Serrate

DEVELOPMENTAL BIOLOGY

Embryo

The Drosophila salivary gland is a simple tubular organ derived from a contiguous epithelial primordium, which is established by the activities of the homeodomain-containing proteins Sex combs reduced (Scr), Extradenticle (Exd), and Homothorax (Hth). EGF signaling along the ventral midline specifies the salivary duct fate for cells in the center of the primordium, while cells farther away from the source of EGF signal adopt a secretory cell fate. EGF signaling works, at least in part, by repressing expression of secretory cell genes in the duct primordium, including fork head (fkh), which encodes a winged-helix transcription factor. Fkh, in turn, represses trachealess (trh), a duct-specific gene initially expressed throughout the salivary gland primordium. trh encodes a basic helix-loop-helix PAS-domain containing transcription factor that has been proposed to specify the salivary duct fate. In conflict with this is the idea that trh specifies salivary duct fate: three genes, dead ringer (dri), Serrate (Ser), and trh itself, are expressed in the duct independently of trh. Expression of all three duct genes is repressed in the secretory cells by Fkh. Ser in the duct cells signals to the adjacent secretory cells to specify a third cell type, the imaginal ring cells. Thus, localized EGF- and Notch-signaling transform a uniform epithelial sheet into three distinct cell types. In addition, Ser directs formation of actin rings in the salivary duct (Haberman, 2003).

In trh mutants, salivary duct cells fail to invaginate and remain clustered on the embryo surface. Based on this phenotype and the loss of expression of other duct genes in trh mutant embryos, it has been proposed that trh is required to establish salivary duct identity. Based on this model, expression of all duct genes would be dependent on trh, even trh itself. Indeed, trh activity is required to maintain trh expression in the trachea. It was asked if trh is required to maintain its own expression in the salivary duct as well. In embryos mutant for two EMS trh alleles, trh¹ and trh², trh RNA was absent from the trachea at late stages. However, trh RNA was still observed at approximately wild-type levels in the salivary duct cells, indicating that trh does not autoregulate in the salivary duct as it does in the trachea (Haberman, 2003).

Two other genes, dead ringer (dri; also known as retained) and Serrate (Ser), are expressed to high levels in the salivary duct. dri encodes an ARID-box transcription factor whose role in the salivary duct has not yet been determined. Ser encodes a ligand for the Notch receptor, whose role in this tissue is also unknown. Expression levels of both dri and Ser are unaffected in trh mutants. Dri protein is present in the uninvaginated salivary duct cells that remain on the surface of trh mutants. Similarly, both Ser RNA and ß-galactosidase expressed under the control of a Ser enhancer (Ser-lacZ) are expressed in salivary duct cells in trh mutants. Thus, trh is neither required for its own expression nor for the expression of at least two other salivary duct genes (Haberman, 2003).

Since dri and Ser are expressed independently of trh, it was asked whether there is any regulatory relationship among the three genes. trh expression is not altered in embryos mutant for dri or Ser. Similarly, Ser expression is not altered in dri mutants, and Dri expression is not altered in Ser mutants. Thus, all three genes are expressed in the salivary duct independently of the other two (Haberman, 2003).

trh is initially expressed throughout the salivary gland, in both duct and secretory cell primordia, but becomes restricted to the duct cells by fkh. It has been suggested that Fkh acts through repression of trh to limit expression of all duct genes to only the ventral preduct portion of the salivary gland primordium. Since it has been shown that expression of at least three genes is trh-independent, it is unclear how their expression is limited to the duct. Whether or not expression of the trh-independent duct genes is affected by Fkh was tested. Since salivary gland cells undergo apoptosis in fkh mutants, the experiments were performed in the background of the H99 deficiency, which blocks apoptosis by removing the apoptosis-activating genes hid, grim, and reaper. As in fkh mutants alone, all salivary gland cells remain on the surface of the embryo in fkh H99 embryos. In these embryos, secretory cells express the secretory marker Pasilla (PS) and Trh is expressed in all salivary gland cells. Similarly, expression of both Dri and Ser expanded into the secretory cells of fkh H99 embryos, suggesting that fkh is required to prevent secretory cell expression of multiple duct genes independently. Expression of all three genes is also observed throughout the salivary gland primordium of fkh mutants without the H99 deficiency, demonstrating that the observed expression profiles are not affected by the H99 deficiency. Also, expression of all of these genes is unchanged in H99 homozygous embryos, further indicating that the changes in gene expression are due to fkh (Haberman, 2003).

Given the role of trh in salivary duct morphogenesis, what is the role of the two Trh-independent salivary duct genes? Staining of dri mutants with the duct markers Trh, Ser, or Crb did not reveal any overt morphological changes from wild-type embryos. Staining of Ser mutants with Dri revealed only a subtle, partially penetrant defect, where the distal ends of the individual ducts are slightly enlarged. Differences between Ser and wild-type embryos in the distal ends of the salivary ducts are more apparent with staining for cytoplasmic Ser-lacZ, which reveals that the ends of the individual ducts are splayed in the region where they contacted the secretory cells (Haberman, 2003).

To test for any potential cell fate changes at the ends of the individual ducts in Ser mutants, expression was analyzed of several salivary gland markers. By coimmunofluorescence with Ser-lacZ, it was found that the cells at the duct ends still express Dri and do not express the secretory cell markers dCrebA and PS. Thus, the change in duct morphology is likely not due to a change in duct cell fate. No change in staining for the phosphorylated form of histone H3 was detected, indicating that loss of Ser does not cause a change in cell proliferation (Haberman, 2003).

Ser transcripts are first detected in the duct cell primordia during embryonic stage 11, when most of the salivary gland precursors are still on the embryo surface. Notch, the gene encoding the receptor for Ser, is transiently upregulated in the secretory primordia prior to invagination. At this stage, the cells are arranged so that Ser could signal to the approximately 10-11 secretory cells in direct contact with either side of the duct primordium. The enlarged distal ends of the ducts of late stage Ser mutants suggest that the duct cells are either in contact with more cells or with larger cells than in wild-type. Thus, Ser could be signaling to the adjacent secretory cells to either undergo programmed cell death or to be smaller than the other secretory cells. If the role of Ser signaling in this system is to induce programmed cell death, it is expected that blocking programmed cell death would give the same phenotype as observed in Ser mutants. Staining with duct markers reveals that the ends of the individual ducts are normal in H99-deficient, indicating that the Ser phenotype is probably not due to a defect in apoptosis (Haberman, 2003).

