Furthermore, some identified hopper neuroblasts have been found in equivalent positions in Drosophila and shown to originate some similar neurons via comparable lineages (Doe et al., 1988a). These lineages also appear to be cell autonomous, as in vitro cultured neuroblasts divide normally to produce clones of ganglion mother cells and neurons and the proportion of the latter expressing serotonergic and dopaminergic phenotypes is similar to the number found in vivo (Huff et al., 1989).

While hopper neuroblasts complete their divisions in the embryo to generate the mature adult nervous system, fly neuroblasts continue to divide in larval life after a period of ‘dormancy’ (Truman and Bate, 1988; Prokop and Tech-nau, 1991). During Drosophila’s larval development, neuroblasts resume their activity in a fashion characteristic of the late portion of the grasshopper’s embryogenesis.

In summary, the development of the nervous systems of hoppers and flies follows a different tempo but its mode appears to be identical.

Genes required for neurogenesis were classically identified in genetic screens where the integrity of the nervous system was scored by histological methods. More recently, geneticists started to apply molecular biology techniques to monitor gene expression patterns and these studies revealed a considerable amount of genetic activity in the developing nervous system. Subsequently, ‘pattern searches’ by the enhancer trap technique (O’Kane and Gehring, 1987) yielded additional candidates. However, most of the genes selected by their pattern of expression have not yet been analysed genetically and, therefore, their role in neurogen-esis remains obscure.

One reason for this lack of genetic data is the coinci-dence of the ectoderm as common anlage for neural and epidermal development; mutants of genes required in both pathways will be defective in neurogenesis primarily because of abnormal ectodermal patterning. Testing the requirement of such genes in neurogenesis, independently of their epidermal function, is a major challenge to the cur-rent research and one of considerable technical difficulty. The correlation between patterns of gene expression and neural ontogeny, however, has served several useful pur-poses: as a guideline to select genes that are likely to con-tribute to those aspects of neurogenesis under scrutiny, as a high resolution phenotypic trait and as an excellent his-tological marker to probe morphology beyond the level of resolution of standard techniques.

I will begin this account of the genetic control of neu-rogenesis by discussing postneuroblast gene activity fol-lowed by preneuroblast development.

Gene activity in postneuroblast development

Postneuroblast development is centred around the charac-teristic lineage of these precursors and their progeny. The developmental potential of the neuroblasts is reflected by the complexity of gene activity in these cells and their ganglion mother cell progeny. Two groups of genes active in these cells can be distinguished, depending on whether or not their spatial domains of expression are inherited from the underlying ectoderm.

The homeotic and segment polarity genes are first expressed and required in the ectoderm and then introduced into the nervous system by segregating neuroblasts, with precise preservation of their spatial boundaries. The domains of expression of these genes, therefore, are dictated by their ectodermal function, regardless of whether their regulation is actively or passively maintained in the nervous system.

The homeotic genes are expressed and required in mul-timetameric portions of the ectoderm for the establishment and maintenance of specific segmental identities (Lewis, 1978; Kaufman et al., 1990). Segmental differences in neural development are most obvious between, for example, thoracic and abdominal ganglia of the larva. The production of more neuronal progeny by thoracic neuroblasts is the main mechanism responsible for this differential development (Bate, 1976; Truman and Bate, 1988; Shepherd and Bate, 1990), and the homeotic genes are the most likely candidates to regulate this aspect of neuroblast activity. This is, however, a feature of postembryonic development. During early neurogenesis, the homeotic genes have been proposed to control the small differences in the number of neuroblasts segregating in each metamere by regulating the activity of neuroblast generating genes (Doe, 1992).

Other aspects of homeotic gene function, such as their expression in embryonic neurons, remain elusive. This is a particularly intriguing problem, given the conservation in genomic cluster arrangement, protein sequence and patterns of expression of these genes in the nervous systems of such a wide variety of animals, from flies to mammals (Graham et al., 1989).

The segment polarity genes are required in the ectoderm for normal intrasegmental patterning of epidermal derivatives of the larval cuticle (Nüsslein-Volhard and Wieschaus, 1980). The expression of some segment polarity genes has been studied and shown to occur in one or two stripes per metamere (DiNardo and Heemskerk, 1990). The analysis of the nervous system of mutants for some of these genes led to the speculation that their wild-type alleles could con-tribute to the control of neuroblast fate (Patel et al., 1989). However, the patterns of ectodermal expression of many segment polarity genes are tightly interdependent, such that the absence of any one gene leads to complex rearrangements of the normal boundaries of the others (DiNardo and Heemskerk, 1990). In these circumstances, the phenotypes observed are, most likely, the result of a re-patterned ectoderm and thus the role of the segment polarity genes in neural development remains unclear.

In contrast to homeotic and segment polarity genes, a second group of genes is specifically activated in neurob-lasts and ganglion mother cells (Table 1). Although many members of this group belong to the gap and pair-rule classes of segmentation genes, or are part of the dorsoven-tral pathway (Nüsslein-Volhard and Wieschaus, 1980; St. Johnston and Nüsslein-Volhard, 1992), this early require-ment entails only transient expression. When redeployment of these genes occurs in neural precursors, they do so in patterns totally unrelated to their previous boundaries of expression. This suggests that these genes constitute a spe-cial group, specifically deployed in close connection with a distinct neural function.

Table 1.

