BIOLOGICAL OVERVIEW

Morphogenesis of the eye, involving as it does the action of dozens and perhaps hundreds of genes, provides a front row view of development in process. Of particular interest is the morphogenetic furrow, a band of differentiating cells which moves from posterior to anterior in the eye imaginal disc. Behind the furrow, elegantly organized eight cell structures called ommatidia form. These are the facets of the compound eye. In front of the furrow, changes are already taking place, induced by secreted proteins made by cells behind the furrow.

The art of producing the right number of ommatidia begins with proteins involved in intercell communication. Two neurogenic genes have an essential role. Scabrous is secreted and Delta is bound to the cell surface. They both help to insure that neighboring cells do not acquire the photoreceptor fate. Thus Delta and Scabrous must be supplied at the right time and in the correct amounts to assure that only 800 ommatidia are formed per eye. Scabrous is first to be enlisted because Delta and its receptor Notch, acting without Scabrous, would give rise to a disorganised pattern of ommatidia. The targets of Scabrous are R8 cells, the first of the eight cells comprising the ommatidia. All furrow cells pass through an R8-competent stage. Scabrous activity, and subsequently action by Notch and Delta, restricts R8 differentiation to alternating groups of cells and confers a regular pattern on the R8 cells.

Scabrous is probably a second ligand for Notch, preparing the way for the primary ligand Delta to fulfill its role in lateral inhibition, a process restricting the number of ommatidia in the developing eye. Either an overexpression of Scabrous or its deficiency causes the same phenotype, a rough eye with oversized and fused ommatidia. This suggests that the concentration of Scabrous is a critical determining element in ommatidial differentiation (Mlodzik, 1990 and Baker, 1995).

R8 photoreceptors are specified by a bHLH transcription factor, Atonal. The Epidermal growth factor receptor (Egfr) has a primary function in regulating R8 spacing. The receptor's activation within nascent ommatidia induces the expression of a secreted inhibitor that blocks atonal expression, and therefore ommatidial initiation, in nearby cells. The identity of the secreted inhibitor remains elusive but, contrary to previous suggestions, it has been shown that this inhibitor is not Argos. This Egfr-dependent inhibition acts in parallel to the inhibition of atonal by the secreted protein Scabrous. The activation of the Egfr pathway is dependent on Atonal function via the expression of Rhomboid-1. Therefore, it was concluded that Egfr's role in promoting cell survival is largely independent of its role in photoreceptor recruitment; even when cell death is blocked, most photoreceptors fail to form. Based on these data and those of others, a model is proposed for R8 spacing that comprises a self-organizing network of signaling molecules. This model describes how successive rows of ommatidia form out of phase with each other, leading to the hexagonal array of facets in the compound eye (Baonza, 2001).

R8 cell spacing is disrupted in clones of cells mutant for Egfr. However, cell death is substantially elevated in these clones, and therefore it is not possible to tell whether the spacing defects are a direct consequence of Egfr loss or are secondary to cell death. To examine this, Egfr^- clones were generated in a genetic background in which cell death was blocked in the eye by expressing the baculovirus p35 gene under the control of the eye-specific GMR enhancer. In these clones, R8 cells differentiate but their spacing is still disrupted. This result implies that the Egfr function in spacing is not secondary to cell death. The abnormal spacing is seen first as a failure of Atonal to become modulated into proneural clusters in the furrow; a broad band of fairly uniform Atonal is expressed until just posterior to the furrow. This eventually resolves into isolated Atonal-expressing cells that form a disorganized array. Since atonal expression does ultimately resolve to single cells despite the lack of proneural clusters, it is imagined that lateral inhibition mediated by Notch still occurs in the Egfr^- clones (Baonza, 2001).

The Drosophila Egfr signals principally through the Ras/MAPK signal transduction pathway. The observation that Egfr signaling has a direct role in spacing the proneural clusters in the eye imaginal disc is therefore consistent with the fact that MAPK activity within proneural clusters is necessary for the repression of atonal expression in cells between proneural clusters. The MAPK signal transduction pathway is activated by a wide range of receptor tyrosine kinases; therefore, tests were made to see whether Egfr is responsible for the observed MAPK activation in the furrow. In clones of cells carrying an Egfr null mutation the receptor is autonomously required for all MAPK activation. From this it is concluded that the Egfr is the only RTK that detectably activates MAPK in the morphogenetic furrow (Baonza, 2001).

Interestingly, clones lacking the Egfr ligand, Spitz, have normal R8 cell spacing and MAPK activation. This finding suggests that another ligand for the receptor may be responsible for this function. A single novel Spitz-like ligand has recently been discovered in the completed Drosophila genome sequence (Keren), and it is speculated that this could provide the missing function in R8 cell spacing. Testing this prediction awaits the identification of loss-of-function mutations in the spitz-2 gene (Baonza, 2001).

The results described above imply that Egfr has a primary function in ommatidial spacing and suggest that it is the only RTK that activates MAPK in the proneural clusters. Since it has been shown that the transcription factor Atonal is also required for this MAPK activation, how Atonal and Egfr activation are related was investigated. One possibility is that Atonal directly activates the expression of rhomboid-1, a principal activator of Egfr signaling, as it does in the embryonic chordotonal organs. Therefore whether ectopic Atonal can activate rhomboid-1 expression was investigated. In wild-type cells, rhomboid-1-lacZ is expressed only in photoreceptors R8, R2 and R5. UAS-atonal was expressed under the control of sevenless-Gal4, which is expressed in all ommatidial cells except R8, R2 and R5. When atonal is thus misexpressed, it was found that rhomboid-1 expression (as detected by rhomboid-1-lacZ) is activated in ectopic photoreceptor cells. Thus, consistent with the model in which the activation of MAPK via Atonal depends on the activation of rhomboid-1 expression, atonal expression can induce rhomboid-1 expression, which in turn activates Egfr (Baonza, 2001).

In some tissues, the expression of the Egfr activator, Rhomboid-1, is dependent on Egfr signaling itself. This dependency thereby constitutes a positive-feedback loop. If this were the case in the furrow, the Atonal-triggered expression of rhomboid-1 could not initiate Egfr signaling (as it would itself depend on prior signaling). Therefore, the expression of rhomboid-1-lacZ in Egfr^- clones was examined. Since both the clone marker and the detector for rhomboid-1 expression are the lacZ gene, both are labeled in the same color. However, the ß-galactosidase that marks the clone is cytoplasmic, whereas the one that indicates rhomboid-1 expression is confined to the nucleus. In this way the rhomboid-1-expressing cells can be distinguished from the Egfr-positive cells; this is particularly obvious in transverse optical sections of the eye disc. Egfr^- cells can initiate rhomboid-1 expression, and it is concluded that the initiation of rhomboid-1 expression in the furrow does not require Egfr activity (Baonza, 2001).

The results described so far indicate that within proneural clusters, Atonal activates the expression of rhomboid-1, which in turn leads to the activation of the Egfr/Ras/MAPK pathway. This MAPK activity in proneural cells leads to a nonautonomous inhibition of atonal expression in the cells between proneural clusters. Scabrous is a secreted protein, expressed within clusters, that is also required for the inhibition of atonal expression between clusters. A possible prediction of the model is that scabrous expression in proneural cluster cells would be activated by Egfr/Ras/MAPK signaling -- in other words, that Scabrous is the inhibitory signal secreted by cluster cells in response to Egfr signaling. To test this, Scabrous expression was analyzed in Egfr^- clones by using a monoclonal antibody against the Scabrous protein. Normal levels of Scabrous were observed in Egfr^- clones, and this finding implies that Scabrous expression is not dependent on the Egfr pathway. The pattern of Scabrous expression was nevertheless altered, which reflects the abnormal spacing of cells in the furrow in Egfr^- clones (Baonza, 2001).

This result suggests that the inhibitory factor regulated by Egfr signaling works in parallel to Scabrous in repressing atonal between proneural clusters. A consequent prediction is that when both Scabrous and Egfr signaling are removed, the spacing defects in the furrow should be worse than those caused by either mutation alone. Conversely, if the Egfr-dependent inhibition is mediated by Scabrous, the double mutants should have the same phenotype as the single mutants. Complete loss of Scabrous alone causes a relatively mild defect in spacing. Clones doubly mutant for Egfr and scabrous were examined and they have reproducibly more severe spacing defects than do Egfr mutant clones alone. When the cells were stained with anti-Boss to label the R8 cells, this was particularly clear within the morphogenetic furrow; in Egfr^- clones the spacing is irregular, but the overall number or R8s is not substantially increased over that of the wild type, while in Egfr^-;sca^- double-mutant clones, more R8s form in the furrow and typically produce a very closely spaced row of cells that is not seen in the single-mutant clones. This additive effect of removing Egfr and Scabrous supports the notion that they mediate two parallel pathways: each contributes to the inhibition of atonal expression between proneural clusters (Baonza, 2001).

It has been proposed that the secreted Egfr antagonist, Argos, could be the inhibitor of R8 determination between preclusters. This suggestion has been directly tested in argos loss-of-function clones. The arrangement and spacing of the developing ommatidia are completely normal, even in very large clones induced in a minute background; some examples cover more than half of the eye disc. This result implies that Argos cannot be significantly involved in regulating ommatidial spacing. Consistent with this result, it has previously been shown that whole eyes mutant for eye-specific argos alleles do not have substantially disrupted precluster spacing (Baonza, 2001).

A fairly simple model can be proposed for how R8 spacing is controlled; this model synthesizes the work of several groups. A key feature is that it is a self-organizing system; once atonal expression is initiated at the posterior of the disc, the pattern spreads across the whole retinal primordium without further input from signals other than those generated by the spacing mechanism itself. The first stage of ommatidial determination is the activation of a broad, uniform band of atonal expression anterior to the morphogenetic furrow. This is initiated at least in part by the secreted protein Hedgehog, which emanates from the more posterior, already differentiating, ommatidia. In the model, this band of Atonal expression becomes modulated by the combined action of two diffusible inhibitors, Scabrous and an unidentified inhibitory factor dependent on Egfr-induced signaling through the MAPK pathway, both of which are dependent on Atonal (Baonza, 2001 and references therein).

atonal expression is upregulated by an autoregulatory loop, just as the proneural clusters become apparent. It is at this point that it is proposed that Scabrous and the Egfr-dependent inhibitory factor act. They diffuse toward the anterior and inhibit atonal expression in those cells closest to the inhibitory source. By this mechanism, only the cells farthest from clusters in the previous row retain atonal expression. This produces the characteristic staggered arrangement of R8s in successive rows. Therefore, the central patterning event in establishing the overall arrangement of the ommatidia is the transformation of uniform Atonal expression into modulated expression, as controlled by a combination of Scabrous and the Egfr-dependent inhibitory signal (Baonza, 2001).

Once Atonal expression is initially modulated by these inhibitory factors, well-defined proneural clusters are formed by a combination of the same inhibitory signals and the autoregulatory positive feedback loop that maintains and increases atonal expression within the clusters. It is suggested that this autoregulation makes the proneural cluster cells refractory to the inhibitory signals they themselves are producing (Baonza, 2001).

There is no obvious candidate for either the Egfr-dependent inhibitory factor or for the signaling pathway it uses. It can be inferred, however, that it triggers the expression of the homeodomain protein Rough, since Rough expression is lost in Egfr^- clones. Rough is a transcription factor that represses atonal expression and Rough is normally expressed in a complementary pattern to Atonal within the furrow. Rough expression is not affected by the loss of Scabrous, which is consistent with the idea that the Rough-mediated inhibition of atonal expression is instead controlled by the Egfr-dependent inhibitory factor. It was originally suspected that Scabrous, which regulates Notch signaling, could be the inhibitory factor. As described above, coupled with the phenotype of scabrous mutants alone, the results on Scabrous expression and the scabrous, Egfr double mutations imply that this is not the case. Nevertheless the possibility that Notch activity has a role in regulating precluster spacing has not been ruled out; it is clearly involved in the later process of lateral inhibition that inhibits atonal expression in all but one of the proneural clusters, so its function in nonautonomous atonal inhibition is well established. Despite this possibility, current evidence does not provide a convincing link between the Egfr-dependent inhibition and the Notch pathway (Baonza, 2001).

A neural progenitor mitotic wave is required for asynchronous axon outgrowth and morphology

Spatiotemporal mechanisms generating neural diversity are fundamental for understanding neural processes. This study investigated how neural diversity arises from neurons coming from identical progenitors. In the dorsal thorax of Drosophila, rows of mechanosensory organs originate from the division of sensory organ progenitor (SOPs). In each row of the notum, an anteromedial located central SOP divides first, then neighbouring SOPs divide, and so on. This centrifugal wave of mitoses depends on cell-cell inhibitory interactions mediated by SOP cytoplasmic protrusions and Scabrous, a secreted protein interacting with the Delta/Notch complex. Furthermore, when this mitotic wave was reduced, axonal growth was more synchronous, axonal terminals had a complex branching pattern and fly behaviour was impaired. The temporal order of progenitor divisions influences the birth order of sensory neurons, axon branching and impact on grooming behaviour. These data support the idea that developmental timing controls axon wiring neural diversity (Lacoste, 2022).

To study how functional neuronal diversity can be generated from a homogenous set of neural precursors, advantage was taken of the invariant way in which sensory organs are located on the dorsal epithelium of Drosophila. This spatial configuration greatly facilitated the study of the relative timing of SOP division and the identification of a distinct temporal wave of SOP mitosis. Asynchrony in mitotic reactivation timing has been described in Drosophila larva neuroblasts. This differential timing is related to two cell cycle arrests: one population of neuroblasts is arrested in G2 while another population is arrested in G0. G2-arrested neuroblasts resume mitosis earlier than those in G0-arrest. As in this system, it has been proposed that this particular order of division ensures that neurons form appropriate functional wiring. It is relevant that other temporal processes controlling the wiring of peripheral receptors with the central nervous system have been described in the Drosophila eye, another highly organised structure. It is conceivable that these temporal patterning mechanisms of neurogenesis, to date identified only in organised tissues, could be more widespread (Lacoste, 2022).