Another explanation for the enlarged distal ends of the ducts in Ser mutant embryos is that the cells they contact are larger than in wild-type embryos. In the mature larval salivary gland, the ends of the duct are in direct contact with the salivary gland imaginal ring. While the duct and secretory cells are large and polytenized at larval stages, the imaginal ring cells are small and diploid. If, in Ser embryos, the imaginal cells were absent or transformed into secretory cells, the duct connected to them would have to spread wider to make a complete connection. Unfortunately, adequate markers for the embryonic salivary gland imaginal ring are not currently available. Moreover, Ser mutant larvae do not survive to the third larval instar stage, where the imaginal ring population is easily distinguished from the duct and secretory cell populations by the dramatic differences in nuclear size. Nonetheless, differences in nuclear size can be distinguished as early as the second instar stage in wild-type larvae, and Ser mutants do survive to the second larval instar. Although the Ser mutant larvae are smaller and have smaller cells than wild-type larvae at the same stage, distinctions between the different salivary gland cell populations are readily observed (Haberman, 2003).

To test for the presence of the imaginal ring in Ser mutants, the salivary glands were dissected from wild-type and Ser mutant second instar larvae and stained with Hoechst or Sytox to visualize nuclei. In wild-type and Ser heterozygous glands, the diploid imaginal ring nuclei are clearly present as small nuclei positioned between the polytenized duct and secretory nuclei. In Ser homozygous glands, the diploid nuclei are absent and all salivary gland nuclei are large and polytenized. Second instar salivary glands were simultaneously stained with Hoechst or Sytox and Texas Red-conjugated phalloidin to visualize the cortical F-actin of the salivary gland cells. Confocal optical sections of wild-type and Ser heterozygous glands revealed small imaginal nuclei surrounded by a tight ring of actin, demonstrating the smaller size of the imaginal cells compared with the duct and secretory cells. Ser homozygous glands are missing these small cells. Thus, the imaginal ring cells are missing in second instar salivary glands of Ser mutants, potentially accounting for the enlarged distal tips of the individual ducts observed in late embryonic stages (Haberman, 2003).

Phalloidin staining of second instar salivary glands reveals a unique organization of F-actin in the salivary ducts. While faint levels of cortical actin are observed in salivary duct cells, intense phalloidin staining is observed in bands that form regular rings around the lumen of the salivary duct and imaginal ring. Three-dimensional reconstruction of the salivary duct reveals that these actin rings encircle the entire salivary duct and imaginal ring of wild type and Ser heterozygous glands. These actin rings were missing from the ducts of Ser homozygous second instar glands. The trachea had similar phalloidin-staining rings in every branch examined, although the rings in the trachea are unaffected by mutations in Ser (Haberman, 2003).

Thus, trh is not the primary determinant of duct cell fate. Instead, these findings support an earlier model in which trh is required for the morphogenesis of the tubes that comprise the salivary duct, in keeping with its role in the trachea and filzkörper. In all three of these tissues, the primordial cells fail to invaginate and form tubular organs, although other tissue-specific markers are still expressed. Thus, it is expected that, in the salivary duct, Trh regulates expression of genes required for tube morphogenesis, as has been shown for the trachea. Indeed, btl, which encodes the FGF-receptor required for tracheal branch migration, is a Trh target in both the trachea and salivary duct, although its role in the salivary duct is unclear, since the loss of btl does not overtly affect salivary gland formation (Haberman, 2003).

A role for Fkh as a master regulator of secretory cell fates has been rejected by multiple groups. A model is proposed where the salivary gland fate and the distinction between duct and secretory fate within the primordium is initiated by the coordinate system provided by the early patterning genes. Scr/Exd/Hth in the absence of Dpp- and EGF-signaling specifies the secretory cell fate, and Scr/Exd/Hth in the absence of Dpp-signaling and in the presence of EGF-signaling specifies the salivary duct fate. As a consequence of this combinatorial system for cell fate specification, multiple different genes are activated in the different salivary gland cell types. It is the combined activities of these downstream genes that make secretory cells different from duct cells. Moreover, since Scr and hth expression disappears from the salivary gland quite early, the downstream target genes must maintain as well as elaborate on these cell fate decisions (Haberman, 2003).

fkh has many roles in secretory cell development. Fkh prevents secretory cell apoptosis, mediates apical constriction during invagination, regulates its own expression, maintains expression of dCrebA, and regulates expression of the ecdysone-stimulated glue genes sgs3 and sgs4. Fkh has been found to represses expression of all tested duct genes in the secretory cells. In fkh mutants, trh, Ser, and dri are expressed throughout the salivary gland primordium in both duct and secretory cells. It is unclear whether Fkh directly regulates duct gene expression or regulates expression through some currently unidentified upstream activator(s). The 4-kb Ser salivary duct enhancer used in these studies contains several potential Fkh binding sites, indicating that Fkh repression of Ser could be direct. Fkh repression of duct gene expression suggests a role for Fkh in reinforcing the secretory cell fate. Fkh is required to maintain the distinction between duct and secretory primordia that is initially established by EGF-signaling. First, EGF-signaling initiates the distinctions between duct and secretory cells by blocking expression of secretory-specific genes in the duct primordium. Then, the genes whose duct expression is blocked by EGF-signaling, specifically fkh, maintain this distinction by repressing duct gene expression and maintaining their own expression, thus sharpening the boundaries between duct and secretory primordia by interpreting the gradient of EGF-signaling into a binary cell fate decision (Haberman, 2003).

The gradient of EGF signal from the ventral midline initiates early differences in duct versus secretory cell populations. The boundary then becomes more firmly established by Fkh. It is proposed that the salivary gland imaginal ring cells are then specified at the boundary between the duct and secretory cells in the salivary gland primordium. While the assay for imaginal ring specification analyzed salivary glands two days after embryogenesis, two lines of evidence suggest that imaginal ring specification occurs during embryogenesis. Ser is expressed in the salivary duct cells beginning at embryonic stage 11, when the duct cells and the adjacent secretory cells are still on the surface of the embryo. Notch, the receptor for Ser, is transiently upregulated in the secretory cells at stage 11. Thus, at this stage, the gland has high-level expression of ligand in the duct primordia and high-level expression of the receptor in adjacent secretory cells and, therefore, this is when signaling is likely to occur. Furthermore, the salivary ducts of Ser mutants have abnormal distal ends that can be observed in late stage embryos, indicating that a defect in salivary gland formation has already occurred. While it was not possible to assay for imaginal ring formation in the embryo due to a lack of markers for imaginal ring cells, the evidence suggests that Ser acts during embryogenesis to specify the imaginal ring. Nonetheless, the possibility cannot be ruled out that Ser specifies the imaginal ring at later embryonic stages when the salivary gland has internalized and when the salivary gland cells are in their final relative positions (Haberman, 2003).

It was not possible to directly test the role of Notch in imaginal ring specification because of the other role Notch plays in the ventral ectoderm. In Notch mutants, all the cells of the ventral ectoderm adopt a neuronal fate, as opposed to an epithelial fate, and salivary glands do not form due to an absence of epithelial precursors. This role for Notch in protecting salivary gland cells from becoming neuronal appears to continue even after the salivary gland is specified. In mutants carrying the Ser allele Beaded of Goldschmidt (BdG), which encodes a secreted dominant-negative form of Ser, salivary gland cells are missing, even though BdG is not expressed in or near the salivary gland until the salivary duct cells are specified (Haberman, 2003).