Neural Fate Genes

Neural Fate Genes
Neural Fate Genes

Two genes fushi tarazu (ftz) and even-skipped (eve) are activated in subsets of ganglion mother cells and expressed subsequently in neurons. The requirement of these two genes has been tested by monitoring the differentiation of the RP1/RP2 and aCC/pCC pairs of sibling neurons. At the non-permissive temperature, an eve temperature-sensitive allele, often leads to the appearance of two RP1s and two pCCs, whereas a ftz chimera, lacking a promoter element necessary for the expression of this gene in the nervous system, similarly produces two RP1s but leaves the CC cells unaltered (Doe et al., 1988a; 1988b).

The phenotypic analysis of eve and ftz mimics the sibling ablation experiments carried out in the grasshopper embryo (Kuwada and Goodman, 1985). This correlation indicates that the binary fate of the ganglion mother cells is under genetic control and suggests that the interactions between the two sibling neurons analysed depend on the activity of these two genes. A more recent and detailed interpretation of the genetic results above suggests that the role of eve and ftz is to confer ganglion mother cell identity; two ganglion mother cells immediately related by their ancestry would attain identical fates in the absence of either of these two genes (Doe, 1992).

Although the importance of these findings is obvious, there are aspects that will require further study. For instance, in the absence of ftz, most neurons appear to dif-ferentiate normally (Hiromi and Gehring, 1987); in partic-ular, six well-studied cases showed wild-type axon mor-phology (Doe et al., 1988a). Although this suggests that ftz expression may not be required by most neural lineages, scoring other traits might reveal more subtle roles for this gene.

Finally, it is clear that a reasonable understanding of gan-glion mother cell and sibling determination will involve the identification of many more genes; some 125 ganglion mother cells would have to be specified for the generation of the approximately 250 neurons present in each embry-onic neuromere.

Fourteen genes have been found that are activated in neu-roblasts (Table 1), most of which are also expressed in ganglion mother cell and neurons. The genetic requirements of three members of this group (prospero, runt and cut) have been analysed. prospero RNA is expressed by most neu-roblasts and ganglion mother cells but not by neurons (Doe et al., 1991; Vässin et al., 1991). prospero protein, how-ever, accumulates in ganglion mother cells only (Vässin et al., 1991). The earliest effect of prospero null alleles is detected in the pattern of gene expression of eve and ftz. Indeed, despite the fact that neuroblasts segregate normally and ganglion mother cell progeny are produced, the number of ganglion mother cells expressing eve and ftz is reduced. Conversely, the expression of another ganglion mother cell marker gene, engrailed, is widened (Doe et al., 1991). These effects appear to be elicited specifically at the gan-glion mother cell level, for the expression of the neuroblast marker gene hunchback appears to be normal.

Consequent to the influence of prospero in ganglion mother cell gene expression, prospero embryos display aberrant neuronal fasciculation (Doe et al., 1991; Vässin et al., 1991). The prospero gene, therefore, is required for the normal expression of, at least, a subset of the neuroblast traits. Other aspects of the neuroblast lineage, like the number of cell divisions and characteristic neuronal progeny, can not be easily scored with the available techniques and, thus, a full account of the function of prospero will require further work.

runt is a segmentation gene also expressed by subsets of neuroblasts, ganglion mother cells and neurons (Duffy et al., 1991). Its function in the nervous system has been tested with a transgenic construct that is normally expressed during segmentation but is defective, at least in part, in neural expression. It has also been examined with the aid of a temperature-sensitive allele. As some of the neurons that express runt co-express eve, the role of the former in the nervous system was assessed by scoring eve-expressing cells. It was shown that, indeed, the absence of runt pro-duces a dramatic reduction in the number of eve-express-ing neurons suggesting that runt is required for some step of cell identity downstream in the neuroblast lineage (Duffy et al., 1991).

The requirement of cut has been analysed only in the peripheral nervous system where its absence was shown to provoke a clean transformation between two types of sensory structure: the so-called ‘external’ into ‘chordotonal’ organs (Bodmer et al., 1987). Both types of organ are generated from single precursor cells that produce distinct lineages (Bodmer et al., 1989) and thus the transformation caused by the lack of cut is thought to stem from mis-spec-ification at the precursor level (Blochlinger et al., 1990). In line with this, cut protein is expressed in the precursors of the ‘external’ organs and subsequently in all their progeny, but is absent from the ‘chordotonal’ organ lineage (Blochlinger et al., 1990). This is consistent with a model in which cut would impose a new differentiation pathway to a precursor cell already specified to a more basal fate. Whether this mechanism of fate determination is a partic-ular feature of the more simple sensory organ lineages is not yet known. The comparative distribution of genes activated in neuroblasts is suggestive of a similar combinatorial code of gene expression that would implement alternative fates to different neuroblasts.

Finally, the fact that for neuroblast-specific gene activation occurs without apparent relationship to the ongoing process of ectodermal segmentation suggests the existence of a mechanism distinctly responsible for this operation. It is clear, however, that such a mechanism can not be independent of ectodermal positional cues, because the nervous system is metamerically organised and is generated in a precise region of the embryo in coordination with other body structures for their normal functional interaction. How this remarkable pattern of neuroblast-specific gene expression is activated concerns preneuroblast development.