A core aspect of this work was to link cellular level of complexity (timing of SOP division) with uppermost level (behaviour). In this context, evidences are presented showing that the cleaning reflex was impaired when the SOP mitotic wave was disrupted. The cleaning reflex has been traditionally analysed after stimulation of macrochaetes rather than microchaetes as in the present work. Macro- and microchaetes have different patterns of terminal axon arborisation. As such, it is remarkable that this fly behaviour was significantly affected by altering the timing of microchaete precursor division in the dorsal thorax. This study showed that the SOP mitotic wave leads to a progressive neurogenesis along each row of microchaetes. This, in turn, would likely induce a particular pattern of microchæte axon arrival in the thoracic ganglion required for the proper organisation of the neuropila in the central nervous system. Although this study has documented this progressive axonogenesis, the strict pattern of axon arrival into the ventral ganglion is not known. It would depend on the order of birth of neurons, and on the geometry of axon projections that fasciculate to form the dorsal mesothoracic nerves in the ganglion. In any case, this study shows that, when sca function was specifically downregulated during the SOP mitotic wave, axonogenesis occurs almost simultaneously in each row of microchaetes. This certainly impairs the pattern of axon arrival into the ganglion leading to ectopic axon branching and changes in fly behaviour. It would be interesting to know whether these impairments are specifically due to neurogenesis occurring simultaneously. To test this, it is necessary to find a way to induce different patterns of SOP mitotic entry, for instance, a centripetal wave or a random order. If the observed effect is specifically due to the simultaneity, normal behaviour would be expected to be associated with other patterns of SOP division (Lacoste, 2022).

We observed that the first SOP to divide (SOP0) was always located in the anteromedial region of each row. This may reflect the existence of a pre-pattern that causes SOPs located in that region to start dividing earlier than the others. Although the anteromedial region corresponds approximately to the posterior limit of expression of the transcription factor BarH1, no factors specifically expressed in this region have yet been identified. Alternatively, as the location of SOP0 is modified when Sca function was impaired, an interesting possibility is that SOP0 is selected by an emergent process related to cell-cell interaction in the epithelium, rather than by a passive pre-pattern that organises the first events in the notum (Lacoste, 2022).

This study presents evidence indicating that the secreted glycoprotein Scabrous, which is known to interact with the N-pathway to promote neural patterning, controls the kinetics of SOP mitosis in the notum. In proneural clusters, cells that express high levels of Dl and Sca become SOPs, while surrounding epithelial cells activate the N-pathway to prevent acquisition of a neural fate. In eye and notum systems, Sca modulates N-activity at a long range. Indeed, during eye development, sca is expressed in intermediate clusters in the morphogenic furrow and transported posteriorly in vesicles through cellular protrusions to negatively control ommatidial cluster rotation. Similarly, in the notum, SOP protrusions extend beyond several adjacent epithelial cells in which Dl and Scabrous are detected. The current data show that shorter protrusions (obtained after rac1^N17 overexpression conditions) as well as loss of function of Dl or sca make the mitotic wave more synchronous. Since, no reduction of the global level of sca expression associated with the wave progression was observed, it is plausible that Sca, required to maintain SOPs in G2 arrest, is delivered focally through protrusions that are difficult to follow with in vivo analysis. Although this possibility is favored, it cannot be formally ruled out that Rac1^N17 overexpression affects Sca secretion per se without affects sca expression (Lacoste, 2022).

As in neuroblasts, G2 arrest in SOP cells is due to the downregulation of the promitotic factor Cdc25/String. Thus, overexpression of string in SOPs induces a premature entry into mitosis, while overexpression of negative regulators, like Wee1, maintain these cells in arrest. Possibly Sca negatively regulates string expression, perhaps through the N-pathway that it is known to control the level of String. Alternatively, it has been recently shown that the insulin-pathway also regulates String level. Moreover, in muscle precursors, cell proliferation is induced by the insulin-mediated activation of the N-pathway. These observations raise the interesting possibility that, in this system, insulin activates the N-pathway and Sca modulates this activation. Further investigations will be required in order to identify the link between Scabrous, the N/Dl- and insulin-pathways in the resumption of mitosis in SOPs (Lacoste, 2022).

During nervous system development, the complex patterns of neuronal wiring are achieved through the interaction between neuronal cell surface receptors and their chemoattractive or repulsive ligands present in the environment. An essential condition for proper axon guidance is the competence of neurons to respond to these environmental clues. It is generally agreed that neuron competence depends on the specific expression of transcriptional factors regulating their identity. This study shows that the timing of neuron formation is also a factor controlling their terminal morphology. It is proposed that the SOP mitotic wave induces a particular pattern of arrival of microchæte axons in the thoracic ganglion. This pattern establishes a specific framework of guidance cues on which circuits will be built and ultimately influencing an organism's behaviour. These findings support the idea that, in addition to genetic factors, neurogenic timing is a parameter of development in the mechanisms controlling neural branching (Lacoste, 2022).

A neural progenitor mitotic wave is required for asynchronous axon outgrowth and morphology

Spatiotemporal mechanisms generating neural diversity are fundamental for understanding neural processes. This study investigated how neural diversity arises from neurons coming from identical progenitors. In the dorsal thorax of Drosophila, rows of mechanosensory organs originate from the division of sensory organ progenitor (SOPs). In each row of the notum, an anteromedial located central SOP divides first, then neighbouring SOPs divide, and so on. This centrifugal wave of mitoses depends on cell-cell inhibitory interactions mediated by SOP cytoplasmic protrusions and Scabrous, a secreted protein interacting with the Delta/Notch complex. Furthermore, when this mitotic wave was reduced, axonal growth was more synchronous, axonal terminals had a complex branching pattern and fly behaviour was impaired. The temporal order of progenitor divisions influences the birth order of sensory neurons, axon branching and impact on grooming behaviour. These data support the idea that developmental timing controls axon wiring neural diversity (Lacoste, 2022).

Size of gene - 12 kb

cDNA clone length - 3 kb

Bases in 5' UTR -376

Exons - 4

PROTEIN STRUCTURE

Amino Acids - 774

Structural Domains and Evolutionary Homologies

Scabrous is a secreted protein with similarity to the beta and gamma chains of the fibinogens in the carboxy-terminal protions (Baker, 1990 and Mlodzik, 1990). Scabrous is not expected to polymerize, since it lacks crosslinking regions of fibrinogen (Baker, 1990). The fibrinogen domain also has homology to a globular domain found in tenascin.

Mutations at the scabrous locus affect cell-cell signaling during neural development. Twenty-one mutant alleles of scabrous have been analyzed; many synthesize no SCA protein. In others, a defective protein is arrested intracellularly. Two mutants in which protein is not arrested must affect SCA protein function outside the cell. Both affect the fibrinogen related domain (FReD), a 200-amino acid segment conserved in fibrinogen, tenascins, and other proteins. In fibrinogen, this region is involved in protein interactions and is altered in human mutations affecting blood clotting. In sca(UM2), an invariant Asp residue is replaced by Asn. The sca(MSKF) allele has dominant negative properties, indicating that the truncated amino-terminal portion interferes with the function of some other gene product. These mutations show that the conserved FReD is essential for wild-type SCA function, but suggest that the amino-terminal domain also interacts with other proteins, although other neural mutations are without effect (Hu, 1995).

Promoter Structure

scabrous has four E-boxes in its proximal promoter. These are thought to be targets of achaete-scute proneural genes (Singson, 1994).

Scabrous is a target for Ultrabithorax. Parts of the last intron and exon of the scabrous gene contain five ATTA sequences, the core sequence shared by most homeodomain binding sites. Mutation of Ubx results in the ectopic transcription of sca in the first abdominal segment. Transcript localization in several combinations of deficiencies in the bithorax complex indicates that sca is downregulated by abdominal A and Abdominal B. It appears to be a common target for the three genes of BX-C (Graba, 1992).

Transcriptional Regulation

Hedgehog acts upstream of glass, scabrous, hairy and decapentaplegic in the developing eye (Ma, 1993). The activator of scabrous in the eye optic disc may be Atonal, a bHLH transcription factor that heterodimerizes with Daughterless and binds to E-boxes (Jarman, 1993). In the wing disc, scabrous is controlled by genes of the achaete-scute complex (Mlodzik, 1990).

Response to the insect hormone ecdysone is mediated by a nuclear receptor complex containing Ultraspiracle (Usp) and the Ecdysone Receptor (EcR). Among other phenotypes, loss of functional Usp in Drosophila eye development results in an accelerated morphogenetic furrow, although loss of ecdysone arrests the furrow. Usp both represses and activates a gene affecting furrow movement, the ecdysone-responsive Z1 isoform of Broad-Complex. Using targeted replacement of Usp to rescue usp mutant clones in the eye, various USP functions have been mapped and whether the USP nuclear receptor has an activating as well as a repressive effect on furrow movement has been tested. Furrow movement and related phenotypes are rescued by the presence of Usp in a limited domain near the furrow, while other phenotypes are rescued by Usp expression posterior to the furrow. These data indicate roles for Usp activity at multiple developmental stages and help explain why loss of functional Usp leads to furrow advancement while loss of ecdysone stops furrow movement (Ghbeish, 2002).

Loss of functional Usp affects multiple genes involved in cell determination in the eye such as scabrous, cut and tramtrack as well as neuronal seven-up, elav and spalt. For most of these markers, expression occurs prematurely, although cellular differentiation appears to occur normally. For example, in usp mutant regions, the Cut protein is first expressed toward the posterior of the third instar eye disc, while in wild-type eye discs Cut expression in the same region is seen later during pre-pupal development. In the wing, multiple markers of various stages of bristle complex determination and differentiation, such as Elav, Cut, ttk-lacZ and Cyclin B, are expressed prematurely at the site of the future triple row of bristles. These data suggest that the usp gene behaves as a timer causing cellular differentiation to occur at the proper pace in imaginal disc development (Ghbeish, 2002).

Scabrous protein misexpression posterior to the furrow in usp^- clones appears to be an exception to the model of premature but normal differentiation in usp mutant clones. The sca gene is required for spacing and specification of the R8 photoreceptors and expression associated with this function is normal in usp^- tissue. In contrast, usp mutant regions show Sca positive cells posterior to the furrow. No Sca expression is seen in these regions in wild-type larval or pre-pupal discs, and no additional roles in eye development have been reported for sca. Thus, this phenotype appears to be true misexpression rather than premature expression. It remains possible that ttk-lacZ expression in usp mutant clones represents ectopic as well as premature expression, suggesting increased numbers of differentiating cone cells. However, examination of Cut expression suggests merely premature differentiation, not extra cone cells (Ghbeish, 2002).

usp mutant phenotypes occur at multiple regions of the eye disc. Using targeted Usp protein in usp mutant clones, various usp mutant phenotypes could be rescued and the different regions in which Usp protein is needed have been mapped. The initiation of Usp protein expression in a limited domain near or in the furrow is able to rescue a series of phenotypes, some of which are revealed posterior to the furrow. These include premature furrow advancement and neuronal differentiation, as well as disarrangement of the ommatidial clusters. Usp protein expression initiating posterior to the furrow will not rescue these phenotypes. Other functions, such as maintaining the appropriate number of Cyclin B positive cells, and posterior repression of ttk-lacZ, cut, and sca can be rescued by targeting Usp protein posterior to the furrow (Ghbeish, 2002).

This and previous studies clearly show that the ultraspiracle gene functions as a repressor of differentiation in both the eye and wing imaginal discs. Loss of functional Usp results in the premature differentiation of multiple cell types including the photoreceptors and the cone cells in the eye. Furthermore, Usp represses the ectopic expression of at least one gene, scabrous. This repressive function may or may not be due to direct activity of the Usp nuclear receptor on target genes either as part of the EcR complex, alone, or as part of another nuclear receptor complex (Ghbeish, 2002).

Split ends, acting through Scabrous, antagonizes the Notch and potentiates the EGFR signaling pathways during Drosophila eye development

The Notch and Epidermal Growth Factor Receptor (EGFR) signaling pathways interact cooperatively and antagonistically to regulate many aspects of Drosophila development, including the eye. How output from these two signaling networks is fine-tuned to achieve the precise balance needed for specific inductive interactions and patterning events remains an open and important question. The gene split ends (spen) functions within or parallel to the EGFR pathway during midline glial cell development in the embryonic central nervous system. This study shows that the cellular defects caused by loss of spen function in the developing eye imaginal disc place spen as both an antagonist of the Notch pathway and a positive contributor to EGFR signaling during retinal cell differentiation. Specifically, loss of spen results in broadened expression of Scabrous, ectopic activation of Notch signaling, and a corresponding reduction in Atonal expression at the morphogenetic furrow. Consistent with Spen's role in antagonizing Notch signaling, reduction of spen levels is sufficient to suppress Notch-dependent phenotypes. At least in part due to loss of Spen-dependent down-regulation of Notch signaling, loss of spen also dampens EGFR signaling as evidenced by reduced activity of MAP kinase (MAPK). This reduced MAPK activity in turn leads to a failure to limit expression of the EGFR pathway antagonist and the ETS-domain transcriptional repressor Yan and to a corresponding loss of cell fate specification in spen mutant ommatidia. It is proposed that Spen plays a role in modulating output from the Notch and EGFR pathways to ensure appropriate patterning during eye development (Doroquez, 2007).

This study demonstrates that loss of spen perturbs the normal balance between the EGFR and Notch pathways as evidenced by the patterning disruptions and aberrant expression of multiple pathway components. These findings raise the question of whether Spen functions primarily in the Notch pathway, primarily in the EGFR pathway, or as a critical component of both. Although definite resolution is difficult given the extensive and intricate feedback regulation within and between these two signaling networks, a model is proposed in which Spen-mediated antagonism of the Notch pathway regulates the signaling flow through the EGFR pathway to achieve proper retinal cell fate specification (Doroquez, 2007).

Loss of spen results in hyperactivation of the Notch pathway as evidenced by elevated levels of both Notch and its transcriptional targets, the E(spl)-bHLHs. Therefore, a normal function of Spen in the developing eye is to limit the activity of Notch. Consistent with this model, heterozygous reduction of spen was shown to be sufficient to suppress the heterozygous Notch wing margin phenotype. However, loss of spen does not lead to the anti-neurogenic phenotypes typically associated with overexpression/overactivation of canonical members of the Notch pathway, suggesting that although Notch signaling output is elevated, the increase is below the threshold needed to achieve such phenotypes. Consistent with this interpretation, recruitment of the initial R8 photoreceptor neuron, a process influenced at multiple stages by Notch signaling, occurs normally in the absence of spen (Doroquez, 2007).