Since Ser directs secretory cells to adopt the imaginal ring fate, it was expected that ectopic Ser would transform more secretory cells into imaginal ring cells. However, ectopic expression of Ser in secretory cells does not have any discernable effect on salivary glands. This lack of phenotype is attributed to the observation that Notch signaling in the embryo depends on relative levels of ligand instead of absolute or threshold levels. Overexpression of Ser using a heat-shock promoter has no effect on wild-type cuticles, but overexpression of Ser in Ser mutants causes malformations of the mouth hooks and gut. Similarly, overexpression of Ser in the secretory cells may fail to disrupt the relative levels of Notch signaling in the salivary gland, resulting in a wild-type gland (Haberman, 2003).

Ser-mediated specification of the salivary gland imaginal ring is another example of a role for Notch signaling mediating boundary formation, the process of specifying a new cell type between two adjacent groups of cells by intercellular signaling. However, specification of the imaginal ring is different from the Notch-mediated boundary formation that occurs in imaginal discs. In the eye and wing imaginal discs, a boundary is formed at the interface between Dl and Ser-expressing cells. Similarly, both ligands are also involved in joint specification in the leg discs. In contrast, only Ser appears to be involved in specification of the imaginal ring. This situation is analogous to that of the embryonic hindgut, where Delta is the only ligand required to specify the boundary cells in the large intestine. Thus, boundary formation in the salivary gland and hindgut represent a new class of boundary formation mediated by unidirectional Notch signaling (Haberman, 2003).

The salivary duct and trachea have regularly spaced actin rings encircling their lumena. These actin rings are unlike actin structures described in other tissues, suggesting that they may be part of unique structures found only tubular tissues. The rings in the trachea may be associated with the taenidia, a series of epithelial folds along the lumen of the trachea that are postulated to give the trachea strength and flexibility. Both the actin rings and the taenidia appear to corkscrew around the lumen of the trachea. Such structures could also give the salivary duct the strength and flexibility it needs to carry secretory products to the larval mouth (Haberman, 2003).

Ser is required for actin rings to form in the salivary duct. While Notch signaling usually affects cell fate decisions, it has been shown to control cellular behaviors without affecting cell fate. Notch signaling directs neurons to arrest or retract neurites, a process involving changes in the actin. Also, Delta1 increases the cohesiveness and reduces the motility of cultured human keratinocytes. Thus, Ser could regulate the cytoskeleton of salivary duct and imaginal ring cells independent of cell fate specification. Though the possibility cannot be ruled out that the loss of actin rings is a secondary effect of the general growth defects observed in Ser mutants, a role in actin ring formation would explain the prolonged expression of Ser in the salivary duct after imaginal ring specification. The fact that Ser does not control actin ring formation in the trachea correlates with the observation that Ser is only expressed in a subset of the trachea, while actin rings were found in every tracheal branch examined. However in the salivary gland, Ser appears to control both the imaginal ring fate and the cytoskeleton of the salivary duct (Haberman, 2003).

Larval and Pupal

Serrate is required for wing and haltere morphogenesis. Targeted ectopic expression of Serrate in imaginal discs results in regionally restricted induction of cell proliferation in ventral tissues in the case of wings and halteres (Speicher, 1994). Serrate is induced in dorsal wing cells at boundaries between dorsal and ventral cell compartments. Serrate in turn triggers the expression of genes involved in wing growth and patterning on both sides of the dorsal-ventral boundary. Ectopic Serrate induces wingless, cut and vestigial expression. Both disc and adult wing outgrowth are restricted to the ventral compartment in Serrate ectopic expression (Kim, 1995).

Molecular-genetic analyses of pattern formation have generally treated imaginal discs as single epithelial sheets. Anatomically, however, discs comprise a columnar cell monolayer covered by a squamous epithelium known as the peripodial membrane. During development, peripodial cells signal to disc columnar cells via microtubule-based apical extensions. Ablation and targeted gene misexpression experiments demonstrate that peripodial cell signaling contributes to growth control and pattern formation in the eye and wing primordia. These findings challenge the traditional view of discs as monolayers and provide foundational evidence for peripodial cell function in Drosophila appendage development (Gibson, 2000).

Imaginal discs are a favored system for understanding how fields of cells can autonomously regulate growth and pattern formation. Originating as invaginations in the embryonic epidermis, discs grow into flattened sacs comprising two distinct cell layers: a columnar epithelium and a peripodial membrane. Little is known about peripodial cells, but they are thought to regulate metamorphic events. As epithelial invaginations, discs must evert during the pupal stages such that the appendages lie on the external surface of the adult fly. Eye peripodial cells have been shown to give rise to parts of the adult head and contraction of the peripodial membrane during metamorphosis is required for eye disc eversion. A similar contraction-eversion function has been reported for wing disc peripodial epithelia in the Lepidopteran Manduca sexta. Subsets of peripodial 'edge' cells employ the JNK signaling cascade to regulate the process of metamorphic interdisc fusion in Drosophila (Gibson, 2000 and references therein).

Previous studies have not addressed possible function of peripodial membranes during early growth and patterning. The adult appendages are almost entirely derived from columnar cells and consequently appendage development has been considered a two-dimensional problem with intercellular signaling restricted to idealized columnar cell monolayers. The discs-as-monolayers concept likely originated from two-dimensional fate maps produced 30 years ago. Peripodial Hedgehog signaling has been shown to induces engrailed expression in leg disc columnar cells during fragmentation-induced regeneration. This observation was an initial indication that peripodial and columnar cells might also interact during normal development. This study now shows that wing and eye disc peripodial cells form microtubule-based 'translumenal' extensions that traverse acellular space and terminate on the surface of developing disc columnar epithelia. These structures are plainly suggestive of communication between peripodial and columnar cells. Consistent with this, ablation experiments have shown that peripodial cells are required for pattern formation in the disc columnar epithelia (Gibson, 2000 and references therein).

The lipophilic membrane marker DiI has been used to describe the morphology of live wing peripodial cells. Optical sectioning obtained by confocal microscopy reveals 'translumenal extensions' that traverse a cellular space (the disc lumen) and terminate on the surface of the columnar epithelium. These structures initiated on the apical surface of each peripodial cell, range from 5–30 µm in length, and often appear to be highly vesiculated. In general, only one extension has been observed per cell. Similar structures have been observed in eye, leg, and haltere discs, suggesting a generalized morphological basis for translumenal signaling. The lumenal cavity of imaginal discs does not form until the early third instar, so it is unlikely that these structures are present during earlier stages and indeed, they were not detected. Further, not all peripodial cells produced extensions. In the wing, extensions are numerous in the presumptive dorsal hinge and notum regions but absent from the presumptive wing blade where peripodial and columnar epithelia are in direct contact (Gibson, 2000).