Gene activity in preneuroblast development

The first sign of gene expression detected in the embryo in relation to neurogenesis is that of the achaete-scute gene complex (ASC) (Cabrera et al., 1987; Romaní et al., 1987; Alonso and Cabrera, 1988; Cabrera, 1990; Martín-Bermudo et al., 1991). The ASC and adjacent proximal chromoso-mal regions constitute a complex locus of qualitatively sim-ilar genetic functions. The histological analysis of individ-uals carrying progressively larger deficiencies of this region showed correspondingly increasing levels of neural hypotrophy. Duplication of the ASC region proved capable of rescuing the largest deficiency to levels quantitatively comparable to the deletion of the ASC alone, suggesting some degree of functional redundancy (Jiménez and Campos-Ortega, 1979). Finding that the ASC comprises four homologous genes [achaete, scute, lethal of scute and asense (Villares and Cabrera, 1987; Alonso and Cabrera, 1988)], together with the genetic results just described, suggested the existence of a clustered gene family (hence-forth referred to as the AS-Group), of which the ASC is a subset. Different members of this family appear to con-tribute in different extents to a common function in neuro-genesis. What is this function?.

Certainly, one functional aspect of the AS-Group is the generation of the neuroblasts: the levels of neural hypotro-phy elicited by progressively larger deficiencies are correlated with an escalating absence of neuroblasts (Cabrera et al., 1987; Jiménez and Campos-Ortega, 1990; Martín-Bermudo et al., 1991). However, the extent of neural hypotrophy always exceeds the levels expected from the number of precursors missing and, in fact, an additional cause for the loss of nervous system is the increased amount of cell death during postneuroblast development (Jiménez and Campos-Ortega, 1990).

These data indicate that the requirements for the AS-Group span pre- and postneuroblast development; hence, either the generation of neuroblasts is linked to their sub-sequent fate or these genes are utilised at two different times during neurogenesis. As expression of the three best known members of the AS-Group (the achaete, scute and lethal of scute genes) occurs continuously from their onset in the ectoderm until the neuroblasts have segregated, when they are switched off, the former seems to be the correct alter-native (Cabrera et al., 1987; Alonso and Cabrera, 1988; Cabrera, 1990 and unpublished data).

The most obvious regulatory role of the AS-Group in the fate of the neuroblasts is the onset of these precursor’s specific patterns of gene expression (Alonso and Cabrera, 1988; Cabrera and Alonso, 1991). Indeed, the ASC prod-ucts form DNA-binding heterodimers with the product of the daughterless gene, these heterodimers bind to specific DNA sequences (Murre et al., 1989; Cabrera and Alonso, 1991; van Doren et al., 1991) and they function as transcriptional activators in a yeast assay system (Cabrera and Alonso, 1991). A motif for ASC/DA binding has been found in the promoter of the hunchback gene (see Table 1) and shown to mediate transcriptional activation in the yeast system (Cabrera and Alonso, 1991). In line with this, embryos deficient for the lethal of scute gene fail to activate hunchback in a third of the neuroblasts generated during the first wave of segregation, suggesting an ASC-dependent activation of hunchback in the neuroblasts (Cabrera and Alonso, 1991). In addition, the three resulting ASC/DA heterodimers display characteristic affinities for a given sequence, suggesting that each ASC gene might be responsible for an unique aspect of this pattern of gene activation (Cabrera and Alonso, 1991). The technical difficulties inherent in the detailed analysis of neuroblast lineages, however, have precluded the stringent assessment of the contribution of each of the components of the locus. Nonetheless, it is conceivable that, if the different members of the AS-Group were expressed in partially overlapping patterns [as is the case between achaete and scute whose expression is totally overlapping and interdependent (Cabrera unpublished data)], the cumulative result of the deletion of several of its components would be reflected in the aberrant specification of those cells still expressing some of these functions. If, however, the full complement of genes expressed in a given cell had been deleted, specification would not occur at all. In the former case the lineage would be abnormal and cell death might ensue; in the latter neu-roblasts would fail to segregate.

Other authors, however, have argued against the involve-ment of the AS-Group of genes in neuroblast fate determination (Jiménez and Campos-Ortega, 1990; Martín-Bermudo et al., 1991; Doe, 1992). Their view is that the AS-Group is composed of redundant functions with only partial requirements in the segregation process, providing ectodermal cells with an unspecified neuroblast state. The lack of nervous system arising from single gene deficien-cies that do not perturb neuroblast segregation is thought to be due to physiological weakness of the resulting pre-cursors (Jiménez and Campos-Ortega, 1990).

The laser ablation experiments carried out with the grasshopper embryo first suggested that the decision to become a neuroblast is independent of the acquisition of its subsequent neural fate (Doe and Goodman, 1985b). The data discussed above suggest that, on the contrary, the gen-eration of the neuroblasts is linked to the acquisition of genetic instructions that influence their subsequent fate (see Fig. 1). The AS-Group appears to be responsible for the acquisition of these fates (Cabrera et al., 1987; Alonso and Cabrera, 1988; Cabrera, 1990; Cabrera and Alonso, 1991). How is the activity of the ASC genes implemented?.