Where might Spen interface with the Notch signaling pathway? The striking increase in Sca expression in spen mutant clones at the MF is consistent with Spen regulating Notch activation by limiting the expression of sca either through transcriptional repression or by destabilizing the transcript. This suggests that in the Drosophila eye Spen may have an upstream role in the Notch pathway in contrast to the downstream role described for Spen mammalian orthologs. In contrast, because of extensive feedback regulation in Notch signaling, it is plausible that Spen interfaces with the network at a more downstream point. For example, ectopic expression of Notch^intra was shown to promote Sca expression, which in turn activates Notch signaling. Additionally, although no such role was detected with respect to yan, it is possible that Spen limits Notch^intra/Su(H)-mediated transactivation at the level of transcriptional repression of other Notch pathway targets, including the E(spl)-bHLHs, as is the case for the mammalian Spen orthologs. This latter mechanism might also be relevant posterior the MF, where Notch signaling remains elevated as judged by increased levels of both Notch and the E(spl)-bHLHs in spen mutant clones, but where Sca is no longer expressed (Doroquez, 2007).

Although pinpointing where Spen interfaces with the Notch signaling pathway remains a challenge, the simplest interpretation of the data is that at the MF, Spen either directly or indirectly regulates Sca expression to restrict Notch pathway output. Posterior to the MF, as discussed below, mutual antagonism between the Notch and EGFR pathways may stabilize the initial signaling imbalance independent of Sca, leading to sustained up-regulation of Notch and down-regulation of EGFR output in spen mutant tissue (Doroquez, 2007).

What might the consequences of a moderate increase in Notch pathway output be? Given the extensive functional antagonism that has been reported between the Notch and EGFR pathways in the eye, a likely outcome is that the increased Notch signaling in spen mutant tissue would dampen EGFR pathway output. Supporting the idea that spen plays a positive role with respect to EGFR signaling, the cell fate specification defects observed in spen mutant clones are highly reminiscent of phenotypes associated with hypomorphic mutants in positive components of the EGFR pathway. Thus, the defective specification of neuronal and non-neuronal cell types and the perturbed R8 spacing adjacent to the MF all suggest reduced, but not ablated, EGFR pathway function in spen clones (Doroquez, 2007).

Lending further support to a model in which elevated Notch signaling in spen clones dampens EGFR pathway output, both Ato and dpERK expression at the MF are reduced. Because previous work has shown that Ato is required for activation of the EGFR pathway at the MF, one possibility is that Spen stabilizes dpERK levels at the MF by antagonizing Notch-mediated lateral inhibition to ensure appropriate Ato expression. Another plausible mechanism for Spen-mediated regulation of dpERK activity would be downstream of or in parallel to Ras. In this scenario, Spen might mediate transcriptional repression of an inhibitor such as a MAPK phosphatase. However, qRT-PCR analysis in imaginal discs predominantly mutant for spen do not indicate a role for Spen in regulating the expression of two characterized Drosophila MAPK phosphatases - dMKP3 and PTP-ER. Thus, validation of such a mechanism will require identification of other MAPK phosphatases or pathway inhibitors that might be regulated by Spen (Doroquez, 2007).

It should be noted that the results of this analysis of spen function in the eye appear contradictory to those from a prior study that suggested spen antagonizes EGFR output and promotes Notch signaling during embryonic neural development (Kuang, 2000). Specifically, elevated EGFR signaling was reported in spen maternal/zygotic null embryos, as evidenced by increased numbers of midline glial cells and loss of Yan expression. However, these results could not be reproduced (F. Chen and I. Rebay, unpublished data reported in Doroquez, 2007). On the contrary, analysis of spen function during midline glial cell development in the embryonic central nervous system was entirely consistent with a role for spen as a positive contributor to EGFR signaling. Thus, at least with respect to EGFR signaling, it is believed that Spen serves an analogous role in multiple developing tissues (Doroquez, 2007).

With respect to Notch signaling, Kuang reported a strong reduction in E(spl)-bHLH expression throughout the embryo but no change in Notch levels, exactly opposite to the current findings in the eye. Additional work will be needed to determine whether and how spen interfaces with the Notch pathway during embryogenesis, and whether distinct or identical mechanisms operate in retinal versus embryonic neural development (Doroquez, 2007).

It is not yet clear whether spen's role in Notch-EGFR interactions posterior to the MF is identical to its role in events occurring at the MF. The failure to down-regulate Yan and the resulting cell fate specification defects show that EGFR signaling posterior to the MF is compromised in spen mutant tissue. Given that Yan up-regulation in spen clones does not result from loss of Spen-mediated transcriptional repression, but rather reflects loss of post-translational control, two models for Spen function seem likely. First, if the inability to detect changes in dpERK protein levels posterior to the MF in situ accurately indicates unaltered dpERK levels, then the ability of dpERK to phosphorylate Yan must be compromised in spen mutants. Alternatively, dpERK levels may be sufficiently reduced to increase Yan stability, but the change may be below the immunohistochemical detection threshold (Doroquez, 2007).

In terms of the signals that impinge on dpERK, whereas Notch signaling and Ato expression are critical for proper dpERK expression in the MF, reiterative EGFR signaling takes over posterior to the MF to maintain dpERK activity. Thus, it is possible that a Spen-dependent, Notch-independent mechanism may regulate EGFR output posterior to the MF. Alternatively, because Notch, E(spl) and Yan expression are all elevated in spen mutant tissue both in and posterior to the MF, Spen-mediated antagonism of Notch signaling may be relevant to EGFR regulation in both contexts. An extension of this idea that results in perhaps the most appealing model is that the initial increase in Notch output at the MF dampens EGFR signaling, which in turn leads to elevated Notch signaling in more posterior regions resulting in reduced EGFR output. In this way, the initial signaling imbalance created by loss of spen at the MF could be maintained over the entire eye disc through mutual antagonism and feedback regulation between the Notch and EGFR pathways (Doroquez, 2007).

In summary, this study analyzed the requirement for spen in regulating the EGFR and Notch pathways during Drosophila eye development, and it is proposed that increased Notch pathway activity upon loss of spen may be sufficient to dampen EGFR signaling, but not to disrupt other downstream effects of Notch signaling. Therefore, because the effects of spen loss appear to be at a threshold below the production of bona fide Notch-related phenotypes, it is suggested that Spen plays a subtle role in the regulation of the Notch pathway or functions redundantly alongside other components. An equally likely hypothesis is that Spen regulates the Notch and EGFR pathways separately and that the phenotypes reported reflect a composite of independent disruptions to both signaling networks (Doroquez, 2007).

Although much of the literature focuses on a primary role for Spen family proteins as co-repressors, recent findings suggest members of this family may also regulate non-coding RNA sequestration, mRNA export, RNA splicing, and proteolysis. Therefore, future identification of the precise molecular mechanisms by which Spen interfaces with the EGFR and Notch pathways may reveal novel modes of interaction between these two critical and conserved signaling networks (Doroquez, 2007).

Targets of activity

Scabrous is dimeric glycoprotein that is secreted and found in soluble form in the tissue culture medium. SCA contains both N- and O-linked carbohydrates and interacts with heparin. sca mutations, along with conditional alleles of Notch and Delta, each affect the pattern of cells expressing atonal, the proneural gene required for R8 differentiation. In normal development, about 1 cell in 20 differentiates into an R8 cell; in the other cells, ato is repressed. N and Dl are required to repress ato in the vicinity of R8 cells, whereas scaproduces effects over several cell diameters. Certain antibodies detect uptake of SCA protein several cells away from its source. The overall growth factor-like structure of SCA protein, its solubility, and its range of effects in vivo are consistent with a diffusible role that complements mechanisms involving direct cell contact. It is proposed that as the morphogenic furrow advances, cell secreted SCA protein controls the pattern of the next ommatidial column (Lee, 1996a).

Protein Interactions

Scabrous is likely to act as a ligand for Notch (Baker, 1995).

Genetic interactions suggest that Canoe, a novel protein containing a GLGF/DHR, motif participates with members of the Notch pathway, including Scabrous, in regulating adhesive cell-cell interactions for the determination of cell fate (Miyamoto, 1995).

It has been suggested that the secreted protein Scabrous (Sca) might be a Notch ligand acting in the peripheral nervous system. The Sca protein was purified and a cell line expressing 18,000 Notch molecules per cell surface was used to test Sca binding by coimmunoprecipitation, cell adhesion assays, and binding with labeled Sca. No interaction was detected between gp300sca and Notch or the related protein Delta, suggesting that Sca acts through a distinct mechanism (Lee, 1996b).

The mechanisms that establish and sharpen pattern across epithelia are poorly understood. In the developing nervous system, the first pattern elements appear as 'proneural clusters'. In the morphogenetic furrow of the immature Drosophila retina, proneural clusters emerge in a wave as a patterned array of 6 to 10 cell groups, which are recognizable by their expression of Atonal, a basic helix-loop-helix transcription factor that is required to establish and pattern the first cell fate. The establishment and subsequent patterning of Atonal expression requires activity of the signaling transmembrane receptor Notch. In vivo and biochemical evidence is presented that the secreted protein Scabrous associates with Notch, and can stabilize Notch protein at the surface. The result is a regulation of Notch activity that sharpens proneural cluster boundaries and ensures establishment of single pioneer neurons (Powell, 2001).

In the morphogenetic furrow, Atonal's expression can be divided into four steps:(1) it is expressed as a broad, unpatterned stripe; (2) expression is then upregulated into evenly spaced proneural clusters;(3) in these proneural clusters, a 2 to 3 cell 'R8 equivalence group' emerges, and (4) expression narrows to identify a single cell in this group as the R8 photoreceptor neuron, the first cell type of the developing retina. In each step, as cells lose Atonal expression they concurrently gain expression of negative regulators, such as members of the E(spl) (Enhancer of Split) complex. Expression of E(spl) initially requires the presence of Atonal, and is subsequently amplified by the Notch signaling pathway to downregulate proneural bHLH expression and function. E(spl), therefore, represents one reporter of Notch activity (Powell, 2001).

Patterning of Atonal and E(spl) expression requires the normal activity of Scabrous, a secreted fibrinogen-related protein with a potential for association with components of the extracellular matrix. In the retina, Scabrous protein first appears in the proneural clusters, mirroring Atonal expression by narrowing to the R8 equivalence group, and eventually R8 alone. This expression is dependent on Atonal activity, which indicates that the scabrous locus may be a direct target of Atonal (Powell, 2001).

Genotypically null sca^BP2 proneural clusters are poorly spaced with poorly defined borders. Broadened E(spl) expression throughout much of the proneural cluster region is one potential cause of this imprecision, suggesting that Notch activity is altered in sca ^BP2 mutants. These observations suggest that initial broad, low-level Atonal expression activates broad, low-level E(spl) expression, and that Scabrous is required to refine the complementary Atonal and E(spl) expression in the proneural cluster region -- events also associated with Notch activity (Powell, 2001).

A 1 hour pulse of ectopic Scabrous results in rapid loss of Atonal within 2 h; E(spl) shows low, diffuse expression that is lost within 4 h. This ectopic expression of Scabrous leads to aberrant patterning of R8s in a manner similar to that of the phenotypes observed in scabrous loss-of-function alleles. The loss of the initial broad stripe of Atonal, a Notch-dependent step, suggests that ectopic Scabrous can lead to a disruption of Notch function, a result consistent with overexpression studies in the Drosophila wing (Powell, 2001).

Subsequent resolution of Atonal expression in the equivalence group, also a Notch-dependent step, is delayed in sca^BP2 mutants. Delay of Atonal resolution can result in ectopic R8s; indeed, 40% of sca^BP2 ommatidia contain two or three R8s derived from the R8 equivalence group. R8s were observed to be both adjacent and non-adjacent, indicating that selection may be stochastic (Powell, 2001).

These results suggest that Scabrous influences the establishment or maintenance of boundaries of Notch activity during both proneural-cluster and R8 equivalence-group maturation. To investigate the nature of this influence, transient expression of full-length Notch protein (N350) in Drosophila S2 tissue-culture cells was examined by Western analysis. Transiently induced N350 is depleted gradually over 7 h; this loss is accelerated in the presence of Delta. In contrast, addition of Scabrous leads to an accumulation of N350 and smaller Notch fragments, even in the presence of Delta. Accumulation of Notch is not due to induction of the endogenous Notch gene, which is rearranged in S2 cells. Cell-surface biotinylation analysis indicates that the presence of Scabrous significantly stabilizes N350 at the cell surface; however, the functional nature of this stabilization remains unknown (Powell, 2001).

Genetic and histological studies have suggested a function for Scabrous as a secreted Notch ligand. However, attempts to show this interaction biochemically have been unsuccessful. This issue was reexamined using two independent assays. Notch is expressed throughout embryogenesis, whereas Scabrous protein is detected in young (2.5-6.5 h) but not more mature (12.5-16.5 h) embryos. Immunoprecipitation with a mouse polyclonal anti-Scabrous antibody leads to a complex with a relative molecular mass of 300,000 (M_r 300K) that contains both Scabrous and Notch protein. The size and composition of the 300K complex indicates that it may include a truncated form of Notch previously characterized in vivo. As controls, crosslinked complexes recovered by an anti-Delta antibody included Notch, but complexes recovered by an anti-Fasciclin III antibody did not (Powell, 2001).

To investigate binding in vivo, the fact that diffusible ligands can be stabilized specifically in cells expressing their natural receptor was used. In the midpupal eye, Notch is expressed exclusively in a set of interommatidial cells, the secondary and tertiary pigment cells. Scabrous is not expressed in the retina at this developmental stage. Transient heat-shock induction of hs-scabrous pupae results in a ubiquitous pulse of Scabrous protein that is processed, secreted and rapidly lost from most cells in as little as 5 min after the end of heat shock. However, Scabrous is retained in secondary/tertiary cells, the same cells that express the Notch receptor. Retention of Scabrous occurs in apically localized vesicles commonly observed with secreted ligands and their receptors, including Notch. This cell-type-specific retention suggests that a receptor for Scabrous is found specifically in the secondary/tertiary cells at this stage of development (Powell, 2001).

The expression pattern of Notch was altered through an inducible heat-shock promoter. Co-expression of Scabrous and Notch in all retinal cells results in the stabilization of Scabrous in all cells. Similarly, ubiquitous expression of an inactive form of Notch that lacks its intracellular domain (N^deltaICN) is also sufficient to stabilize exogenous Scabrous in all cells, indicating that Scabrous may associate with the extracellular domain of Notch. This also indicates that stabilization of Scabrous does not require the signaling activity of Notch. As predicted, a membrane-tethered form of the intracellular domain (N^deltaECN) does not retain Scabrous (Powell, 2001).