The subcellular morphology of peripodial cells further supports a translumenal signaling function. In wing discs, peripodial cell nuclear membranes are associated with a membranous organelle that tapers into a funnel-shaped sac within each cellular extension. It is speculated that this unusual internal membrane represents a mechanism for targeting specific RNAs or proteins to the translumenal extensions. Mitochondria labeled with Rhodamine 123 localize to the extensions and are suggestive of locally enhanced ATP consumption. To observe peripodial microtubule networks, a peripodial Gal4 driver (c311-Gal4) was identified and the Gal4/UAS system was used to direct expression of a fluorescently tagged microtubule binding protein (UAS-tau-GFP). In live analysis, translumenal extensions are packed with microtubules, demonstrating that they are not cytonemes, threadlike processes observed in disc columnar cells. Based on these observations, it is concluded that peripodial cells possess structurally specialized translumenal extensions. Evidence is provided that these structures mediate peripodial-columnar signaling during development (Gibson, 2000).

In the eye disc, peripodial translumenal extensions correlate with the position of the morphogenetic furrow, a wave of cell division and photoreceptor cluster formation that sweeps across the columnar epithelium during the latter third instar and early pupal development. To test peripodial function in furrow progression, the peripodial membrane was surgically removed during the late third instar and 'naked' eye columnar epithelia were cultured in vivo overnight. This treatment abolishes the mitotic waves normally associated with the furrow and causes a marked reduction in photoreceptor cluster formation relative to intact cultured controls. These results are consistent with the notion that furrow progression requires peripodial signaling (Gibson, 2000).

A parallel experiment was performed in situ by genetic ablation of the peripodial epithelium. The c311-Gal4 driver is peripodial membrane specific in the eye disc and c311-Gal4 and the flp-out technique could be combined to ablate eye peripodial cells with a toxic UAS <w+<RicinA transgene. Peripodial expression of RicinA was induced in late third instar animals and then eye cuticle was recovered from pupae with uneverted eyes. In all cases, peripodial ablation results in significant eye size reduction and severe pattern defects including square ommatidia. The observed size reductions are consistent with a failure in furrow progression (Gibson, 2000).

Further ablation experiments have revealed that peripodial function is not unique to the eye. Surgical ablation of wing peripodial membranes results in specific loss of the specialized bristles and hairs along the wing margin. Translumenal extensions are not observed in the presumptive wing margin region where there is no detectable lumenal cavity between the opposing peripodial and columnar epithelia. Therefore, it is inferred that direct cellular contact mediates peripodial-columnar signaling in the absence of visible translumenal processes (Gibson, 2000).

Together, these ablation experiments reveal a novel requirement for intact peripodial epithelia in patterning both eye and wing primordia. However, it is important to note that the phenotypes observed following ablation could be due to myriad effects other than disruption of signaling through translumenal extensions. For example, ablation of the peripodial epithelium could result in direct exposure of columnar cells to hemolymph or dilution of critical extracellular factors normally concentrated in the enclosed disc lumen. To circumvent the many caveats associated with cell ablation, the Gal4/UAS system was employed to inactivate functional aspects of peripodial cells while maintaining the integrity of the peripodial membrane (Gibson, 2000).

While it is anticipated that microtubule-associated motor function is required to transport signals within the extensions, the observed phenotypes might also result from an indirect structural defect or failure to form extensions at all. Glued mutant flies have reduced, rough eyes and ubiquitously expressed dominant-negative Glued causes columnar-cell mitotic delays (resulting in an increase in mitotic figures), as well as rough and reduced eyes. These phenotypes were not observed in Gal4 driven dominant negative Glued flies and it is therefore likely that the distinctive slow/arrested furrow phenotype reported here is peripodial cell specific (Gibson, 2000).

Several complex genetic networks are known to govern Drosophila eye development. If indeed the peripodial membrane signals to the disc columnar epithelium, then some described eye mutant phenotypes would be expected to derive from failed peripodial function. The action of Fringe (Fng), a Golgi-localized glycosyltransferase known to affect eye development on several levels has been explored. A ventrally restricted Fng expression boundary is thought to regulate eye disc growth; abolishing this boundary by ubiquitous overexpression of UAS-fng in the second instar severely reduces eye size. Surprisingly, clonal analysis demonstrates that while fng is required for growth of the ventral eye, this requirement lies somewhere outside the eye columnar epithelium (Papayannopoulos, 1998). A test was performed to see whether disc growth is sensitive to peripodial fng activity (Gibson, 2000).

fng-LacZ is expressed in a subset of eye peripodial cells in the first and second larval instars, growth stages where peripodial and columnar epithelia appear to be in direct contact. This is consistent with the notion that fng functions in peripodial cells. Further, ectopic expression of UAS-fng throughout the peripodial membrane (using c311-Gal4) results in dramatically reduced eye size and predominantly square-shaped ommatidia similar to those observed following genetic ablation of the peripodial membrane. The eye size reduction is consistent with the idea that proliferative growth of the eye disc is regulated by a Fng expression boundary in the peripodial membrane. An alternative interpretation is that overexpression of Fng in the peripodial membrane interfers with furrow progression, resulting in a smaller eye. However, it has been noted that ectopic fng causes size reduction of the eye disc prior to furrow movement (Papayannopoulos, 1998); thus, the former interpretation is favored. The square ommatidia phenotype is more difficult to explain, but may reflect variant cell numbers within each ommatidia or some other failure in retinal morphogenesis. It is noted that an identical phenotype is obtained by overexpression of a dominant-negative form of the Notch ligand Serrate in a broad domain of the eye. In both cases, uniformly square ommatidia are arranged in a highly regular array (Gibson, 2000 and references therein).

It was questioned whether the small eyes seen in c311-Gal4;UAS-fng flies are due to a peripodial-specific effect since the same phenotype is obtained when UAS-fng is expressed under the control of eyeless-Gal4. How could eyeless- and c311-Gal4 elicit the same fng-dependent phenotype? Intriguingly, eyeless-Gal4 drives UAS-GFP expression in eye columnar and peripodial cells throughout larval development consistent with the idea that peripodial fng activity regulates disc growth (Gibson, 2000).

During normal development Fng acts as a glycosyltransferase that modifies the ligand preference of the extracellular receptor Notch. Accordingly, the results of the previous section indicate that reception of signals via Notch could be a key feature of peripodial cell function in retinal patterning. Consistent with this idea, Notch is expressed in peripodial cells. The involvement of the Notch pathway in peripodial cell signaling was further explored by examining the distribution of its ligands, Delta (Dl) and Serrate (Ser). Dl is expressed extensively in the eye disc columnar epithelium and in some limited regions of the peripodial membrane. In contrast, high levels of Ser are observed in subcellular vesicles throughout the eye peripodial epithelium while only minimal levels are detected in the columnar cell layer. Since Ser is required for eye development, it is proposed that Ser is a strong candidate for a peripodial-to-columnar signal. Consistent with this, peripodial-specific expression of a secreted, dominant-negative form of Serrate [UAS-Ser(s)] results in flies with reduced eyes and highly irregular ommatidial patterning (Gibson, 2000).