The onset of ASC RNA expression germane to neurogenesis occurs in the cellular blastoderm in the form of one dorsoventrally split stripe per metameric unit (Cabrera et al., 1987; Cabrera, 1990 and unpublished data). These stripes extend from the boundary of the mesodermal anlage to the dorsal midline and undoubtedly must result from the positional specification laid down by both the anteroposte-rior and dorsoventral axis forming systems (St. Johnston, D. and Nüsslein-Volhard, 1992). As the germ band extends, the blastoderm stripes of expression are transformed into discrete groups of ASC RNA-expressing cells positioned at the ventral neuroectoderm. Some of these cells will segregate as neuroblasts (see below) and those remaining on the ectoderm cease expressing ASC RNA quickly (Cabrera et al., 1987; Cabrera, 1990). Subsequent ASC expression reap-pears in a complex pattern, both in the central, and later in the peripheral, nervous systems (unpublished data). How this expression is implemented is as yet unknown for, despite a clear dorsoventral component in the arrangement of the peripheral nervous system (Dambly-Chaudière and Ghysen, 1986), the expression of the pair-rule genes has already faded away. This second round of ASC expression labels the precursors of the second and third wave of central neuroblasts and subsequently the peripheral precursors. It is at this stage that an as yet ill-defined ectodermal patterning system, possibly based on cellular interactions, may influence the outcome of neuroblast identity by regulating ASC expression.

In summary, according to our current working model, a combinatorial code of the AS-Group expression determines neuroblast lineage specification. The outcome of this code would entail the activation of different sets of neuroblast-specific genes in correlation with the mosaic of fates char-acteristic of these precursor cells (Fig. 1).

The segregation of the neuroblasts

Laser ablation experiments showed that neuroblast segregation is a stochastic process, governed by interactions amongst groups of equivalent ectodermal cells (Taghert et al., 1984; Taghert and Goodman, 1984; Doe and Goodman, 1985b). Genes of the neurogenic class were first proposed to mediate this process because the phenotype of their lack-of-function alleles mimics the laser ablations (Doe and Goodman, 1985b); that is, their mutations elicit the segregation of supernumerary neuroblasts (Poulson, 1937; Lehmann et al., 1983; Bourouis et al., 1989; Cabrera, 1990; Hoppe and Greenspan, 1990). This is in contrast to the phenotype shown by some ASC mutants in which neurob-lasts fail to segregate (Cabrera et al., 1987; Jiménez and Campos-Ortega, 1990; Martín-Bermudo et al., 1991). It is, therefore, pertinent to ask how does the activity of the neurogenic genes relate to the expression of the ASC genes?.

The first clue to this process arose from the observation that ASC RNA accumulates in groups of cells and that only a fraction of those cells actually segregate as neuroblasts (Cabrera et al., 1987). Subsequently, by using an antibody specific for the lethal of scute protein, it was shown that a fraction of the RNA-producing cells accumulate the protein and that these latter cells correspond to the neuroblasts [Cabrera, 1990; similar results have since been obtained with an achaete-specific antibody (unpublished data)].

In embryos mutant for neurogenic genes the domains of lethal of scute RNA remain normal, but every RNA-con-taining cell in the mutant accumulates protein and the number of cells segregating as neuroblasts increases. This observation suggests that the lethal of scute is regulated post-transcriptionally (Cabrera, 1990).

Antibodies have been generated against two forms of the lethal of scute protein. One of them recognises an epitope in a potential protein kinase phosphorylation site (Villares and Cabrera, 1987; Alonso and Cabrera, 1988; Cabrera, 1990). The second one was raised against the whole pro-tein (Martín-Bermudo et al., 1991). The former antibody recognises a protein that is produced only in a subset of the cells in which the lethal of scute RNA is detected and labels the segregating neuroblasts (Cabrera, 1990). The second antibody detects a product that is present in every cell that accumulates the lethal of scute RNA and therefore labels the segregating neuroblasts as well as those cells failing to segregate (Martín-Bermudo et al., 1991). These observations suggest the following working hypothesis: there are two forms of the lethal of scute protein; (1) an unphos-phorylated or active form detected with the former anti-body, which shows a restricted domain of accumulation that is likely the result of post-transcriptional regulation, and (2) a phosphorylated form, present in every cell that contains the lethal of scute RNA and inactive.

It is conceivable that some of the neurogenic genes encoding membrane-bound products (Notch and Delta;Wharton et al., 1985; Kidd et al., 1986; Knust et al., 1987; Kopczymski et al., 1988) function as receptors of a signal that is involved in the orderly process of neuroblast segregation. This signal would reach the lethal of scute protein in the nuclei by a process of signal transduction. It is note-worthy that the product of another neurogenic gene, shaggy, encodes a potential kinase (Bourouis et al., 1989 and 1990; Siegfried et al., 1990) and it is, therefore, a candidate for involvement in the signal transduction process leading to the phosphorylation/inactivation of ASC proteins.

The neurogenic genes, therefore, provide a device for pattern refinement which has been classically known as ‘lateral inhibition’ (Wigglesworth, 1940). The post-transcrip-tional regulation of the lethal of scute gene provides a means to visualise this process. That is, a group of cells all of which express lethal of scute is capable of regulating its subsequent fate by means of controlling which cells accu-mulate a specific form of the protein. Those cells potentially capable of taking upon a given fate are called ‘equivalence groups’ and were first described during vulva development in the nematode C. elegans (Kimble, 1981; Kimble and White, 1981). This mechanism of cell diversification appears to be a prevalent one and, in addition, has been shown to occur during bristle development in Drosophila (Stern, 1968) and algae (Wilcox et al., 1973). This process may respond to the lack of precision of the underlying molecular machinery to select a single cell out of a complex population and therefore the alternative strategy is used: a group is selected and competition amongst the cells in the group, a different subroutine, resolves the problem.