The extracellular domain of Notch includes 36 epidermal growth factor (EGF)-like repeats. Two demonstrated ligands of Notch, Delta and Serrate, require EGF-like repeats 11 and 12 for binding. Ubiquitous expression of forms of Notch that delete repeats 1-18 or 9-17 (N^delta1-18 and N^delta9-17) is still sufficient to stabilize Scabrous protein in all retinal cells; however, N^delta19-36 and N ^delta17-26 fail to retain Scabrous protein. The inability of N^delta19-36 or N^delta17-26 to retain Scabrous suggests that the site of Notch/Scabrous association lies within repeats 19-26. Consistent with this view, Scabrous protein is able to promote stabilization of N^delta1-18 that was expressed on the surface of S2 cells, but failed to stabilize N^delta19-36 (Powell, 2001).

These data support the view that Scabrous and Notch form a receptor-ligand pair; however, the possibility that binding requires other cofactors or adapters cannot be ruled out. These results indicating stabilization suggest that Scabrous may act to scaffold Notch protein to the extracellular matrix and downregulate Notch activity in at least some experimental situations in order to sharpen local boundaries. Consistent with this view, the sharp boundaries between proneural clusters are lost in sca^BP2 clonal patches, indicating that Scabrous may act over a short distance. Notably, when a sca^BP2 clonal patch crosses the morphogenetic furrow, high E(spl) expression typically coincides with the boundary of the clone (Powell, 2001).

The observed complexity of their in vivo interactions leaves unresolved the precise mechanism by which Scabrous regulates Notch at boundaries. One model suggests that similar to Fringe, Scabrous provides a means by which Notch activity can be altered in the absence of changes in ligand expression, although the tissues and possibly the mechanisms, are different (Powell, 2001).

AP-1 clathrin adaptor and CG8538/Aftiphilin are involved in Notch signaling during eye development in Drosophila melanogaster

Clathrin adaptor protein complex-1 (AP-1) and its accessory proteins play a role in the sorting of integral membrane proteins at the trans-Golgi network and endosomes. Their physiological functions in complex organisms, however, are not fully understood. This study found that CG8538p, an uncharacterized Drosophila protein, shares significant structural and functional characteristics with Aftiphilin, a mammalian AP-1 accessory protein. The Drosophila Aftiphilin was shown to interact directly with the ear domain of γ-adaptin of Drosophila AP-1, but not with the GAE domain of Drosophila GGA. In S2 cells, Drosophila Aftiphilin and AP-1 formed a complex and colocalized at the Golgi compartment. Moreover, tissue-specific depletion of AP-1 or Aftiphilin in the developing eyes resulted in a disordered alignment of photoreceptor neurons in larval stage and roughened eyes with aberrant ommatidia in adult flies. Furthermore, AP-1-depleted photoreceptor neurons showed an intracellular accumulation of a Notch regulator, Scabrous, and downregulation of Notch by promoting its degradation in the lysosomes. These results suggest that AP-1 and Aftiphilin are cooperatively involved in the intracellular trafficking of Notch during eye development in Drosophila (Kametaka, 2012).

AP-1 and GGAs are the major clathrin adaptors that function at the post-Golgi compartments in species ranging from yeast to mammals. After a decade of biochemical and cell biological approaches, however, functional specificity of each adaptor at a molecular level still remains to be solved. The present study showed that Drosophila AP-1 and its novel accessory protein Aftiphilin, but not GGA, are required for eye development, suggesting that the Drosophila AP-1-Aftiphilin protein complex is involved in the intracellular trafficking of specific cargo molecule(s) distinct from those regulated by GGA during eye development. It has previously been reported that the GAE domain of Drosophila GGA lacks major conserved amino acid residues potentially required for interaction with the accessory molecules that possess the tetrapeptide PsiG[PDE][PsiLM] motif. Consistent with this, this study showed that Drosophila GGA failed to interact with Aftiphilin, suggesting that the GAE domain of GGA is not structurally conserved. This finding might also reflect the physiological functional diversity between Drosophila AP-1 and GGA. However, the interaction between AP-1 and GGA was detected in the coimmunoprecipitation analysis, thus Drosophila AP-1 might also have a certain functional mode to form a complex with GGA, as implicated in mammalian cells (Kametaka, 2012).

It has been suggested that CG8538, an ORF in the Drosophila genome, encodes a protein with a limited homology with human Aftiphilin. This study concluded that Drosophila Aftiphilin/CG8538p is a functional counterpart of mammalian Aftiphilin, because of their common characteristics such as the possession of multiple γ-ear binding motifs, specific interaction with the γ-ear of AP-1, and the colocalization with AP-1 at the trans-Golgi compartments. Interestingly, the molecular basis of the interaction between Aftiphilin and the γ-adaptin of the AP-1 complex was also well conserved over species, because ectopically expressed Drosophila Aftiphilin in HeLa cells was also colocalized with γ1-adaptin of AP-1. Thus, the results indicate that Drosophila could serve a good model system to dissect the molecular mechanisms of AP-1 and Aftiphilin functions (Kametaka, 2012).

In the deduced amino acid sequence of Drosophila Aftiphilin/CG8538p, two WxxF-type binding motifs for the α-subunit of AP-2 complex were found. In mammals, Aftiphilin was shown to interact with AP-1 and AP-2, and was also proposed to function with AP-2 at the endocytic pathway in neuronal cells. In S2 cells, Drosophila Aftiphilin is predominantly associated with AP-1-positive Golgi compartments and forms a stable complex with AP-1. Moreover, the molecular interaction between Drosophila Aftiphilin and AP-2 was detected. Although the interaction seems to be minor compared with the interaction with AP-1, it is likely that Aftiphilin has other functions that are not related to AP-1, because the Aftiphilin-depleted fly occasionally showed much smaller eyes with decreased number of ommatidia in addition to the roughened eye phenotype. Precise analysis of the physiological functions of Drosophila Aftiphilin is ongoing (Kametaka, 2012).

Eye-specific depletion of Drosophila Aftiphilin or of any of the sigma1- or mu1-subunits of AP-1 caused misalignment of the photoreceptor neurons due to generation of extra R8 neurons during eye development. A genetic screening for Notch modifier genes suggested that AP47, which encodes the mu1 subunit of Drosophila AP-1, is involved in Notch signaling. Another genome-wide RNAi screening showed that the subunits of Drosophila AP-1 and Aftiphilin/CG8538 are involved in Notch signaling. Recently, it has also been reported that Drosophila AP-1 depletion led to mislocalization of Notch and its regulator Sanpodo (Spdo) to the apical plasma membrane and the adherens junction in the sensory organ precursor (SOP) daughter cells in developing nota in the fly. It was suggested that the altered trafficking of Notch is primarily due to increased recycling of the Notch regulator Spdo from the recycling endosomes to the plasma membrane, and that the mislocalization of Notch to the cell surface caused the gain-of-function phenotype in the AP-1 mutants. By contrast, in the current study a clear loss-of-function phenotype of Notch was observed by depletion of AP-1 or Aftiphilin in the developing eyes (Kametaka, 2012).

This discrepancy is probably due to the different mechanisms by which intracellular trafficking of Notch is regulated in different tissues. This study focused on Scabrous as a candidate for a Notch regulator that is affected by AP-1 or Aftiphilin depletion. Scabrous is a glycosylated secretory protein expressed in the R8 neurons, and sca mutation as well as AP-1-depletion causes duplication of R8 and other photoreceptor neurons. In addition, Scabrous was also shown to bind to the extracellular domain of Notch and to stabilize Notch at the cell surface. Drosophila AP-1 has been shown to function together with clathrin in the biogenesis of mucin-containing secretory granules in the salivary gland (Burgess, 2011). Because Scabrous was shown to accumulate in the intracellular compartments in the AP-1-deficient eye discs, the observations in the current study suggest that a defect in the secretion of Scabrous and/or other regulatory proteins causes the instability of Notch at the cell surface, which leads to degradation of Notch in the endosomal and lysosomal compartments. The decrease in the amount of Notch on the cell surface then causes defects in the lateral inhibition mechanism required for the photoreceptor cell specification during eye development (Kametaka, 2012).

In addition to the tissue-specific regulation of Notch trafficking, Notch signaling could also be regulated in several ways in the intracellular trafficking pathways. In the AP-1-depleted eye antennal discs, Notch was accumulated at the late endosomal-lysosomal compartment upon treatment with the lysosomal inhibitor chloroquine, suggesting that Notch is missorted for its lysosomal degradation. It has recently been showm that defects in endocytic trafficking caused by mutations of vps25, a component of the ESCRT-II complex, caused endosomal accumulation of Notch and enhanced Notch signaling. This suggests that the cellular output of Notch signal could be affected drastically in several ways through alterations in the intracellular transport machineries for Notch protein. Finally, the possibility cannot be excluded that Notch is a cargo molecule for Drosophila AP-1, although no direct interaction between AP-1 and the cytoplasmic tail of Notch has been observed so far (Kametaka, 2012).

In conclusion, Drosophila AP-1 plays a crucial role in Notch stability in vivo. It is inferred that Drosophila AP-1 is involved in the intracellular trafficking of tissue-specific regulators of Notch at the TGN or endosomal compartments, as proposed by Benhra (2011). Notch trafficking can be regulated by several mechanisms, and a particular regulatory mode would predominate according to the context of the development. Further analysis on the precise molecular mechanisms by which Drosophila AP-1 and Aftiphilin are involved in the sorting of these signaling molecules will uncover the physiological functions of these adaptor proteins in vivo (Kametaka, 2012).

DEVELOPMENTAL BIOLOGY

Embryonic

scabrous expression is first detected in the late blastoderm and becomes more prominent during gastrulation [Image] in cells of the neuro-ectoderm. Expression is higher in neuroblasts than in epidermal cells. There is expression in the PNS as well (Mlodzik, 1990).

Larval

During imaginal development, scabrous is expressed in the R8 photoreceptor precursor cells in the eye imaginal disc, and later in R2 and R5 cells. In the wing disc, scabrous expression is coextensive with the anlagen for anterior wing margin bristles and machrochaetae, and is controlled by genes of the achaete-scute complex (Mlodzik, 1990).

Establishment of planar polarity in the Drosophila compound eye requires precise 90° rotation of the ommatidial clusters during development. The morphogenetic furrow controls the stop of ommatidial rotation at 90° by emitting signals to posterior ommatidial clusters. One such signal, Scabrous, is synthesized in the furrow cells and transported in vesicles to ommatidial rows 6-8. Scabrous vesicles are transported through actin-based cellular extensions but not transcytosis. Scabrous functions nonautonomously to control the stop of ommatidial rotation by suppressing Nemo activity in the second 45^o rotation. It is proposed that the morphogenetic furrow regulates precise ommatidial rotation by transporting Scabrous and perhaps other factors through actin-based cellular extensions (Chou, 2002).

In the eye discs of furrow-stop mutants hh¹ and ro^DOM, in which the MF is disrupted, ommatidia overrotate during the second 45° of their rotation. Cell fate specification in hh¹ is not affected by the disrupted MF, since the expressions are normal for markers of all types of photoreceptors and cone cells. Adult eye sections also reveal the full complement and the regular trapezoidal arrangement of photoreceptor clusters. The overrotation phenotype is not simply contributed by a faster speed of second 45° rotation, since the degree of rotation is much more than 90°, and the overrotation phenotype persists to the adult stage, suggesting a defect in the stop of rotation. Overrotation defect is also evident in ommatidial clusters posterior to the smo³ tkv^strII double mutant clones that disrupt MF progression, leading to the hypothesis that the MF controls the stop of ommatidial rotation by emitting signals to its posterior direction (Chou, 2002).

The following evidence suggests that the MF secretes Sca to stop ommatidial rotation: (1) a significant overrotation defect is detected in sca hypomorphic mutants and is strongly enhanced by reducing hh activity; (2) ommatidial clusters posterior to sca^BP2 mutant clones display overrotation defects; (3) expression of sca in the posterior differentiating cells of hh¹ eye discs effectively rescues the overrotation defect. This rescuing effect also suggests that sca exerts the rotation-stop function in the posterior region. In genetic analysis with nmo, it has been demonstrated that sca functions specifically in the second 45° rotation; Sca overexpression suppresses the second but not the first 45° rotation, and in the absence of the second 45° rotation in the nmo mutant, disruption of the MF by hh¹ exhibits a normal first 45° of rotation. Further overexpression analyses suggest that sca likely functions to suppress nmo activity. All these data support the vital role of Sca as a signal for ending the second 45° rotation of ommatidial clusters (Chou, 2002).

These analyses suggest two possible modes for sca to function as a stop signal. One is that the Sca signal instructs the formation of a barrier during the second 45° rotation process. Therefore, in the presence of this barrier, all ommatidial clusters rotate to 90° and then stop. Alternatively, the Sca signal counteracts the forward promoting force of the second rotation. The balance between the promoting force and the Sca 'braking' force regulates the rotation stop at 90°. Results from experiments of Sca overexpression support the latter possibility; Sca overexpression in the posterior region causes underrotation and suppresses the Nmo overexpression phenotype (Chou, 2002).

Sca is transported posteriorly to ommatidial rows 6-8, the region where ommatidial clusters are in the process of the second 45° rotation, which is not complete until much later. Therefore, Sca likely transduces a pathway controlling the eventual rotational stop at a later stage. One possibility is that Sca transduces a signaling pathway in all or subsets of the rotating cells in which the rotation-stop mechanism is subsequently activated. Although Sca is suggested not to be a high-affinity ligand of N, it has been demonstrated that Sca is able to antagonize N signaling in the Drosophila wing. On the cell surface, Sca associates with and stabilizes N. No overrotation phenotype, however, has been observed in N^ts1 mutants at restricted temperature, suggesting that other factors are likely functioning as receptors for Sca in this process. In sca loss-of-function and sca misexpression mutants, it has been shown that the epithelium is disorganized and the distribution of adhesion or junction molecules such as DE-cadherin is disturbed. Abnormal aggregation of F-actin has been observed in phalloidin-stained eye discs with sca or nmo misexpression in the ommatidial clusters. Therefore, it is possible that Sca acts on ommatidial clusters in ommatidial rows 6-8 to modify the extracellular environment or intracellular cytoskeleton organization to facilitate the establishment of a rotation-stop mechanism (Chou, 2002).