While highly suggestive, it is noted that the c311 Gal4;UAS-Ser(s) phenotype does not indicate whether the defective Ser signal travels through translumenal extensions or is released into the lumenal cavity. Further, these results do not allow the authors to state where the dominant-negative ligand is acting. The findings do, however, clearly demonstrate that peripodial expression of a defective N ligand is sufficient to disrupt development of the eye columnar epithelium. Combining this observation with the results of fng overexpression, the peripodial localization of vesicular Ser in wild-type discs, and the requirement for Ser in eye development, it is proposed that components of the Notch signaling pathway participate in translumenal signaling. Presently the data do not directly implicate Ser in this process. However, a peripodial function of Ser could explain an outstanding paradox regarding its role in eye development: Ser mutants have rough and reduced eyes, but Ser^- clones in the columnar epithelium have no apparent phenotype (Gibson, 2000 and references therein).

The identification of peripodial translumenal extensions is new evidence that specialized cellular processes effect long-range signaling during development. Interestingly, many cell signaling pathways have been genetically linked to components of the cytoarchitecture. These genetic connections take on new significance in light of evidence for long-range filopodia with specialized cytoskeletal organization. The characterization of a novel cellular extension in Drosophila will permit detailed analysis of the formation and function of long-range extensions on a molecular and genetic level. Peripodial function in disc development is further revealed by the results show that several secreted signaling molecules (including wingless, decapentaplegic, and hedgehog) are expressed in eye peripodial cells where they function to regulate the early development of the eye columnar epithelium (Cho, 2000). Together, these studies necessitate a reconsideration of the two-dimensional models for disc morphogenesis and suggest a new parallel between appendage development in Drosophila and vertebrates. It now appears that limb and retinal development require functional interactions between opposing epithelial sheets in both systems (Gibson, 2000).

Notch activation at the midline plays an essential role both in promoting the growth of the eye primordia and in regulating eye patterning. Specialized cells are established along the dorsal-ventral midline of the developing eye by Notch-mediated signaling between dorsal and ventral cells. D-V signaling in the eye shares many similarites with D-V signaling in the wing. In both cases an initial asymmetry is set up by Wingless expression. Both Eye and wing cells then go through a distinct intermediate step: in the wing, Wingless represses the expression of Apterous, a positive regulator of fringe (fng) expression; in the eye, Wingless promotes the expression of mirror (mrr), which encodes a negative regulator of fringe (unpublished observations of McNeill, Chasen, Papayannopoulos, Irvine, and Simon, cited by Papayannopoulos, 1998). Both wing and eye cells share a Fng-Ser-Dl-Notch signaling cassette to effect signaling between dorsal and ventral cells and establish Notch activation along the D-V midline. Local activation of Notch leads to production of diffusible, long-range signals that direct growth and patterning, which in the wing include Wingless, but in the eye remain unknown. At least one downstream target of D-V midline signaling, four jointed (fj), is also conserved. four jointed is also expressed in the wing and its expression there is indirectly influenced by Notch (Papayannopoulos, 1998 and references).

During early eye development, fringe is expressed by ventral cells. This expression appears to be complementary to that of the dorsally expressed gene mrr. During early to mid-third instar, additional expression of fng appears in the posterior of the eye disc. This line of posterior fng expression is just in front of the morphogenetic furrow and moves across the eye ahead of the furrow. In the wing disc, Dl and Ser induce each other's expression, and become up-regulated along the D-V border where they can productively signal. Dl and Ser are also preferentially expressed along the D-V midline during eye development. Ser expression, like fng expression, is complementary to that of mrr, whereas Dl expression partially overlaps that of mrr. The spatial relations among fng, Ser, and Dl expression in the eye are thus similar to those in the wing, although in the wing, their expressions are inverted with respect to the D-V axis (Papayannopoulos, 1998).

The four-jointed gene is expressed in a gradient during early eye development, with a peak of expression along the D-V midline. Together with Ser and Dl, Fj serves as a molecular marker of midline fate. Ubiquitous expression of Fng during early eye development, generated by placing fng under the control of an eyeless enhancer, eliminates detectable expression of Ser and Dl along the midline. Conversely, misexpression of Fng in clones of cells, can result in ectopic expression of Ser and fj that is centered along novel borders of Fng expression in the dorsal eye. Ectopic Ser and fj expression can also be detected along the borders of fng mutant clones in the ventral eye. These observations show that Fng expression borders play an essential and instructive role in establishing a distinct group of cells along the D-V midline of the developing eye. Animals with reduced fng activity have small eyes. Moreover, ubiquitous fng expression also results in a dramatic loss of tissue. Tissue loss is detectable in the developing imaginal disc, before the morphogenetic furrow moves across the eye. Moreover, eye loss is observed when fng is ectopically expressed during early development, but not when fng is ectopically expressed behind the furrow. These observations indictate that a Fng expression border is required for eye growth, specifically during early eye development (Papayannopoulos, 1998).

Fng differentially modulates the action of Notch ligands in the eye just as it does in the wing. Clones of cells ectopically expressing Dl can induce Ser expression in ventral, Fng-expressing cells, but not in dorsal cells. Fng alone can induce Ser expression in dorsal cells, but only near the D-V midline. When Fng and Dl are co-misexpressed, Ser expression can be induced in dorsal cells even when the clones are far from the D-V midline. Clones of cells ectopically expressing Ser are able to induce increased expression of Dl in dorsal cells but not in ventral, Fng expressing cells. However, if Ser is ectopicallly expressed in fng mutant animals, it can induce Dl expression in ventral cells (Papayannopoulos, 1998).

Notch function is also necessary for normal D-V midline cell fate. The ability of Ser and Dl to induce one another's expression indicates that the expression of either one is a marker for Notch activation in the eye. Analysis of loss-of-function mutants of Notch and its ligands, as well as ectopic expression studies, indicate that Notch activation also regulates eye growth. Several observations indicate that the D-V midline is the focus of Notch activation required for growth. Moreover, the midline corresponds to a fng expression border, which is essential for growth and modulates Notch signaling during early eye development. Because local activation of Notch has long-range effects on growth and four-jointed expression, it is inferred that Notch induces the expression of a diffusible growth factor at the midline. Notch activation influences ommatidial chirality. fng mutant clone borders within the ventral eye can be associated with reversals of ommatidial chirality, whereas mutant clones that cross the D-V midline disrupt the normal equator. The equatorial bias in the influence of ectopic Notch activation implies that the equator is the normal source of a Notch-dependent, chirality-determining signal (Papayannopoulos, 1998).

The teashirt (tsh) gene has dorso-ventral (DV) asymmetric functions in Drosophila eye development: promoting eye development in dorsal and suppressing eye development in ventral regions by Wingless mediated Homothorax (HTH) induction. A search was carried out for DV spatial cues required by tsh for its asymmetric functions. The dorsal Iroquois-Complex (Iro-C) genes and Delta (Dl) are required and sufficient for the tsh dorsal functions. The ventral Serrate (Ser), but not fringe (fng) or Lobe (L), is required and sufficient for the tsh ventral function. It is proposed that DV asymmetric function of tsh represents a novel tier of DV pattern regulation, which takes place after the spatial expression patterns of early DV patterning genes are established in the eye (Singh, 2004).