That the extra neuroblasts segregated in neurogenic mutants correspond to equivalence groups is supported by two lines of evidence. Firstly, neural precursor markers like ftz display the corresponding enlarged domain of expression in neurogenic mutant embryos (unpublished data). Secondly, the number of differentiated cell types in peripheral sensory organs is increased in these mutants, again indicating that multiple, identical lineages have been generated (Hartenstein and Campos-Ortega, 1986; see Fig. 2). Given that most neuroectodermal cells segregate in neurogenic mutants, the fact that correct lineages are still generated provides a very strong case for the specification of neu-roblasts while still in the ectoderm (see Fig. 1).

Certainly the main theme of neurogenic gene function discussed here is not their exclusive activity. For one thing only a few components of this group have been mentioned (see Lehmann et al., 1983). In addition, requirements for these genes have been demonstrated in mesodermal development (Corbin et al., 1991) and neurogenic gene function is required as well in epidermal cells for their viability (Hoppe and Greenspan, 1990).

Neuroblast segregation is a highly regulated process in which the timing and number of cells involved is controlled (Bate, 1976; Doe and Goodman, 1985a; Campos-Ortega and Hartenstein, 1985; Doe et al., 1988a; Shepherd and Bate, 1990). What triggers neuroblast segregation and therefore the initiation of neurogenesis?. Inductive processes similar to those described in vertebrates have been ruled out (Leptin, 1991; Kosman et al., 1991; Rao et al., 1991). The nature of the signal in Drosophila is unknown. However, the discovery that the membrane-bound Notch and Delta proteins can interact heterotypically in a fashion characteristic of cell adhesion molecules has opened new horizons (Fehon et al., 1990). Indeed, cellular properties such as cell sorting have been shown to result from differences in the amount and the degree of specificity of interaction of the mammalian cell adhesion molecules expressed in tissue culture (Friedlander et al., 1989). Similar modulations could take place during the expression of Notch and Delta. It is conceivable that a signal may be generated in this way and that transduction of this signal to the nuclei activates the ASC proteins, thus triggering neurogenesis.

I will never be able to thank enough Juan Botas for his help and encouragement in getting this review finished. I thank also María, Minia and Renate for so much dedication and care. I am most grateful to Ralph Greenspan for discussions and for bring-ing to my attention the cell adhesion paper and to G. Currie for reading the manuscript. Some of the author’s work mentioned in the text was supported by a MRC Project Grant.