The staining pattern of Sca reveals that Sca vesicles are transported from the MF to ommatidial row 6-8. Two possible mechanisms are surmised for the transportation of Sca vesicles: transportation through exocytosis and endocytosis, such as argosomes or transportation by extended cellular processes, such as cytonemes in the Drosophila wing disc or microtubule-based peripodial extensions. The latter possibility is preferred due to the following data: first, argosomes are freely diffusible without direction. Second, Sca vesicles are still aligned in linear pattern in shi^ts1 clones, which behave semidominantly in blocking endocytosis. Furthermore, immunostaining for endogenous Sca or Sca-GFP reveals extended cellular processes that likely function in transporting vesicles from the MF to posterior cells (Chou, 2002).

The cellular extension for Sca transportation is sensitive to treatment with cytochalasin D. Also, the cellular extension is labeled with Actin-GFP when expressed in the MF, suggesting that this structure is likely F-actin based. Polymerization and depolymerization of F-actin is very dynamic and is often utilized in rapidly changing structures, like filopodia in the growth cones of axons that constantly probe the environment and change their structure accordingly. This dynamic property of the extension allows it to cope with the rapid advance of the MF during development. Also, the character of this F-actin-based structure makes it unlike the peripodial extension in the Drosophila eye disc, which is microtubule based. In contrast to the cytoneme that extends from the signal-receiving cells, the MF cells extend cellular process to deliver signals to posterior cells (Chou, 2002).

Interestingly, the findings that Sca can be transported into posterior sca^BP2 or ato¹ mutant clones suggest that this Sca pattern is truly emitted from the MF. Moreover, even lacking receiving cells in the ato¹ mutant clone, in which no cell differentiates, the Sca pattern can still be detected in the mutant cells, suggesting that the receiving cells do not induce the cellular extensions from the MF cells. The protrusion into the clones, however, is not linear; they were found wriggling around within the mutant clones, suggesting that this structure is affected by the correct differentiation of posterior clusters. One possibility is that this linear structure is guided or maintained by posterior differentiating cells, such as photoreceptors. Sca may be released to the extracellular space and captured by receptors on the receiving cells or into the intracellular cytosol of the receiving cells. At the end of the string-like Sca pattern, infusions of Sca are often observed into the cytosol of particular cells that may be the receiving cells. Understanding which cells within the ommatidial cluster receive the Sca signal and how they respond will be important to further dissect the mechanism in controlling ommatidial rotation (Chou, 2002).

Effects of Mutation or Deletion

scabrous mutants have a visible phenotype in adults consisting of duplicate bristles and a rough eye due to irregular spacing of too many ommatidial preclusters formed in the morphogenetic furrow (Baker, 1990 and Mlodzik, 1990).

Interactions are described between the Notch locus of Drosophila, and two other loci, scabrous and vestigial, which respectively affect the eyes and wings, were examined. The Notch locus is responsible for mediating decisions of cell fate throughout development in many different tissues. Mutations and duplications of vestigial and scabrous alter the severity of phenotypes associated with Notch mutations and duplications in a manner that is essentially tissue- and allele-specific. These interactions indicate that the products of vestigial and scabrous act in conjunction with Notch to stimulate the differentiation of specific cell types (Rabinow, 1990).

The magnitude of segregating variation for bristle number in Drosophila exceeds that predicted from models of mutation-selection balance. To evaluate the hypothesis that genotype-environment interaction (GEI) maintains variation for bristle number in nature, the extent of GEI was quantitated for abdominal and sternopleural bristles among 98 recombinant inbred lines, derived from two homozygous laboratory strains, in three temperature environments. There is considerable GEI for both bristle traits. A genome-wide screen was conducted for quantitative trait loci (QTLs) affecting bristle number in each sex and temperature environment, using a dense (3.2-cM) marker map of polymorphic insertion sites of roo transposable elements. Nine sternopleural and 11 abdominal bristle number QTLs were detected. Significant GEI is exhibited by 14 QTLs, but there was heterogeneity among QTLs in their sensitivity to thermal and sexual environments. To further evaluate the hypothesis that GEI maintains variation for bristle number, estimates of allelic effects across environments at genetic loci affecting the traits are required. This level of resolution may be achievable for Drosophila bristle number because candidate loci affecting bristle development often map to the same location as bristle number QTL, including achaete-scute, scabrous, hairy, and Delta (Gurganus, 1998).

The scabrous (sca) gene encodes a secreted dimeric glycoprotein with putative coiled-coil domains N-terminally and a C-terminal region related to the blood clot protein fibrinogen. SCA belongs to an expanding class of secreted proteins related in part to the blood clot protein fibrin. These proteins share a C-terminal fibrinogen-related domain (FReD), and they differ in their N-terminal regions, which often contain coiled coils. In Sca, N-terminal apparent coiled coils are required for dimerization, and they are linked to the C-terminal FReD by a short proline-rich region. Other class members include fibrinogen itself, the extracellular matrix proteins of the tenascin group (all of which are hexameric) and the newly described angiopoietins, which are ligands interacting with tyrosine kinase receptors and are of unknown multimerism. An evolutionary tree constructed for FReD sequences implies a common ancestor for the various FReDs. In none of these proteins is the specific function of the FReD known, despite its conservation and the medical importance of fibrin as the protein directly responsible for myocardial infarction and stroke. Homozygous sca mutants have extra bristle organs and rough eyes. A GAL4-based expression system is described for testing rescue of the sca mutant phenotype by altered Sca proteins and for misexpression. Deletion of the FReD domain greatly decreases Sca function, confirming the importance of this conserved region. Sca function could not be restored by FReDs from human fibrinogen chain genes. However, proteins lacking any FReD still show some function in both rescue and misexpression experiments, suggesting that putative effector-binding regions lie outside this domain. Consistent with this, proteins expressing only the FReD have no rescuing activity but are recessive negative; i.e., they enhance the phenotype of sca mutations but have no phenotype in the presence of a wild-type sca allele. This suggests that the FReD contributes to Sca function by binding to other components of the bristle determination pathway, increasing the activity of the linked N-terminal region (Lee, 1998).

Characterization of mutant alleles of sca has suggested that each of its constituent domains might interact with other proteins. The FReD has been found to be essential for proper Sca function because its mutation or deletion lead to loss of function. Three possible roles for the FReD in Sca function have been considered: (1) both the FReD and N-terminal region might contact other proteins to effect Sca function; (2) the FReD might be the major functional domain that is attached to the N domain for structural reasons, and (3) the N domain might be the effector domain, with the FReD playing a supporting or regulatory role. The present data favor the third model, in which the FReD supports or regulates activities of the N domain. The N-terminal region of Sca, which lacks the FReD completely, can partially rescue the sca mutant phenotype, as measured by thoracic bristle number. This indicates that sequences capable of mediating sca gene functions, including possible interaction with a hypothesized but unidentified receptor, must lie in the N domain. Because every function observed for wild-type Sca constructs is shown by the N domain to some degree, whereas none are mediated by the FReD alone, the FReD most likely serves to enhance, regulate, or localize N-domain function rather than to perform supplementary functions. The FReD must also be covalently linked to the N domain; coexpression of the FReD as a separate peptide does not enhance rescue by the N domain (Lee, 1998).

The FReD is clearly required for full sca activity, and, therefore, it is concluded that the FReD sequence promotes activities that, to a lesser degree, the N domain can perform alone. Deleting the FReD, substituting Asn for Asp654, or replacement with FReDs from human fibrinogen chain genes all yield defective proteins with less than half the wild-type activity, as judged from bristle numbers. The role of the FReD in sca could not be performed by similar sequences from the human fibrinogen chain genes. The FReD from wild-type Sca must be able to interact with some other gene product because expression of the FReD alone enhances the sca mutant phenotype, implying an interaction with other components of the bristle determination pathway. There is no formal evidence to prove that this reflects a normal interaction of the FReD as part of the wild-type protein. However, if absence of the N domain permits neomorphic FReD interactions, these would also be expected to occur in a wild-type Sca background or after ectopic expression of FReD proteins. Instead, no effect is seen from expressing FReDs in other tissues, and bristle number is increased only when in a sca mutant background. These findings suggest that the FReD interacts with other gene products whose functions are more important in a sca mutant background and, therefore, part of the same pathway of bristle patterning. Such interactions may well be the basis of the FReD contribution to wild-type Sca function. Because the FReDs are presumed to be structurally similar, it may now be possible to map the critical regions within them by replacing smaller regions between Sca and fibrinogen chains (Lee, 1998).

Notch is a receptor for signals that inhibit neural precursor specification. Since N and its ligand Delta are expressed homogeneously, other molecules may be differentially expressed or active to permit neural precursor cells to arise intermingled with nonneural cells. During Drosophila wing development, the glycosyltransferase encoded by the gene fringe promotes N signaling in response to DI, but inhibits N signaling in response to Serrate, which encodes a ligand that is structurally similar to DI. Dorsal expression of Fng protein localizes N signaling to the dorsoventral (DV) wing margin. The secreted protein Scabrous is a candidate for modulation of N in neural cells. Mutations at the scabrous locus alter the locations where precursor cells form in the peripheral nervous system. Unlike fringe, sca mutations act cell non-autonomously. Targeted misexpression of Sca during wing development inhibits N signaling, blocking expression of all N target genes. Sca reduces N activation in response to DI more than in response to Ser. Ligand-independent signaling by overexpression of N protein, or by expression of activated truncated N molecules, is not inhibited by Sca. These results indicate that Sca can act on N to reduce its availability for paracrine and autocrine interactions with DI and Ser, and can act as an antagonist of N signaling (Lee, 2000).

The Drosophila rotund (rn) gene is required in the wings, antenna, haltere, proboscis and legs. Previously identified in the rotund region was a member of the Rac family of GTPases, denoted the RacGAP84C or rotund racGAP gene. However, rotund racGAP is not responsible for the rotund phenotypes. The rotund gene has now been isolated. It is a member of the Krüppel family of zinc finger genes. The adjacent roughened eye locus specifically affects the eye and is genetically separable from rotund. However, roughened eye and rotund are tightly linked, and thanks to this connection, the roughened eye transcript was isolated. Intriguingly, roughened eye is part of the rotund gene but is represented by a different transcript. The rotund and roughened eye transcripts result from the utilization of two different promoters that direct expression in non-overlapping domains in the larval imaginal discs (St Pierre, 2002 and references therein).

Little is known about the genetic cascades within which roe and rn are acting. The results from eye-antennal imaginal discs indicate that roe acts at the morphogenetic furrow, as evident both from its expression and from the effects on Delta and Scabrous expression in roe mutants. Both Dl and sca play roles in spacing the array of ommatidial preclusters in the morphogenetic furrow, and it is interesting to note that the expression of roe at the furrow is not evenly distributed and appears stronger in clusters of cells. Genetic screens for modifiers of the N^spl mutation have identified roe as an enhancer, and sca and Dl as suppressors of the N^spl eye phenotype. Given the dynamics of N signaling, these results support models where Roe acts to either positively or negatively regulate Dl and Sca. A genetic interaction screen for enhancers of glass also identified roe, an interesting finding given that ectopic expression of roe using GMR-GAL4 leads to a glass-like phenotype with a loss of bristles and pigment cells (St Pierre, 2002 and references therein).

The role of scabrous in the evenly spaced bristle pattern of Drosophila

The role of scabrous in the evenly spaced bristle pattern of Drosophila has been explored. Loss-of-function of sca results in development of an excess of bristles. Segregation of alternately spaced bristle precursors and epidermal cells from a group of equipotential cells relies on lateral inhibition mediated by Notch and Delta (Dl). In this process, presumptive bristle precursors inhibit the neural fate of neighboring cells, causing them to adopt the epidermal fate. Dl, a membrane-bound ligand for Notch, can inhibit adjacent cells, in direct contact with the precursor, in the absence of Sca. In contrast, inhibition of cells not adjacent to the precursor requires, in addition, Sca, a secreted molecule with a fibrinogen-related domain. Over-expression of Sca in a wild-type background, leads to increased spacing between bristles, suggesting that the range of signaling has been increased. scabrous acts nonautonomously, and evidence is presented that, during bristle precursor segregation, Sca is required to maintain the normal adhesive properties of epithelial cells. The possible effects of such changes on the range of signaling are discussed. It is also shown that the sensory organ precursors extend numerous fine cytoplasmic extensions bearing Dl molecules, and these structures may play an active role during signaling (Renaud, 2001).

sca functions in the inhibition of the neural fate, since in its absence an excess of neural precursors form. The sca mutant phenotype is very similar to that of hypomorphic N and Dl mutants and indeed Notch and Dl are known to be the main components of the signaling pathway regulating the spaced pattern of bristles. Thus, Sca is likely to positively modulate N signaling. During bristle precursor selection, like Dl, sca acts nonautonomously and is not required for reception of the signal that places its activity upstream of N. What are the respective roles of Dl and sca in the segregation of bristle precursors? Both genes are expressed in proneural domains and then at high levels in the bristle precursors, and their products act nonautonomously on neighboring presumptive epidermal cells. Both proteins associate with N, but so far only Dl has been shown to be an activating ligand. Scabrous is a secreted molecule, whereas data accumulated to date suggest that active Dl is membrane-bound. In the complete absence of Dl, all cells adopt the neural fate and thus bristle precursors arise adjacent to one another. In the complete absence of Sca, there is an excess of bristle precursors but they are never adjacent and are always separated by at least one epidermal cell. This indicates that sca is not needed for the bristles to be spaced apart by a short distance, and that Dl, which is expressed in sca mutants, is able to inhibit cells immediately adjacent to the precursors without any help from Sca. In contrast, in the absence of Sca, Dl is unable to inhibit those cells not in direct contact with the precursor. This suggests a role for Sca in the inhibition of cells not adjacent to the precursor. These two, possibly separable events, are referred to as 'short' and 'long' range signaling (Renaud, 2001).

Formally, there are three possible mechanisms for 'long' range signaling. The first would be that sca is part of a relay mechanism whereby cells immediately adjacent to the precursor are inhibited by Dl and then relay the signal farther out by means of Sca. This hypothesis is very unlikely, however, because one would expect sca to be expressed in cells adjacent to the precursor following activation of N by Dl. In fact, sca protein is only detectable in the precursor itself, although sca is earlier expressed at low levels in the proneural domain. Furthermore, sca is expressed in the absence of Dl and therefore does not require a prior signaling event mediated by Dl (Renaud, 2001).