Dl is expressed preferentially in the dorsal eye. Misexpression of Dl anterior to morphogenetic furrow in the hairy domain (hairy>Dl) accelerates photoreceptor differentiation but does not result in eye enlargement. bi>Dl (bifid-Gal4, UAS Delta) does not affect eye size. However, coexpression of tsh with Dl (bi>tsh+Dl) results in eye enlargements on both dorsal and ventral margins. Act>tsh+Dl clones in both dorsal- and ventral-eye also causes enlargements. These results suggest that Dl can provide the dorsal cue for tsh function (Singh, 2004).

Dl function was blocked by a dominant-negative form of DL, DL^DN. bi>Dl^DN causes reduction of eye on both dorsal and ventral margins whereas coexpression of tsh and Dl^DN (bi>tsh+Dl^DN) further enhances this phenotype. A dorsal Act>tsh+Dl^DN clone suppresses eye development. Act>tsh+Dl^DN clones also non-autonomously suppress eye development, a phenotype also seen in Act>Dl^DN clones. These phenotypes suggest that Dl is also required for the dorsal function of tsh in eye. In the absence of Dl, tsh exerts its ventral function in dorsal eye (Singh, 2004).

Ser is preferentially enriched in the ventral eye until late second instar of larval development. Misexpression of Ser (bi>Ser) does not suppress eye development whereas coexpression of tsh+Ser(bi>tsh+Ser) suppresses eye development on both dorsal and ventral margins. Despite the suppression of photoreceptor differentiation, bi>tsh+Ser eye disc shows overall enlargement. The adult eyes were also enlarged and folded despite the suppression of photoreceptors differentiation on the dorsal and ventral margins. These results suggest that Ser can provide the ventral cue for the eye suppression function of tsh but does not affect its early function in promoting growth (Singh, 2004).

The dominant-negative form of Ser, Ser^DN was used to block Ser function. In bi>Ser^DN, the eyes are suppressed on both dorsal and ventral margins. This phenotype is partially blocked in bi>tsh+Ser^DN eye. Similar results were observed in Act>tsh+Ser^DN clones. Thus, tsh requires Ser for its ventral function (Singh, 2004).

These results show that tsh requires several early DV eye patterning genes for its dorsal and ventral specific functions in the eye. The requirement for these DV patterning genes is specific, because not all the DV patterning genes have similar effects. Eye suppression by tsh is prevented in the dorsal eye region. This function requires the normal dosage of both Iro-C and Dl genes, because the reduction of either Iro-C or Dl allows tsh to suppress eye development even in the dorsal eye. However, when ectopically expressed in the ventral eye, either Iro-C genes or Dl can block the ventral function of tsh, suggesting that the two genes may play similar roles (Singh, 2004).

The genes involved in early DV eye patterning can be categorized in two classes: (1) genes that are preferentially expressed in dorsal (e.g., Iro-C, Dl) or ventral (e.g. Ser, fng) and (2) genes that are uniformly expressed but function only in one domain (e.g., L). It is proposed that tsh comprises a new class of genes, which is expressed symmetrically but perform asymmetric functions in dorsal and ventral eye (Singh, 2004).

Although tsh is expressed ubiquitously in the early eye disc, the DV asymmetric functions of tsh can be uncovered only after the expression of early DV patterning genes is established. These results suggest that early expression of tsh may be responsible for its growth function only, whereas for the DV asymmetric functions the expression of early DV patterning genes is a prerequisite. Therefore, TSH function in eye represents a new tier of DV pattern regulation, which functions in interpreting the DV spatial cues in eyes. It would thus be interesting to identify other members of this class. Interestingly, two orthologs of tsh have been identified in mouse but their function in eye is not yet known. Since there is evolutionary conservation in patterns of gene expression and functions, it would be interesting to look for the role of tsh during eye development in higher organisms (Singh, 2004).

Effects of Mutation or Deletion

Asymmetric divisions allow a precursor to produce the four distinct cells of Drosophila sensory organ lineages (SOLs). The sensory organ precursor (SOP) first divides into two different secondary precursor cells, IIA and IIB, which gives rise to one shaft-producing cell (trichogen) and one socket-producing cell (tormogen), and one neuron and one sheath cell (thecogen), respectively. Although this process requires cell-cell communication via the Notch (N) receptor, mitotic recombination that removes the N ligand Delta (Dl) or Serrate (Ser) in the SOL has mild or no effect. N mutant clones generated on the central region of the adult scutum are devoid of any external bristle structures, such as shafts and sockets, similar to the Nts mutant phenotype at a restrictive temperature. Whereas loss of N function during the process of lateral inhibition produces supernumerary SOPs, this balding phenotype is probably due to the requirement of N in asymmetric divisions. Without N activity the supernumerary SOPs divide symmetrically, giving rise to two IIB cells and, consequently, no external sensory structures. Dl clones typically produce a tuft of densely packed bristles in the interior of the clone. These tufts of bristles are likely due to a failure of lateral inhibition, resulting in overproduction of SOPs. The presence of the external bristle structures in these Dl mutant clones indicates that, unlike N clones, most of the supernumerary SOPs in the Dl mutant clones produce IIA cells that divide to form shaft and socket cells. Clones homozygous for three Ser null alleles produce normal external bristle structures. In contrast, clones with loss of both Dl and Ser function produce epidermal cells but not external bristle structures. This balding phenotype is clearly different from the phenotypes of the Dl or Ser mutant clones but is indistinguishable from that of N mutant clones, suggesting that Ser and Dl have overlapping functions in the N signaling pathway. Dl and Ser also have redundant functions in patterning wing veins. In contrast, Dl and Ser are known to serve distinct functions in specifying dorsal-ventral compartment boundary of the wing (wing margin). Ser in dorsal cells signals to N in ventral cells, and Dl in ventral cells signals to N in dorsal cells. For Dl and Ser to provide distinct signals from one compartment to the other without generating signals among cells within the same compartment, it may be necessary to involve other factors such as those encoded by the dorsally expressed gene fringe (fng), which inhibits a cell's ability to respond to Ser and potentiates a cell's response to Dl. It is concluded that Dl and Ser are redundant in mediating signaling between daughter cells to specify their distinct sensory organ cell fates (Zeng, 1998). Ser mutant larva often die, exhibiting a failure to differentiate the anterior spiracles, poorly developed mouth-hooks and a severe reduction in the size of wing and haltere primordia. The few mutants that successfully pupate develop into pharate adults, almost completely lacking in wings and halteres (Speicher, 1994). The dominant Ser mutation causes a gap in the posterior wing tip and margin, and a portion of the blade (Jack, 1992).