Alonso
,
M. C.
and
Cabrera
,
C. V.
(
1988
).
The achaete-scute gene complex of Drosophila melanogaster comprises four homologous genes
.
EMBO J
.
7
,
2585
2591
.
Anderson
,
D.
(
1972
).
The development of hemimetabolous insects
.
In Developmental Systems: Insects
vol.
1
(ed.
S.J.
Counce
and
C.H.
Waddington
), pp.
96
165
.
London
:
Academic Press
.
Bate
,
C. M.
(
1976
).
Embryogenesis of an insect nervous system I. A map of the thoracic and abdominal neuroblasts in Locus migratoria
.
J. Embryol. Exp. Morph
.
35
,
107
123
.
Bauer
,
V.
(
1904
).
Zur inneren Metamorphose des Zentralnerveusystems der Insekten
.
Zool. Jhb. Anat. Omtd. Tiere
.
20
,
123
150
.
Baumgartner
,
S.
,
Bopp
,
D.
,
Burri
,
M.
and
Noll
,
M.
(
1987
).
Structure of two genes at the goosberry locus related to the paired gene and their spatial expression during Drosophila embryogenesis
.
Genes and Development
1
,
1247
1267
.
Blochlinger
,
K.
,
Bodmer
,
R.
,
Jan
,
L. Y.
and
Jan
,
Y. N.
(
1990
).
Patterns of expression of cut, a protein required for external sensory organ development in wild-type and cut mutant Drosophila embryos
.
Genes and Development
4
,
1322
1331
.
Blochlinger
,
K.
,
Bodmer
,
R.
,
Jack
,
J.
,
Jan
,
L. Y.
and
Jan
,
Y. N.
(
1988
).
Primary structure and expression of a product from cut, a locus involved in specifying sensory organ identity in Drosophila
.
Nature
333
,
629
635
.
Bodmer
,
R.
,
Barbel
,
S.
,
Sheperd
,
S.
,
Jack
,
J. W.
,
Jan
,
L. Y.
and
Jan
,
Y. N.
(
1987
).
Transformation of sensory organs by mutations of the cut locus of D. melanogaster
.
Cell
51
,
293
307
.
Bodmer
,
R.
,
Carretto
,
R.
and
Jan
,
Y. N.
(
1989
).
Neurogenesis of the peripheral nervous system in Drosophila embryos: DNA replication patterns and cell lineages
.
Neuron
3
,
21
32
.
Bopp
,
D.
,
Jamet
,
E.
,
Baumgartner
,
S.
,
Burri
,
M.
and
Noll
,
M.
(
1989
).
Isolation of two tissue-specific Drosophila paired box genes, Pox meso and Pox neuro
.
EMBO J
.
8
,
3447
57
.
Boulay
,
J. L.
,
Dennefeld
,
C.
and
Alberga
,
A.
(
1987
).
The Drosophila developmental gene snail encodes a protein with nucleic acid binding fingers
.
Nature
330
,
395
398
.
Bourouis
,
M.
,
Heitzler
,
P.
,
El Messal
,
M.
and
Simpson
,
P.
(
1989
).
Mutant Drosophila embryos in which all cells adopt a neural fate
.
Nature
341
,
442
445
.
Bourouis
,
M.
,
Moore
,
P.
,
Ruel
,
L.
,
Grau
,
Y.
,
Heitzler
,
P.
and
Simpson
,
P.
(
1990
).
An early embryonic product of the gene shaggy encodes a serine/threonine protein kinase related to the CDC28/cdc2+ subfamily
.
EMBO J
.
9
,
2877
2884
.
Cabrera
,
C. V.
(
1990
).
Neuroblast determination and segregation in Drosophila: the interactions between scute, Notch and Delta
.
Development
109
,
733
742
. Reprinted in vol. 110(1).
Cabrera
,
C. V.
and
Alonso
,
M. C.
(
1991
).
Transcriptional activation by heterodimers of the achaete-scute andaughterless gene products of Drosophila
.
EMBO J
.
10
,
2965
2973
.
Cabrera
,
C. V.
,
Martínez-Arias
,
A.
and
Bate
,
M.
(
1987
).
The expression of three members of the achaete-scute gene complex correlates with neuroblast segregation in Drosophila
.
Cell
50
,
425
433
.
Campos-Ortega
,
J. A.
and
Hartenstein
,
V.
(
1985
).
The Embryonic Development of Drosophila melanogaster
Berlin Heidelberg
:
Springer-Verlag
.
Caudy
,
M.
,
Grell
,
E. H.
,
Dambly-Chaudière
,
Ch
,
Ghysen
,
A.
,
Jan
,
L. Y.
and
Jan
,
Y. N.
(
1988a
).
The maternal sex determination gene daughterless has zygotic activity necessary for the formation of peripheral neurons in Drosophila
.
Genes and Development
2
,
843
852
.
Caudy
,
M.
,
Vässin
,
H.
,
Brand
,
M.
,
Tuma
,
R.
,
Jan
,
L. Y.
and
Jan
,
Y. N.
(
1988b
).
daughterless, a Drosophila gene essential for both neurogenesis and sex determination, has sequence similarities to myc and the achaete-scute complex
.
Cell
55
,
1061
1067
.
Corbin
,
V.
,
Michelson
,
A. M.
,
Abmayr
,
S. M.
,
Neel
,
V.
,
Alcamo
,
E.
,
Maniatis
,
T.
and
Young
,
M. W.
(
1991
).
A role for the Drosophila neurogenic genes in mesoderm differentiation
.
Cell
67
,
311
323
.
Cronmiller
,
C.
,
Schedl
,
P.
and
Cline
,
T. W.
(
1988
).
Molecular characterization of daughterless, a Drosophila sex determination gene with multiple roles in development
.
Genes and Development
2
,
1666
1676
.
Dambly-Chaudière
,
Ch
and
Ghysen
,
A.
(
1986
).
The sense organs in the Drosophila larva and their relation to the embryonic pattern of sensory neurons
.
Roux’s Arch. Dev. Biol
.
195
,
222
228
.
Dick
,
T.
,
Yang
,
X.
,
Yeo
,
S.
and
Chia
,
W.
(
1991
).
Two closely linked Drosophila POU domain genes are expressed in neuroblasts and sensory
Proc. Natl. Acad. Sci. USA
88
,
7645
7649
.
DiNardo
,
S.
and
Heemskerk
,
J.
(
1990
).
Molecular and cellular interactions responsible for intrasegmental patterning during Drosophila embryogenesis
.
Seminars in Cell Biology
1
,
173
183
.
Doe
,
Ch Q.
(
1992
).