A second possible mechanism is that there may be two independent signals, one acting at a 'short' range (Dl) and another at a 'long' range (Sca). Two observations suggest that this, too, is unlikely: (1) in the absence of Dl, all cells adopt the neural fate and so the process of bristle spacing is completely abolished. scabrous is expressed in cells mutant for Dl so this result indicates that Sca alone is unable to repress the neural fate; (2) the results indicate that sca is not involved in the process of lateral inhibition whereby a pattern of alternating neural and epidermal cells is generated. Along a border between wild-type cells and cells mutant for Dl, the precursors are nearly always selected from the pool of wild-type cells, rather than from the mutant cells. This is thought to be because the mutant cells produce little or no signal and are inhibited by the Dl-producing wild-type cells. Along sca mosaic borders, the mutant cells can adopt the neural fate, indicating that they are not defective in the production of the activating ligand Dl. Furthermore, since wild-type precursors also form along the mosaic border, neural precursors can be chosen from cells of either genotype. Thus, cells choose the neural or epidermal fate regardless of whether or not they express sca (Renaud, 2001).

This would explain the fact that precursors are never adjacent in sca mutants and is not inconsistent with the observed sca phenotype of excess bristles. During segregation of both the normal component of precursors, as well as the additional precursors, adjacent cells have to choose between epidermal and neural fates. Any failure of this process would result in the presence of adjacent bristle precursors, a phenotype characteristic of N and Dl mutants, but not sca mutants. Segregation of single neural precursors surrounded by epidermal neighbors is thus probably mediated by Dl alone, which would explain why bristle spacing is completely abolished in the absence of Dl, even though Sca is present (Renaud, 2001).

The third possibility is that 'long' range signaling requires both Dl and Sca. Under this hypothesis, Dl would be the activating ligand in 'long' range signaling, but would require Sca in order to inhibit cells not directly in contact with the precursor. This could be the case regardless of whether the signal originates exclusively in the precursor or additionally in proneural cells. One observation in favour of this hypothesis is afforded by examination of flies mosaic for Dl. Along the edges of Dl mutant clones, mutant cells are able to differentiate as epidermis under the influence of an inhibitory signal from neighboring wild-type cells. This 'rescue,' due to expression of Dl in the wild-type neighbours, can extend up to four cell diameters. If there is no relay mechanism and no other signal, then Dl must somehow be able to activate N several cell diameters away from the cell in which it is produced. Scabrous is of course present in both the Dl⁺ and the Dl^- cells, but is unable to effect any rescue of cells mutant for Dl that are situated more than about four cells away from the wild-type Dl-expressing cells. Examination of Dl mutant clones in a mutant sca background would indicate the range of Dl signaling in the absence of Sca (Renaud, 2001). Clones doubly mutant for sca and Dl, however, fail to differentiate the cuticular components of the bristles (for all allelic combinations tested) and so are uninformative (Renaud, 2001).

It is suggested that bristle spacing may be the result of two signaling events. The first step of lateral signaling involves Dl alone and allows a group of cells to choose a single neural precursor that will inhibit adjacent cells. This step proceeds normally in the absence of Sca. The second step would act to inhibit cells that are not adjacent to the precursor and would require the activity of both Dl and sca. This step is impaired in both sca and Dl mutants. In Dl mutants, in the absence of the inhibitory signal, lateral inhibition fails, leading to adjacent precursors and a loss of epidermal cells. In sca mutants, an excess of precursors form, but they segregate singly and are spaced by at least one epidermal cell through the activity of Dl. Examination of the nascent precursors with neu-lacZ, fail, to reveal two temporally separate waves of precursor formation in sca mutants. Staining of wild-type pupal nota with DE-cadherin, however, may provide a visual correlate of the two groups of target cells. A rosette-like ring of wedge-shaped cells surrounds the bristle precursor; it is reminiscent of the cell preclusters that precede segregation of the R8 photoreceptor during development of ommatidia. These supra-cellular structures may allow more cells to enter into direct contact with the precursor during the first step of inhibitory signaling (Renaud, 2001).

The results demonstrate a requirement for local discontinuities in the levels of Sca between cells during inhibitory signaling. This was also reported to be the case for eye development. Uniform expression of Sca under experimental conditions is unable to rescue the sca mutant phenotype. So the relative quantitative differences in the level of Sca between neighbouring cells is an essential feature of the spacing mechanism. Over-expression of Sca in a wild-type fly causes a phenotype opposite that of the loss-of-function phenotype. The distance between bristles actually increases and there are correspondingly fewer bristles in the domain of over-expression. In these flies, although exogenous Sca is uniformly distributed, regulation of the endogenous gene will provide local differences in levels of the protein. The greater distance between bristles suggests an extension in the range of the inhibitory signal. Scabrous is likely to be present in a gradient of decreasing concentration around each precursor. The molecule has been shown to have a very short half-life. Thus, it would decay rapidly after secretion and this would help maintain a graded distribution from the source. It is postulated that such a gradient could define the range of the inhibitory signal (Renaud, 2001).

A rescue of up to four cell diameters has been observed at the edges of clones of cells mutant for Dl. This suggests a signaling range that exceeds that required in wild-type flies, where bristles are usually only four to five cells apart. It is possible, however, that the signal range is longer than usual in this experimental situation because very large amounts of Sca are likely to be present in these clones due to the hyperplasia of precursors. It is, however, noteworthy that in other drosophilid species, such as D. ararama, the bristles are spaced by eight or nine cells, suggesting a possibly greater signaling range (Renaud, 2001).

One property of Sca is to modulate adhesive parameters of the epithelial cells surrounding the precursor. Discrete changes in the distribution of DE-cadherin, Discs large, and other junctional proteins are seen in the epidermis of sca mutants, that are associated with mislocalization of the adherens and septate junctions and some disruption of epithelial organization. The monolayered nature of the epithelium is retained, consistent with the fact that the morphogenetic changes of metamorphosis are not impaired in sca mutants. This phenotype is only observed in late third instar discs and early pupae, when bristle precursors are forming. If ac-sc activity is removed, the late third instar disc epithelium is wild type. In ac-sc mutants, the absence of Ac and Sc entails a loss of Sca, whose expression is dependent on Ac-Sc (Renaud, 2001).

The epithelial defects resulting from a lack of Sca protein thus coincide with ac-sc expression and the process of lateral inhibition. Scabrous may therefore be required to maintain normal epithelial integrity by counteracting the effects of a protein(s) activated during precursor segregation. It is therefore likely to act through association with a protein(s) whose expression is regulated by Ac and Sc. Notch is ubiquitously expressed in the epithelium, but is activated and probably up-regulated in future epidermal cells surrounding the precursors. Powell (2001) demonstrated that Sca binds N and as a result N is stabilized at the cell surface in S2 cells. Interestingly, in N^ts1 mutants at 29°C, where the activity of N is strongly reduced, the distribution of DE-cadherin in the disc epithelium is also discretely altered and the epithelium appears similar, but not identical to that of sca. This suggests that the epithelial defects seen in sca mutants may be the result of a failure to stabilize N protein in epithelial cells. These results are consistent with the idea that Sca acts through N, and with earlier observations linking N to epithelial cell adhesion (Renaud, 2001).

scabrous is not required for inhibition of cells that are adjacent to the precursor. Furthermore, the requirement for sca in the fly is quite restricted: it is not expressed in many other tissues where Notch signaling takes place in its absence. While it is expressed in neural precursors in the embryo, loss of the protein there seems to be without consequence, perhaps because the interactions involve adjacent cells. This leads to the hypothesisis that the effects of Sca on cell adhesion and the stabilization of N may be specifically required when the levels of Dl are limiting (Renaud, 2001).

Delta is expressed in proneural domains and can still be detected in presumptive epidermal cells after precursor segregation. Nevertheless the amount of Dl remaining in presumptive epidermal cells appears to be insufficient, by itself, to repress the neural fate. In the absence of Sca, or in flies carrying hypomorphic alleles of Dl, the space between bristles is decreased. Furthermore, in epidermal cells, the transcription of Dl progressively declines due to repression of its regulators ac and sc following N activation, whereas in the bristle precursors the levels of Dl increase as the levels of Ac and Sc rise. This leaves two possibilities. One is that all of the signal originates in the precursor cell, in which case Dl must be transported, by some as yet unknown means, from the precursor to cells not in direct contact with the latter. The other, is that the concentration of Dl molecules remaining on the presumptive epidermal cells is insufficient to inhibit by itself, but can do so if helped by Sca. Dl from both groups of cells may participate in the wild type. In either case, stabilization of N may therefore be a means to increase the chances of receptor activation in the presence of limiting amounts of Dl (Renaud, 2001).

Changes in cell adhesion could also be the cause of the abnormal bristle organs seen in sca mutants, where two or more cells of the bristle organ lineage adopt the same fate at the expense of the others. In the wild type, spatial arrangements of the cells of the bristle organs are stereotyped as a result of the nonrandom orientation of the mitotic spindles at each division. In sca mutants, the cells are often randomly arranged and in some cases appear to drift apart from one another. This would be likely to prevent the precise cell-cell interactions, mediated by the N signaling pathway, necessary for the assignment of the correct cell fates (Renaud, 2001).

The precursor cells have a quite distinctive shape, reminiscent of neurons with a number of filopodial-like extensions that fan out in a planar orientation. It is not known whether, during bristle precursor segregation, the epidermal cells of the notum extend similar filopodia. Oriented epidermal outgrowths have been described in the epidermis of other insects and also in the wing pouch and peripodial membrane of Drosophila imaginal discs, where it has been suggested that they may function during signaling. Some of these structures project basally and others extend long, straight and polarized structures. Their morphology differs from that of the extensions observed (Renaud, 2001).

It is not known whether Dl from the precursor cell is able to reach cells that are not adjacent to the Dl-expressing precursor, but one means by which this could occur is via cytoplasmic extensions. This has been suggested for Lag-2 signaling in the nematode germ line. Indeed, the presence of Dl molecules can be detected on the filopodia. Although formation of filopodia will depend on properties of the neural precursor itself, subtle changes in junctional contacts and adhesion between surrounding epithelial cells may help to orient or stabilize these structures. The changes in the bristle density of both wild-type and sca flies that are seen after expression of a dominant negative form of DE-cadherin, suggest a role for adhesion molecules in bristle spacing. Preliminary observations indicate that Sca is not required for the extension of filopods. This is consistent with the nonautonomy of sca mutant cells. If filopodia are the means whereby nonadjacent cells are inhibited, and if Sca were to be required for extension of filopodia from the Sca-producing bristle precursors, then sca would be expected to behave autonomously (Renaud, 2001).

Further studies are necessary to determine the molecular basis of Sca function, but one possibility is that binding of Sca to N leads to discrete modifications in epithelial structure that allow Dl molecules on the cytoplasmic extensions to form stable ligand-receptor complexes. The colocalisation of Sca and Dl in cytoplasmic vesicles may indicate cellular trafficking of protein complexes that include Dl, N, and Sca (Renaud, 2001).

In Drosophila, microchaetes (small bristles) are regularly spaced and form five straight rows in the acrostichal region of the adult notum. Microchaetes develop from sensory organ precursors that arise as single, evenly-spaced cells during pupal development. This article addresses the question of how the precursor cells remain aligned throughout pupal development, in spite of continued division of the intervening epidermal cells. Using in vivo imaging it has been shown that bristle precursors move about continuously throughout development, covering distances of up to one or two cell diameters. During this process, they remain aligned in wild-type flies, suggesting that the movement may be regulated. Flies mutant for scabrous (sca) have a disorganized pattern of bristles with little or no alignment. In vivo observations of sca mutants indicate that the precursor cells move around more than in the wild type, but that, in spite of this the precursor cells and resulting bristles never become well aligned. They appear to follow a more complex path, suggesting that the movement is not co-ordinated. Moreover, analysis of the alignment of precursor cells in vivo in wild-type and sca mutant flies indicate that mutant animals are not able to maintain the pattern of precursor cells during development. Analysis of mosaic flies confirmed the time-lapse observations and showed furthermore that bristles preferentially move towards high levels of Scabrous. It is suggested that, by altering the properties of epithelial cells in a graded fashion, Scabrous may provide cues that allow the precursors to remain evenly spaced after they have segregated (Renaud, 2002).

Sensory organ precursors (SOP) were monitored in vivo during pupal development in a neuralised-GAL4, UAS-GFP strain. The neuralised-GAL4 driver was used as a marker for the precursors: neuralised is expressed very early, exclusively in the SOP and not in epidermal cells. SOPs emerge sequentially in rows. Rows 1 and 5 emerge first, then row 3 appears in between, and finally rows 2 and 4 are intercalated between rows 1/3 and 3/5. Within any particular row, the time of formation of individual SOPs was found to be quite variable; they do not appear to arise in any particular order. Furthermore, some SOPs arise quite late, particularly in rows 2 and 4. The SOPs of each row are aligned from the beginning and they remain aligned throughout pupal development during which time the epidermal cells undergo further division [from 14 to 18 h after puparium formation (APF)]. Pupae used for the time-lapse experiment develop normally and so the resulting imagos were collected and the nota mounted. It was thus not possible to determine the final position of each bristle that had been monitored during development. The spacing between microchaetes is maintained both within and between the rows, indicating that the bristles have remained within their respective rows. Thus the pattern is maintained throughout development until the adult stage, suggesting the existence of a mechanism to maintain the alignment (Renaud, 2002).

GFP expression and SOP movement were followed from 13 to 24 h APF. Movement probably continues after this point, up to about 36 h APF, when the adult cuticle starts to be secreted; this probably impedes further movement. The position of each precursor was plotted onto the same drawing every 30 min using different color combinations. To quantify the distance of displacement of a SOP between 13 h and before its first division, the shortest distance between the position of a SOP at the beginning and the end of the experiment was measured. The average distance observed was 5.15±0.27 m m, and the maximum distance was 19.8 mm. The average diameter of an epidermal cell is 11.71±0.36 mm. At this stage, SOPs are separated by about two epidermal cells (this increases to between 4 and 5 by the end of the development), so a displacement of the SOP of one cell diameter or more is more than sufficient to modify the alignment of the final pattern (Renaud, 2002).

No relationship was observed between SOP movement and displacement of the neighboring epidermal cells, suggesting that SOPs are not simply pushed away by epidermal cell displacement. In addition, epidermal cell displacement is very limited and could not explain the distances measured for SOP movement (Renaud, 2002).