Flies hetero- or homozygous for the dominant mutation SerD exhibit scalloping of the wing margin due to cell death during pupal stages. SerD is associated with an insertion of the transposable element Tirant in the 3' untranslated region of the gene, resulting in the truncation of the Ser RNA, thereby eliminating putative RNA degradation signals located further downstream. This leads to increased stability of Ser RNA and higher levels of Serrate protein. Wing discs of SerD third instar larvae exhibit additional Serrate protein expression in the edge zone of the future wing margin, where it is normally not detectable. Expression of wing margin specific genes, such as cut and wingless, is repressed in these cells (Thomas, 1995).

Ectopic expression of Serrate during wing development induces ectopic outgrowth of ventral wing tissue and formation of an additional wing margin. In order for Serrate to elicit these responses the concomitant expression of wingless seems to be required. Ectopic expression of Delta provokes wing outgrowth and induction of a new margin, both on the dorsal and ventral side. Serrate acts downstream of apterous and induces expression of wing margin patterning genes. Serrate also has the potential to repress margin-specific genes such as wg and cut. This repression results in the failure to differentiate a proper wing margin, visible as a notch in the distal-most part of the wing margin. Thus ectopic Serrate provokes one of two responses, depending on the time and place of expression: outgrowth of wing tissue and induction of a new margin, or repression of margin specific genes that results in a nick in the wing margin. Actions of both Serrate and Delta are mediated by Notch, suggesting that the effects of the two Notch ligands depend on the cellular context, since the capability to activate Notch is spatially and temporally restricted, and expression of ligands at other times and in other places results in repression of Notch activity (Jönsson, 1996).

To ask whether the wing abnormalities caused by reducing Beadex levels might be due to an effect on Apterous activity, the effects of the Bx^hdp excision mutants were examined on Ap target gene expression. In early third-instar fng-lacZ and Serrate (Ser) are expressed evenly throughout the dorsal compartment of the wing disc and are thought to be regulated by Ap. In Bx^hdp mutant discs, the size of the dorsal compartment is considerably reduced, consistent with the small wing phenotype. fng-lacZ expression is not affected in Bx^hdp discs. Ser expression is elevated in the dorsal compartment and does not resolve normally into stripes along the DV boundary and wing veins. Ser expression in the ventral compartment appears normal. The stripes of Ser expression along the DV boundary and wing veins are both dorsal and ventral and are under different regulation than the early dorsal-specific domain. The abnormal pattern of Ser in the dorsal compartment of the Bx^hdp may be due to superimposition of the early and late expression patterns. It is suggested that this reflects a failure to down-regulate Ap activity as the disc matures. To ask whether elevated Ser levels might contribute to the defects observed in Bx^hdp mutant wings, Ser was overexpressed in the dorsal compartment of an otherwise wild-type disc using ap-gal4 to direct UAS-Ser expression. The resulting wings are small and show thickened veins but do not show the abnormalities in vein pattern observed in the Bx^hdp mutant wings. Overexpression of fng using ap-gal4 in a wild-type background produces no phenotype. These observations suggest that Ser overexpression contributes to the abnormalities observed in Bx^hdp mutant wings but that there are likely to be additional factors. Thus, both gain-of-function and loss-of-function Bx mutant phenotypes can be attributed to abnormal regulation of Ap activity. It is concluded that Ap induces dLMO expression in the wing disc and that dLMO then functions as part of a feedback system to regulate the level of Ap activity (Milan, 1998).

Serrate is an essential gene in Drosophila melanogaster, best known for the Ser dominant (SerD) allele and its effects on wing development. Animals heterozygous or homozygous for SerD are viable and exhibit loss of wing margin tissue and associated bristles and hairs. The Beaded of Goldschmidt (BdG) allele of Ser, when heterozygous to wild type, will also produce animals exhibiting loss of wing margin material. However, animals homozygous for BdG exhibit a larval lethal phenotype comparable to animals homozygous for loss-of-function Ser alleles. BdG is a partial duplication of the Ser locus with a single 5' Ser-homologous region and two distinct 3' regions. Meiotic recombination between BdG and a wild-type Ser chromosome demonstrates that only one DNA lesion, caused by the insertion of a transposable element into the coding regions of the Ser transcript, appears capable of generating BdG phenotypes. Due to the insertion, the protein product is predicted to be prematurely truncated and lack an extracellular cysteine-rich region along with the transmembrane and intracellular domains found within the normal SERRATE (SER) protein. The loss of these protein domains apparently contributes to the antimorphic nature of this mutation (Hukriede, 1997a).

Specification of the dorsal-ventral compartment boundary in the developing Drosophila wing disc requires activation of Notch from its dorsal ligand Serrate and its ventral ligand Delta. Both Notch ligands are required in this process: one cannot be substituted for the other. In the wing disc, expression of BD(G), a dominant-negative, truncated form of Serrate, is capable of inhibiting Notch activation in the ventral but not the dorsal compartments. BD(G) can act as a general antagonist of both Serrate and Delta mediated Notch interactions. However, BD(G) retains the Serrate protein domain targeted by Fringe, hence BD(G)'s antagonistic effects are restricted in the dorsal wing disc. Implicit in these results is the suggestion that binding of a ligand to Notch is not sufficient for Notch activation. The specificity of the Notch signal generated by interactions with Serrate and Delta originates from regions residing outside of the Notch binding domains of these molecules; other properties attributable to Notch ligands are required for Notch activation. Thus, ligand binding to Notch is a necessary but insufficient step toward Notch activation (Hukriede, 1997b).

The legs of Drosophila are cylindrical appendages divided into segments along the proximodistal axis by flexible structures called joints, with each leg having 9 segments. The separation between segments is already visible in the imaginal disc because folds of the epithelium and cells at segment boundaries have a different morphology during pupal development. The joints form at precise positions along the proximodistal axis of the leg; both the expression patterns of several genes in the leg and the results of regeneration experiments suggest that different positions along the proximodistal axis have different identities. Two signaling molecules, wingless (wg) and decapentaplegic (dpp) play a central role in patterning the leg discs. These genes are activated in complementary anterior dorsal (dpp) and anterior ventral (wg) sectors in response to the secreted protein Hedgehog, which is only expressed in posterior cells. The asymmetry of dpp and wg expression is maintained by mutual repression: dpp and wg act antagonistically to regulate several genes involved in generating differences along the dorsoventral axis. It is therefore likely that the proximodistal patterning system initiated by wg and dpp determines the localization of presumptive joints in developing leg discs, but the identity of the gene products mediating this process is unknown (de Celis, 1998 and references).

Although the mechanism underlying joint formation is not understood, the fusion of segments caused by some Notch alleles indicates a requirement for Notch signaling. In the leg imaginal disc most segments form concentric rings, with the most distal in the center of the disc. The exceptions are the distal femur and proximal tibia, which are indistinguishable in the larval imaginal disc and only separate during pupariation. This separation occurs through the formation of lateral invaginations that fuse creating two epithelial tubes constricted at the femur/tibia joint. When Notch activity is compromised in N^ts1 larvae during early and late third instar stage, the legs that develop are misshapen, with some fusion between femur/tibia (early) and tarsal (late) segments (de Celis, 1998).