The generation of neuronal diversity in the Drosophila embryonic central nervous system
.
In Determination of Neuronal Identity (in press)
(eds.
Shankland
,
M.
,
Macagno
,
E.
),
New York
:
Academic Press
.
Doe
,
Ch 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
,
Ch 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
,
Ch 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
,
Ch 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
,
Ch Q.
,
Smouse
,
D.
and
Goodman
,
C. S.
(
1988b
).
Control of neuronal fate by the Drosophila segmentation gene even-skipped
.
Nature
333
,
376
378
.
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
.
Fehon
,
R. G.
,
Kooh
,
P. J.
,
Rebay
,
I.
,
Regan
,
C. L.
,
Xu
,
T.
,
Muskavitch
,
M. A. T.
and
Artavanis-Tsakonas
,
S.
(
1990
).
Molecular interactions between the protein products of the neurogenic loci Notch and Delta, two EGF-homologous genes in Drosophila
.
Cell
61
,
523
534
.
Frasch
,
M.
,
Hoey
,
T.
,
Rushlow
,
Ch
,
Doyle
,
H.
and
Levine
,
M.
(
1987
).
Characterization and localization of the even-skipped protein of Drosophila
.
EMBO J
.
6
,
749
759
.
Friedlander
,
D. R.
,
Mège
,
R-M.
,
Cunningham
,
B. A.
and
Edelman
,
G. M.
(
1989
).
Cell sorting-out is modulated by both the specificity and amount of different cell adhesion molecules (CAMs) expressed on cell surfaces
.
Proc. Natl. Acad. Sci. USA
86
,
7043
7048
.
Gaul
,
U.
,
Seifert
,
E.
,
Schuh
,
R.
and
Jäckle
,
H.
(
1987
).
Analysis of Krüppel protein distribution during early Drosophila development reveals posttranscriptional regulation
.
Cell
50
,
639
647
.
Graham
,
A.
,
Papalopulu
,
N.
and
Krumlauf
,
R.
(
1989
).
The murine and Drosophila homeobox gene complexes have common features of organisation and expression
.
Cell
57
,
367
378
.
Hartenstein
,
V.
and
Campos-Ortega
,
J. A.
(
1986
).
The peripheral nervous system of mutants of early neurogenesis in Drosophila melanogaster
.
Roux’s Arch. Dev. Biol
.
195
,
210
221
.
Hiromi
,
Y.
and
Gehring
,
W. J.
(
1987
).
Regulation and function of the Drosophila segmentation gene fushi tarazu
.
Cell
50
,
963
974
.
Hoppe
,
P. E.
and
Greenspan
,
R. J.
(
1990
).
The Notch locus of Drosophila is required in epidermal cells for epidermal development
.
Development
109
,
875
885
.
Huff
,
R.
,
Furst
,
A.
and
Mahowald
,
A. P.
(
1989
).
Drosophila embryonic neuroblasts in culture: autonomous differentiation of specific neurotransmitters
.
Dev. Biol
.
134
,
146
157
.
Jiménez
,
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
.
Jiménez
,
F.
and
Campos-Ortega
,
J.
(
1979
).
On a region of the Drosophila genome necessary for central nervous system development
.
Nature
282
,
310
312
.
Kaufman
,
T. C.
,
Seeger
,
M. A.
, and
Olsen
,
G.
(
1990
).
Molecular and genetic organization of the antennapedia gene complex of Drosophila melanogaster
.
Adv. Genet
,
27
,
309
62
.
Kidd
,
S.
,
Kelley
,
M. W.
and
Young
,
M. W.
(
1986
).
Sequence of the Notch locus of Drosophila: relationship of the encoded protein to mammalian cotting and growth factors
.
Mol. Cell. Biol
.
6
,
3094
3108
.
Kimble
,
J.
(
1981
).
Alterations in cell lineage following laser ablation of cells in the somatic gonad of Caenorhabditis elegans
.
Dev. Biol
.
87
,
286
300
.
Kimble
,
J.
and
White
,
J. G.
(
1981
).
On the control of germ cell development in Caenorhabditis elegans
.
Dev. Biol
.
81
,
208
219
.
Knust
,
E.
,
Dietrich
,
U.
,
Tepass
,
U.
,
Bremer
,
K. A.
,
Wiegel
,
D.
,
Vässin
,
H.
and
Campos-Ortega
,
J. A.
(
1987
).
EGF homologous sequences encoded in the genome of Drosophila melanogaster, and their relation to neurogenic genes
.
EMBO J
.
6
,
761
766
.
Kopczymski
,
C. C.
,
Alton
,
A. K.
,
Fechtel
,
K.
,
Kooh
,
P. J.
and
Muskavitch
,
M. A. T.
(
1988
).
Delta, a Drosophila neurogenic gene, is transcriptionally complex and encodes a protein related to blood coagulation factors and epidermal growth factor of vertebrates
.
Genes and Development
2
,
1723
1735
.
Kosman
,
D.
,
Ip
,
Y. T.
,
Levine
,
M.
, and
Arora
,
K.
(
1991
).
Establishment of the mesoderm-neuroectoderm boundary in the Drosophila embryo
.
Science
254
,
118
22
.
Kuwada
,
J. Y.
and
Goodman
,
C. S.
(
1985
).
Neuronal determination during embryonic development of the grasshopper system
.
Dev. Biol
.
110
,
114
126
.
Laughon
,
A.
and
Scott
,
M. P.
(
1984
).
Sequence of Drosophila segmentation gene: protein structure homology with DNA-binding proteins
.
Nature
310
,
25
31
.
Lehmann
,
R.
,
Jiménez
,
F.
,
Dietrich
,
U.
and
Campos-Ortega
,
J.
(
1983
).
On the phenotype and development of mutants of early neurogenesis in Drosophila melanogaster
.
Roux’s Arch. Dev. Biol
.
192
,
62
74
.
Leptin
,
M.
(
1991
).
twist and snail as positive and negative regulators during Drosophila development
.
Genes and Development
5
,
1568
1576
.
Lewis
,
E. B.
(
1978
).
A gene complex controlling segmentation in Drosophila
.
Nature
276
,
565
570
.
Martín-Bermudo