The bristles are very disorganized on the notum of scabrous mutants. Homozygous mutant flies display an excess of microchaetes that are poorly, if at all, aligned into rows. It has been shown that the five rows of precursors arise from five stripes of achaete-scute expression. scabrous mutant flies were examined for the presence of these stripes, by staining with an antibody against Achaete. No difference in Achaete expression was seen between wild type and scabrous mutant pupa just prior to the emergence of bristle precursors. This result indicates that the disorganized pattern observed in adult mutant flies is not due to a mis-expression of achaete-scute when SOPs arise and is consistent with the known regulation of scabrous by Achaete and Scute (Renaud, 2002).

The density of microchaetes is higher in scabrous mutant flies and perhaps this feature alone could account for the disruption of the bristle rows. Therefore, flies mutant for hairy (h), extramacrochate (emc) and Hairy-wing (Hw), that also had a higher density of bristles on the notum, were examined. In these cases bristles could still be seen to be roughly aligned in spite of the increased density. Thus rows appear to be specifically disrupted in scabrous mutant flies. This result suggests that the function of scabrous is required for spacing and alignment (Renaud, 2002).

A number of differences were noted between scabrous and wild-type flies. From early time points on, the SOPs are poorly aligned. They emerge in greater numbers and earlier than in the wild type. This has also been observed for the larval sense organs in Notch mutants. The order of appearance of the rows is less strict than in the wild type, although rows 1 and 5 do seem to appear first. The only row that displays, more or less, a correct initial alignment is row 1, which is closest to the midline. The disorganized nature of the SOP pattern is never resolved during pupal development and becomes progressively more chaotic (Renaud, 2002).

SOPs are perfectly able to move in a scabrous null mutant. As in the wild type, the precursors move around constantly. The average distance covered, when the initial and final points are compared, was found to be 3.22±0.20 mm. The greatest distance measured was 12.7 mm. These distances are smaller than those observed in wild-type animals, but, as before, they do not reflect the total distance covered since the SOPs follow a roundabout path. The total distance travelled by a SOP (obtained by adding together all distances measured between each pair of data points) between 14 and 21.5 h APF was found to be higher than in the wild type (24.09±1.25 mm). This result demonstrates that in scabrous mutants the bristle precursors move around even more (Renaud, 2002).

Thus, although the SOPs move around more in scabrous mutants than in wild-type flies, the actual distance between the beginning and end positions of a SOP is considerably smaller. This suggests that they may follow a more complex, roundabout path. In order to verify this, drawings were made of the path of each SOP by joining the points occupied at each recording made between 14 and 21.5 h APF. The resulting drawings were then divided into two categories, those in which the path travelled crossed back over itself at least once, and those in which it did not. Nearly half of the scabrous mutant SOPs, but only a quarter of wild-type SOPs fell into the former category. Thus SOPs appear to follow a more complex path in the mutant (Renaud, 2002).

Therefore, while SOP movement may be oriented in the wild type (the small movements may be in response to the proximity of neighboring precursors, a feature that will vary with time and place), movement appears to be random in scabrous mutants. Indeed, in spite of their constant movement, the bristles in scabrous mutants do not end up in a regularly spaced pattern. The example of row 1 is striking. At 14 h APF the SOPs in row 1 are almost aligned, but by 21.5 h APF they are much more disorganized. Row 3, which forms by intercalation between rows 1 and 5, appears to be wider than normal and very disorganized in scabrous mutant flies. Thus from the beginning of precursor segregation in this row, the alignment of bristles is impaired. It is concluded that SOPs form in a disorganized manner in scabrous mutants and that this disorganization is amplified in the course of notum development (Renaud, 2002).

Thus, during the development of wild-type flies, bristle precursors and organs move around continuously whereas epidermal cells show almost no movement. SOPs can move up to several cell diameters but do not display any apparent preferred direction. SOPs arise in a fairly well spaced pattern and this is maintained throughout development, in spite of all the movement. It is suggested that the constant movement allows SOPs to continuously readjust their positions relative to one another, and this enables them to remain equidistant. The direction of movement would then be determined by the relative distance to each of the surrounding SOPs. Bristle precursors do not arise simultaneously, so earlier-born ones may have to readjust their position as new ones are added. Although the distances covered are small, it should be noted that bristles, both within and between rows, are normally separated by only four to five cell diameters. Thus displacement of one cell diameter is sufficient to disrupt the alignment (Renaud, 2002).

In scabrous mutants, the precursors also move around continuously, even more so than in the wild type. However, in this case, in spite of normal expression of ac-sc in stripes and some initial alignment of the precursors (for example row 1 is well aligned with respect to the midline), spacing becomes progressively more chaotic and by the time of differentiation the five rows are not distinguishable. So if bristle movement is indeed a means of maintaining regular spacing, it seems that in the absence of Scabrous, SOPs are unable to 'sense' the position of surrounding SOPs and adjust their own position accordingly. They are never found to be adjacent, however, suggesting that factors other than Scabrous, such as Delta, prevent them from moving too close together (Renaud, 2002).

Scabrous has been shown to act non-autonomously. It is expressed in the SOP, but is a secreted protein and functions to alter the adhesive properties of surrounding epidermal cells. In mosaics, bristles mutant for scabrous preferentially move out of mutant territories into surrounding wild-type tissue. This suggests that the mutant bristles move in response to the presence of Sca in the surrounding tissue. If so, Sca could provide directional cues and the SOPs may move to the area of highest concentration. In the wild type, Sca is secreted by the bristle precursor itself and levels would be expected to decrease with distance from this source. The very short half life of Sca would help to maintain these concentration differences. Thus the highest levels of Sca will be in the immediate vicinity of each SOP where and when it forms. Since precursors arise in an evenly spaced pattern, they would never need to move much and would remain near their site of origin within a pool of Sca. This would explain why, in the wild type, precursors generally do not move more than about one cell diameter. Thus, in the wild type, high levels of Sca at the site of origin, would prevent SOPs from moving away from this site and help them to remain in the spaced pattern in which they arise. In contrast, in the mutants, bristles would tend to stray from their site of origin and move around indiscriminately (Renaud, 2002).

One possible mechanism for movement would be that the SOPs are able to detect and measure the levels of Scabrous that they themselves have produced, and then move so as to remain close to the highest levels of this protein. This would be consistent with the fact that Scabrous itself is not required for movement per se, and that mutant SOPs devoid of Scabrous tend to move into wild-type Scabrous-producing tissue in mosaics. However, the alternative hypothesis that bristles are passively displaced due to specific adhesive properties of epidermal cells is favored, since these are the cells that are modified by Sca. It has been shown that Scabrous alters the adhesive nature of epithelial cells. Transmitted electron microscopy of the epithelium of the scabrous pupal notum shows that the organization of the epithelium is disrupted and that the distribution of DE-cadherin, Armadillo and Disc large is altered. If cells adhere more tightly at higher levels of Sca this might prevent bristle displacement. This would account for the fact that the bristles move less in wild-type scabrous-expressing tissue, than in mutant tissue. Over-expression of a dominant negative form of DE-cadherin in the acrostichal domain of the notum also leads to a disruption of the microchaete spacing and alignment, not unlike that seen in scabrous mutants. Alignment is not affected when expression is driven exclusively in the precursor cells. This provides strong evidence that levels of adhesion between epithelial cells is a requirement for the regular spacing of bristles. Notch and Dl also affect cell adhesion: cells in culture expressing N adhere to cells expressing Dl. The fact that Sca associates with, and stabilizes Notch, may be the mechanism by which Sca could regulate the adhesive properties of the surrounding cells (Renaud, 2002).

These observations uncover a hitherto unrecognized role of SOP movement in the final pattern of evenly distributed bristles. Scabrous may be the crucial factor in this process by regulating indirectly the adhesive properties of the epidermal cells surrounding the SOP. Further details on the epithelial modifications brought about by Sca may clarify these processes (Renaud, 2002).

Interaction of Scabrous and Gp150

Notch and Delta are required for lateral inhibition during eye development. They prevent a tenfold excess in R8 photoreceptor cell specification. Mutations in two other genes, Scabrous and Gp150, result in more modestly increased R8 specification. Their roles in Notch signaling have been unclear. Both sca and gp150 are required for ectopic Notch activity that occurs in the split mutant. Similar phenotypes show that sca and gp150 genes act in a common pathway. Gp150 was required for all activities of Sca, including inhibition of Notch activity and association with Notch-expressing cells that occur when Sca is ectopically expressed. Mosaic analysis found that the gp150 and sca genes were required in different cells. Gp150 concentrates Sca protein in late endosomes. A model is proposed in which endosomal Sca and Gp150 promote Notch activation in response to Delta, by regulating acquisition of insensitivity to Delta in a subset of cells (Li, 2003).

A Notch mutation, split, specifically elevates Notch activity in the neural cells. The split mutation alters glycosylation of the N extracellular domain and leads to inappropriate N activity within R8 precursor cells in the developing eye. Specifically, a Ile578-->Thr578 substitution responsible for the split mutation introduces a new site for O-fucosylation on EGF repeat 14 of the Notch extracellular domain. This suggests functional differences between Notch in neural and non-neural cells. Thus factors specifically regulate the inactivity of N in neural cells and contribute to the spatial pattern of neurogenesis (Li, 2003).

Genetic studies have identified several genes whose mutations interact with the split allele. One gene has been reported where deletion of a single allele is sufficient to suppress the spl phenotype. This gene encodes the secreted protein Scabrous. In addition, in the homozygous absence of sca, the spl mutation has no detectable effect, i.e. spl mutant and wild-type N behave indistinguishably. Conversely duplications of sca enhance the spl phenotype. These results indicate that activity of N in neural cells depends critically on sca. By contrast, none of the well-known components of N signaling behave as such dose-sensitive genetic modifiers of spl. Special alleles of Dl were also recovered as dominant spl suppressors, consistent with the finding that in spl the N activity in neural cells is ligand dependent (Li, 2003).

The molecular role of Scabrous in the Notch pathway is not yet clear. Mutations of sca cause defects in the spacing and number of sensory mother cells in the epidermis and of R8 precursor cells in the retina, two founder cell types for adult peripheral nervous system. The sca mutations act cell nonautonomously. Because N acts cell autonomously in the specification of these same cell types it was suggested that sca encodes an extracellular ligand for the receptor protein N. This hypothesis proved difficult to confirm, however, since sca mutations affect only a subset of Notch functions, have weaker effects than N null mutations, and since no direct interaction between the Sca and N proteins has been demonstrated. Other ideas have been proposed: that Sca acts to scaffold N to the extracellular matrix to downregulate N activity; that it acts to preserve epithelial structure within proneural regions and so enhance function of other N ligands, or acts independently of N to arrest ommatidial rotation (Li, 2003).

Other findings strongly suggest that Sca and N proteins are closely associated in vivo. When Sca is overexpressed in the developing wing, N activity and specification of the wing margin are prevented, even though wing margin specification is independent of Sca function in the wild type. Sca protein appears to prevent Dl from activating of N in this ectopic expression assay. The results strongly suggest that Sca protein targets N signaling, although not defining the exact role of Sca in normal development. When ectopically expressed in pupal retina, Sca protein is preferentially stabilized in cells expressing N and such stability depends on EGF repeats 19-26 of the N extracellular domain. Dl and Ser signal through EGF repeats 10-12. The association with Sca occurs independently of N signaling activity. Chemical crosslinking of Drosophila embryos detects Sca protein in a complex with N, consistent with a close association between the proteins in vivo. Sca protein also appears to stabilize N protein on the surface of tissue culture cells. It remains uncertain, however, whether the interaction is direct or mediated by other proteins, or where in the cell it occurs (Li, 2003 and references therein).

Another gene required for proper eye and bristle patterning has recently been described. Mutations at the Gp150 locus cause defects in ommatidial development and cuticular bristle development that are similar to those seen in sca homozygotes (Fetchko, 2002). Gp150 protein was originally isolated biochemically as a phosphoprotein target of the receptor tyrosine phosphatase DPTP10D. Recent work shows that Gp150 is located in endosomes and interacts with the Notch pathway (Li, 2003).

This study explores the relationship of Sca and Gp150. Gp150 is required for neural Notch activity in the spl mutant, and it is concluded that the Sca and Gp150 proteins must act in a common pathway, with Gp150 acting downstream in cells that respond to secreted Sca protein. Gp150 is required for all Sca activities yet identified, including those of ectopic expression and association with Notch in vivo. Sca is localized to endosomes along with Gp150. It is proposed that an endosomal pathway downregulates N activity in neural cells, and that Sca and Gp150 oppose this pathway to permit N activity in a subset of non-neural cells. Accordingly, Sca and Gp150 activate N indirectly, via effects on N downregulation (Li, 2003).

To explore how gp150 was required for sca function, attempts were made to identify the cells in which gp150 is required using mosaic analysis. Mosaic analysis using sca mutations shows that the likelihood of normal ommatidial assembly is reduced unless the R8 cell is genetically wild type for sca, consistent with a nonautonomous role for sca in lateral inhibition. Previous mosaic analysis with gp150 provided only limited data for R8 cells (Fetchko, 2002). Sections were cut through eyes containing gp150³ homozygous clones, and 90 ommatidia that were phenotypically normal were scored. No specific photoreceptor cell type was found to be important for gp150 function, and ommatidia with R8 cells mutant for gp150 developed normally with the same probability as ommatidia with R8 cells wild type for the gp150 locus. Similar results were obtained from a smaller number of gp150¹ and gp150² mosaics. The mosaic results show that gp150 is not required in the same cells as sca, at least during eye development. They would be consistent with gp150 function in cells that take many fates other than R8, so that no requirement is detected in any specific ommatidial cell. The data rule out the model that Gp150 is required for Sca protein synthesis, but are consistent with Gp150 being required for the localization or reception of Sca by other cells (Li, 2003).

One piece of evidence that Sca interacts with N comes from colocalization studies in the pupal retina. Notch protein distribution is unusually asymmetric in pupal retina, being excluded from the differentiating ommatidia but expressed in the surrounding pigment cell lattice. Sca expressed transiently and uniformly from the hsp70 promoter accumulates in N-expressing cells, implying an interaction of some kind between the proteins (Li, 2003).