To distinguish which elements of the Notch pathway are required during leg development, clones of homozygous mutant cells were generated, using lethal alleles in fng, Dl and Su(H) as well as a deficiency of the E(spl) complex. Lethal Ser alleles can survive into adults and they have a low frequency of joint fusions. The phenotype of Dl and Su(H) mosaics are similar to each other and, like Notch, result in a failure to make joints when mutant cells are in the position where a joint should have formed. Again, the wild-type cells near the clones can still form joints, but the length of the leg is reduced when the mutant clones are large and span more than one segment. In contrast, mutant cells homozygous for a deficiency that removes the E(spl)bHLH genes form normal joints even when they span more than one segment and are characterized by the differentiation of a vast array of ectopic sensory organs. These develop without intervening epidermal cells, indicating that E(spl) is required for the lateral inhibition mechanism that allows the spacing between sensory organs. The larger clones cause a slight reduction in the overall size of the leg (12% in area and 8% in length), but it is likely that these effects are due to the differentiation of ectopic sensory organs rather than direct effects on growth. Cells mutant for fng also result in fusions between segments. However, these effects are position dependent. Thus, with clones spanning the boundary between the femur and tibia the phenotypes are indistinguishable from those of Notch and Su(H), resulting in a fusion of these two segments and shortening of the leg, whereas in more distal segments defects in the joint can only be detected between the proximal two tarsal segments. The fact that fng is important in leg segmentation suggests that boundaries similar to the wing dorsal-ventral boundary are being created in at least some of the presumptive joints (de Celis, 1998).

In the developing wing the localized activation of Notch can be detected by the activation of certain target genes such as E(spl) and vestigial. Furthermore, the domains of expression of Dl and Ser are important in creating this localized activation of Notch. The expression of Ser, Dl, fng, Notch and E(spl)m beta were therefore examined during leg development. Heterogeneities in the expression of all these genes are detected in the third instar imaginal disc, where Dl and E(spl)m beta RNA are expressed in narrow concentric rings. In evaginating leg discs (0-4 hours APF) and in pupal legs, when the separation between leg segments becomes more evident, E(spl)m beta expression is localized to a ring of distal cells in each leg segment, suggesting that larval expression of E(spl)m beta also defines the distal end of each segment. The expression of fng is also restricted, and is only detected in several broad rings localized to the presumptive tibia and first tarsal segment, and in two groups of distal cells in the fifth tarsal segment that could correspond to the presumptive claws. At this stage, no heterogeneity could be detected in the expression of Notch RNA, but by 24 hours after puparium formation the levels are higher in the places where the joints are being formed, which appear to be the same cells where E(spl)m beta is expressed. At these later stages, Dl also accumulates in rings of cells located at the distal end of each segment and at the separation between the femur and tibia, as well as in many clusters of cells that correspond to developing sensory organs. Expression of E(spl) genes is dependent on Notch activity and hence the localization of E(spl)m beta mRNA to rings of cells in the imaginal and pupal leg disc indicates that there are high levels of Notch activation in the distal-most set of cells in each segment. To determine more precisely the relationship between the E(spl)m beta-expressing cells and the expression of other components of the Notch pathway, a reporter gene was generated in which 1.5 kb of genomic DNA upstream of E(spl)m beta was used to drive expression of a rat cell surface protein, CD2. As a landmark for the segment boundaries an enhancer trap in the bib gene, bib lacZ was used, which is expressed at higher levels in single-cell wide rings at the distal end of each leg segment during both larval and pupal development. The expression of E(spl)m beta-CD2 is localized to a narrow ring, 1-2 cells wide, which coincides with the cells expressing bib lacZ and with cells that have higher levels of lacZ expression in the N lacZ1 enhancer trap line. The expression of N lacZ1 at the dorsoventral boundary and at vein-intervein boundaries is dependent on Notch activity itself. Thus the coincident Notch, E(spl)m beta and bib expression indicates that high levels of Notch activation during imaginal leg development are restricted to the most distal cells of each segment. The accumulation of Notch ligands is also localized within the developing leg segments, with the highest levels of Dl and Ser detected in a narrow stripe of cells localized proximally to those expressing bib lacZ both in the larval imaginal disc and at pupal stages (de Celis, 1998).

Overall the effects produced by ectopic Dl and Ser are similar: the altered morphology of the resulting legs includes both fusion of segments and ectopic joints. However there are positional differences in the way the ligands exert their effects. Thus, the strongest effects of mis-expressing Dl are observed in the tarsal segments, where joint formation is perturbed resulting in foreshortened fused tarsi. This resembles Notch loss-of-function phenotypes suggesting that the levels or position of Dl expression are interfering with normal Notch activity. In addition, an abnormal structure forms at the junction between the first and second tarsal segments, which seems to consist of a partial perpendicular joint. The strongest effects of Ser mis-expression are suggestive of dominant negative effects, since the tibia is foreshortened and forms abnormal joints with the femur and tarsi. In addition, incomplete ectopic joints can be observed at low frequency in distal tarsal segments. Thus, the phenotypes indicate that both activation and repression of Notch occurs when high levels of Notch ligands are expressed. It is likely that the differential effects of misexpression of Dl and Ser are related to the distribution of fng, because the strongest dominant negative effects of Ser occur in the tibia, where fng expression is maximal, and those of Dl occur in distal tarsal segments, where fng is absent or expressed at low levels. Similar effects occur when the ligands are expressed in the wing using the GAL4 system, where the outcome is in part determined by interactions between Notch and Fng (de Celis, 1998).

The functions of artificially constructed secreted forms of the two known Drosophila Notch ligands, Delta and Serrate, were examined by expressing them under various promoters in the Drosophila developing eye and wing. The phenotypes associated with the expression of secreted Delta (DlS) or secreted Serrate (SerS) forms mimic loss-of-function mutations in the Notch pathway. Both genetic interactions between DlS or SerS transgenics and duplications or loss-of-function mutations of Delta or Serrate indicate that DlS and SerS behave as dominant negative mutations. Expression of DlS and SerS in the eye results in a rough eye phenotype. This phenotype is enhanced by loss-of-function Delta and gain-of function Suppressor of hairless. These observations were extended to the molecular level by demonstrating that the expression of Enhancer of split mdelta, a target of Notch signaling, is down-regulated by SerS. The antagonistic nature of the two mutant secreted ligand forms in the eye is consistent with their behavior in the wing, where they are capable of down-regulating wing margin specific genes in an opposite manner to the effects of the endogenous ligands. For example, wingless expression is down-regulated where a SerS expressing stripe crosses the dorsal/ventral boundary. The secreted ligands also interfere with wing vein specification. This analysis uncovers secreted molecular antagonists of Notch signaling and provides evidence of qualitative differences in the actions of the two ligands DlS and SerS (Sun, 1997).

Home page: The Interactive Fly © 1997 Thomas B. Brody, Ph.D.

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