,
M. D.
,
Martínez
,
C.
,
Rodríguez
,
A.
and
Jiménez
,
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.
,
Webe
,
U.
,
Goodman
,
C.
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
.
Murre
,
C.
,
McCaw
,
P. S.
,
Vässin
,
H.
,
Caudy
,
M.
,
Jan
,
L. Y.
,
Jan
,
Y. N.
,
Cabrera
,
C. V.
,
Lassar
,
A. B.
,
Weintraub
,
H.
and
Baltimore
,
D.
(
1989
).
Interactions between heterologous Helix-Loop-Helix proteins generate complexes that bind specifically to a common DNA sequence
.
Cell
58
,
537
544
.
Nüsslein-Volhard
,
Ch
and
Wieschaus
,
E.
(
1980
).
Mutations affecting segment number and polarity in Drosophila
.
Nature
287
,
795
801
.
O’Kane
,
C. J.
and
Gehring
,
W. J.
(
1987
).
Detection in situ of genomic regulatory elements in Drosophila
.
Proc. Natl. Acad. Sci. U.S.A
.
84
,
9123
9127
.
Patel
,
N. H.
,
Schafer
,
B.
,
Goodman
,
C. S.
and
Holmgren
,
R.
(
1989
).
The role of segment polarity genes during Drosophila neurogenesis
.
Genes and Development
3
,
890
904
.
Poulson
,
D. F.
(
1937
).
Chromosomal deficiencies and the embryonic development of Drosophila melanogaster
.
Proc. Natl. Acad. Sci. USA
.
23
,
133
137
.
Prokop
,
A.
and
Technau
,
G. M.
(
1991
).
The origin of postembryonic neuroblasts in the ventral nerve cord of Drosophila melanogaster
.
Development
111
,
79
88
.
Rao
,
Y.
,
Vaessin
,
H.
,
Jan
,
L. Y.
and
Jan
,
N. Y.
(
1991
).
Neuroectoderm in Drosophila embryos is dependent on the mesoderm for positioning but not for formation
.
Genes and Development
5
,
1577
1588
.
Romaní
,
S.
,
Campuzano
,
S.
and
Modolell
,
J.
(
1987
).
The achaete-scute complex is expressed in neurogenic regions of Drosophila embryos
.
EMBO J
.
6
,
2085
2092
.
Rosenberg
,
U. B.
,
Schröder
,
Ch
,
Preiss
,
A.
,
Kienlin
,
A.
,
Coté
,
S.
,
Riede
,
I.
and
Jäckle
,
H.
(
1986
).
Structural homology of the product of Drosophila Krüppel gene with Xenopus transcription factor IIIA
.
Nature
319
,
336
339
.
Shepherd
,
D.
and
Bate
,
C. M.
(
1990
).
Spatial and temporal patterns of neurogenesis in the embryo of the locust (Schistocerca gregaria)
.
Development
108
,
83
96
.
Siegfried
,
E.
,
Perkins
,
L. A.
,
Capaci
,
T. M.
and
Perrimon
,
N.
(
1990
).
Putative protein kinase product of the Drosophila segment-polarity gene zeste-white3
.
Nature
345
,
825
829
.
Stern
,
C.
(
1968
).
Genetic Mosaics and Other Essays
.
Cambridge, Mass
:
Harvard University Press
.
St. Johnston
,
D.
and
Nüsslein-Volhard
,
Ch
(
1992
).
The origin of pattern and polarity in the Drosophila embryo
.
Cell
68
,
201
219
.
Taghert
,
P. H.
,
Doe
,
Ch Q.
and
Goodman
,
C. S.
(
1984
).
Cell determination and regulation during development of neuroblasts and neurones in grasshopper embryo
.
Nature
307
,
163
165
.
Taghert
,
P. H.
and
Goodman
,
C. S.
(
1984
).
Cell determination and differentiation of identified serotonin immunoreactive neurones in the grasshopper embryo
.
J. Neuroscience
4
,
989
1000
.
Tautz
,
D.
,
Lehmann
,
R.
,
Schnürch
,
H.
,
Schuh
,
R.
,
Seifert
,
E.
,
Kienlin
,
A.
,
Jones
,
K.
and
Jäckle
,
H.
(
1987
).
Finger protein of novel structure encoded by hunchback, a second member of the gap class of Drosophila segmentation genes
.
Nature
327
,
383
389
.
Thomas
,
J. B.
,
Bastiani
,
M. J.
,
Bate
,
M.
and
Goodman
,
C. S.
(
1984
).
From grasshopper to Drosophila: a common plan for neuronal development
.
Nature
310
,
203
207
.
Truman
,
J. W.
and
Bate
,
M.
(
1988
).
Spatial and temporal patterns of neurogenesis in the central nervous system of Drosophila melanogaster
.
Dev. Biol
.
125
,
145
157
.
Vaessin
,
H.
,
Grell
,
E.
,
Wolff
,
E.
,
Bier
,
E.
,
Jan
,
L. Y.
and
Jan
,
N. Y.
(
1991
).
prospero is expressed in neuronal precursors and encodes a nuclear protein that is involved in the control of axonal outgrwth in Drosophila
.
Cell
67
,
941
953
.
van Doren
,
M.
,
Ellis
,
H. M.
and
Posakony
,
J. W.
(
1991
).
The Drosophila extramachrochaetae protein antagonizes sequence specific DNA binding by daughterless/achaete-scute protein complexes
.
Development
113
,
245
255
.
Villares
,
R.
and
Cabrera
,
C. V.
(
1987
).
The achaete-scute gene complex of D. melanogaster: conserved domains in a subset of genes required for neurogenesis and their homology to the myc proteins
.
Cell
50
,
415
424
.
Wharton
,
K. A.
,
Johansen
,
K. M.
,
Xu
,
T.
and
Artavanis-Tsakonas
,
S.
(
1985
).
Nucleotide sequence from the neurogenic locus Notch implies a gene product that shares homology with proteins containing EGF-like repeats
.
Cell
43
,
567
581
.
Wheeler
,
W. M.
(
1891
).
Neuroblasts in the arthropod embryo
.
J. Morphol
.
4
,
337
343
.
Wigglesworth
,
V. B.
(
1940
).
Local and general factors in the development of “pattern” in Rhodnius prolixces (Hemiptera)
.
J. Exp. Zool
.
17
,
180
200
.
Wilcox
,
M.
,
Mitchinson
,
G. J.
and
Smith
,
R. J.
(
1973
).
Pattern formation in the blue-green alga, Anabaena: I. Basic mechanisms
.
J. Cell. Sci
.
12
,
707
723