To test whether Gp150 was required for Sca to associate with N, Sca was expressed in pupal retinas from gp150 mutants. Prior to heat shock, pupal retinas from HS-sca transgenic flies lack Sca protein until specification of interommatidial bristle precursors begins. Within 20 minutes of mild heat shock newly synthesized Sca protein was cytoplasmic and uniformly distributed. As Sca protein was secreted and decayed, protein transiently accumulates in the Notch-expressing pigment cell lattice, usually between 40-80 minutes after heat shock. Notch protein is still expressed in the pigment cell lattice of gp150 mutants, but heat-shock induced Sca protein shows no accumulation in these cells. Thus, gp150 is required for Sca to accumulate in N-expressing cells in this assay (Li, 2003).

Sca deletion proteins were used to investigate further how Sca associates with N. Sca comprises an N-terminal coiled-coil, previously found to be sufficient for sca function, and the C-terminal fibrinogen related domain (FReD) that increases the activity of the protein. Flies transgenic for truncated Sca proteins under control of the heat shock promoter were prepared. Neither the ScaDelta41-514 protein encoding the FReD nor the N-terminal sequences encoded by ScaDelta513-773 accumulates in N-expressing cells to the same degree as does full-length Sca. There seems to be more accumulation with the ScaDelta41-514 protein, as if the FReD makes more contribution to Sca accumulating in N-expressing cells (Li, 2003).

These obervations were subsequently revised. Whereas full-length Sca associated with the lattice of pigment cells that express the Notch protein at high levels, ScaDelta41-514 associates predominantly with other cells. The results indicate that the N-terminal 514 amino acids of Sca mediate colocalization with Notch, not the Fibrinogen-Related Domain contained within amino acids 515-774 (Li, 2004).

Gp150 is located in endosomes where it may interact with endocytosed extracellular proteins (Fetchko, 2002). Attempts were made to determine whether Sca protein is also found in endosomes. Although Sca is quantitatively secreted from tissue culture cells, antibodies detect Sca protein only within cells in epithelial tissues. Immunoelectron microscopy studies locate Sca within large intracellular vesicles. There is evidence that at least some such vesicles contain endocytosed Sca (Li, 2003).

Double labelling using markers for particular parts of the endocytic pathway were examined by confocal microscopy. One such marker was Rab7, which associates with the cytoplasmic face of late endosomes. In tubGal4>rab7-GFP eye discs, most of the Sca protein detected by confocal microscopy is located in late endosomes surrounded by rab7-GFP. A second marker was HRS, a protein found in early endosomes and required for maturation of endosomes into multivesicular bodies. HRS and Sca protein distributions do not overlap in eye disc cells, showing that Sca is not stably retained in early endosomes. Gp150 protein also overlaps with rab7-GFP, although Gp150 is found separately from GFP-rab7 in addition, perhaps in other parts of the endosome pathway. Further double labelling showed directly that Gp150 is present in the same late endosomes that are the major location of Sca protein (Li, 2003).

The distribution of Sca protein is altered in clones of cells lacking gp150. The largest and most intensely labeled intracellular bodies are absent, although lower Sca levels are still detected. Similar results were observed in gp150 mutant eye discs. It is not known whether this change corresponds to lower Sca levels within the endosomes, or absence of Sca from the endosomal compartment but presence of lower levels at other locations. In any case, Gp150 is present in the same late endosomes as Sca and partly responsible for Sca concentration or stability there (Li, 2003).

It is suggested that neural cells in the spl mutant mimic a subset of non-neural cells that approach neural fate in wild-type development, and that Sca and Gp150 chiefly contribute to N signaling in such cells. It is proposed that during lateral inhibition to select neural precursor cells, activation of N signaling is only one part of the story. Inactivation of N signaling in cells taking the neural fate is also required. It is suggested that neural cells in which N is inactive have passed through a transient stage in which a low level of incipient N signaling is a normal occurrence prior to neural determination. In this model, Sca and Gp150 normally function to sustain N activity in potential neural cells (or to block or delay N inactivation in potential neural cells). Accordingly, Sca and Gp150 increase N signaling by the same mechanism both in wild-type cells on the verge of neural specification and in spl mutant cells struggling to maintain N inactivity. This model predicts that absence of Sca or Gp150 could lead to N inactivity in too many cells and specification of extra neural precursor cells. This is consistent with the sca and gp150 mutant phenotypes. This model is consistent with the presence of Sca and Gp150 in endosomes, as it posits that they regulate inactive N molecules, not the process of N activation that occurs at the cell surface (Li, 2003).

The model suggests two slightly different routes for the inhibition of neural fate by N. In some cells, activation of N by Dl is sufficient. As a by-product of the protection of future neural cells from Dl, there appear to be other cells that are also at risk for protection from Dl. By antagonizing protection, Sca and Gp150 promote N activity in such cells and prevent too many cells taking neural fate (Li, 2003).

The current data focuses attention on possible roles of endosomes in N signaling. Both Sca and Gp150 proteins are found predominantly in endosomes, where Gp150 is required for Sca location or stability, and for Sca function. This suggests that Sca and Gp150 promote N function, or prevent N inactivation, through an effect on endosomes. Gp150 is thought to be transported to late endosomes directly from the Golgi. Sca is thought to reach late endosomes after uptake from outside the cell, because in cultured cells all the Sca is secreted. Several studies indicate that Sca can be taken up into other cells in vivo. Notably, the subcellular distribution of Sca proteins shows little dependence on dynamin function, suggesting a dynamin-independent mode of uptake (Li, 2003).

The pathway of N activation in which ligands trigger proteolytic cleavages to release the intracellular domain is thought to occur at the cell surface, and none of these reactions is thought to involve endosomes. N activation by trans-endocytosis of the N extracellular domain has been proposed, but this involves endosomes in the signal sending cell, which is not where mosaic analysis finds Gp150 to be required. Endocytosis has been proposed both to downregulate N activity and to promote N activity by removing inactive and inhibitory forms of both N and its ligands from the cell surface. Although the current data are probably consistent with previous models for Sca function in increasing the sensitivity or range of N signaling, both the idea that sca and gp150 are most important in cells where N signaling would otherwise be downregulated, and the location of their products away from the cell surface supports the view that these proteins specifically affect a downregulatory mechanism, rather than acting directly on N activation. Since the ectopic N activity in the spl mutant depends on Dl, it is inferred that sca and gp150 promote ligand-dependent N activation (Li, 2003).

Several new models can be proposed. One model is that either before or after Dl binding, endocytosis reduces the amount of surface N available for activation. Sca and Gp150 might antagonize such endocytosis, or permit endocytosed N to be activated, either by permitting gamma-secretase to act on endocytosed intermediates or by their return to the cell surface. A second model incorporates the observation that in addition to activating N signaling on neighboring cells, N ligands can `cis-inactivate' N signaling in the same cell. Protection of neural cells from N activation by Dl might reflect an increased cis-inactivation in neural cells. In this model, Sca and Gp150 would antagonize cis-inactivation, e.g. by removing Dl or N from cis-inactivatory interactions at the cell surface or in endosomes. Interestingly, Dl is also present in Gp150-positive vesicles. Elevated intracellular Dl levels have been observed in gp150 mutants, suggesting that intracellular Dl may antagonize N signaling (Li, 2003).

One problem for these models is that changes in the cell surface levels of N or Dl have not been detected during the selection of neural cells. It remains possible that there are changes in subsets of the detectable N or Dl proteins that are somehow particularly important for signaling. It is interesting to note that endocytosis is also implicated in N regulation within neural stem cell lineages. Asymmetric divisions during sensory organ lineages deliver Numb protein to particular daughter cells, where Numb then inhibits N signaling through binding to N and to alpha-adaptin, an adaptor for endocytosis via clathrin-coated pits. Although presumed to promote N endocytosis, numb and alpha-adaptin result in no detectable reduction in N protein levels despite blocking N activity. In nematodes, endocytosis has been proposed to permit downregulation of the N-homolog lin-12 by Ras. Perhaps endosomes provide an environment where N signaling components are neither degraded nor removed permanently from the cell surface, but rerouted or modified to change their signaling properties (Li, 2003).

REFERENCES

Baker, N.E., Mlodzik, M. and Rubin, G.M. (1990). Spacing differentiation in the developing Drosophila eye: a fibrinogen-related lateral inhibitor encoded by scabrous. Science 250: 1370-1377. PubMed Citation: 2175046

Baker, N.E. and Zitron, A.E. (1995). Drosophila eye development: Notch and Delta amplify a neurogenesis pattern conferred on the morphogenetic furrow by scabrous. Mech Dev 49: 173-189. PubMed Citation: 7734391

Baonza, A., Casci, T. and Freeman, M. (2001). A primary role for the epidermal growth factor receptor in ommatidial spacing in the Drosophila eye. Curr. Biol. 11: 396-404. 11301250

Benhra, N., (2011). AP-1 controls the trafficking of Notch and Sanpodo toward E-cadherin junctions in sensory organ precursors. Curr. Biol. 21: 87-95. PubMed Citation: 21194948

Burgess, J., et al. (2011). AP-1 and clathrin are essential for secretory granule biogenesis in Drosophila. Mol. Biol. Cell 22: 2094-2105. PubMed Citation: 21490149

Chou, Y.-H. and Chien, C.-T. (2002). Scabrous controls ommatidial rotation in the Drosophila compound eye. Dev. Cell 3: 839-850. 12479809

Doroquez, D. B., Orr-Weaver, T. L. and Rebay. I. (2007). Split ends antagonizes the Notch and potentiates the EGFR signaling pathways during Drosophila eye development. Mech. Dev. 124(9-10): 792-806. PubMed citation: 17588724

Ellis, M.C., Weber, U., Wiersdorff, V. and Mlodzik, M. (1994). Confrontation of scabrous expressing and non-expressing cells is essential for normal ommatidial spacing in the Drosophila eye. Development 120: 1959-1969. PubMed Citation: 7925001

Fetchko, M., Huang, W., Li, Y. and Lai, Z. C. (2002). Drosophila Gp150 is required for early ommatidial development through modulation of Notch signaling. EMBO J. 21: 1074-1083. 11867535

Ghbeish, N. and McKeown, M. (2002). Analyzing the repressive function of ultraspiracle, the Drosophila RXR, in Drosophila eye development. Mech. Dev. 111(1-2): 89-98. 11804781

Graba, Y., et al. (1992). Homeotic control in Drosophila; the scabrous gene is an in vivo target of Ultrabithorax proteins. EMBO J 11: 3375-84

Gurganus, M. C., et al. (1998). Genotype-environment interaction at quantitative trait loci affecting sensory bristle number in Drosophila melanogaster. Genetics 149(4): 1883-1898

Hu, X., Lee, E. C. and Baker, N. E. (1995). Molecular analysis of scabrous mutant alleles from Drosophila melanogaster indicates a secreted protein with two functional domains. Genetics 141: 607-617

Jarman, A.P. et al. (1993). atonal is a proneural gene that directs chordotonal organ formation in the Drosophila peripheral nervous system. Cell 73: 1307-1321

Kametaka, S., et al. (2012). AP-1 clathrin adaptor and CG8538/Aftiphilin are involved in Notch signaling during eye development in Drosophila melanogaster. J. Cell Sci. 125(Pt 3): 634-48. PubMed Citation: 22389401

Kuang, B., Wu, S. C., Shin, Y., Luo L. and Kolodziej, P. (2000). split ends encodes large nuclear proteins that regulate neuronal cell fate and axon extension in the Drosophila embryo. Development 127: 1517-1529. PubMed Citation: 10704397

Lacoste, J., Soula, H., Burg, A., Audibert, A., Darnat, P., Gho, M. and Louvet-Vallee, S. (2022). A neural progenitor mitotic wave is required for asynchronous axon outgrowth and morphology. Elife 11. PubMed ID: 35254258

Lee, E. C., et al. (1996a). The scabrous gene encodes a secreted glycoprotein dimer and regulates proneural development in Drosophila eyes. Mol. Cell. Biol. 16: 1179-1188

Lee, E. C. and Baker, N. E. (1996b). gp300sca is not a high affinity Notch ligand. Biochem. Biophys. Res. Commun. 225(3): 720-5

Lee, E. C., et al. (1998). Functional analysis of the fibrinogen-related scabrous gene from Drosophila melanogaster identifies potential effector and stimulatory protein domains. Genetics 150(2): 663-73

Lee, E. C., Yu, S. Y. and Baker, N. E. (2000). The scabrous protein can act as an extracellular antagonist of notch signaling in the Drosophila wing. Curr. Biol. 10(15): 931-4.

Li, Y., Fetchko, M., Lai, Z. C. and Baker, N. E. (2003). Scabrous and Gp150 are endosomal proteins that regulate Notch activity. Development 130: 2819-2827. 12756167

Li, Y., Fetchko, M., Lai, Z. C. and Baker, N. E. (2004). Corrigendum Scabrous and Gp150 are endosomal proteins that regulate Notch activity. Development 131: 5213

Ma, C., et al. (1993). The segment polarity gene hedgehog is required for the progression of the morphogenetic furrow in the developing Drosophila eye. Cell 75: 927-938

Miyamoto, H., et al. (1995). canoe encodes a novel protein containing a GLGF/DHR motif and functions with Notch and scabrous in common developmental pathways in Drosophila. Genes Dev 9: 612-625

Mlodzik, M., Baker, N.E. and Rubin, G.M. (1990). Isolation and expression of scabrous, a gene regulating neurogenesis in Drosophila. Genes Dev. 4: 1848-1861

Powell, P. A., et al. (2001). Scabrous complexes with Notch to mediate boundary formation. Nature 409: 626-630. 11214322

Rabinow, L., and Birchler, J. A. (1990). Interactions of vestigial and scabrous with the Notch locus of Drosophila melanogaster. Genetics 125(1): 41-50

Renaud, O. and Simpson, P. (2001). scabrous modifies epithelial cell adhesion and extends the range of lateral signaling during development of the spaced bristle pattern in Drosophila. Dev. Biol. 240(2): 361-376. 11784069

Renaud, O. and Simpson, P. (2002). Movement of bristle precursors contributes to the spacing pattern in Drosophila. Mech. Dev. 119: 201-211. 12464433

Singson, A., et al. (1994). Direct downstream targets of proneural activators in the imaginal disc includes genes involved in lateral inhibitory signaling. Genes Dev. 8: 2058-71

St Pierre, S. E., Galindo, M. I., Couso, J. P. and Thor, S. (2002). Control of Drosophila imaginal disc development by rotund and roughened eye: differentially expressed transcripts of the same gene encoding functionally distinct zinc finger proteins. Development 129: 1273-1281. 11874922

date revised: 25 August 2022 

The Interactive Fly resides on the
Society for Developmental Biology's Web server.