The Drosophila EGF receptor (DER) is required for the specification of diverse cell fates throughout development. We have examined how the activation of DER controls the development of vein and intervein cells in the Drosophila wing. The data presented here indicate that two distinct events are involved in the determination and differentiation of wing cells. (1) The establishment of a positive feedback amplification loop, which drives DER signaling in larval stages. At this time, rhomboid (rho), in combination with vein, initiates and amplifies the activity of DER in vein cells. (2) The late downregulation of DER activity. At this point, the inactivation of MAPK in vein cells is necessary for the maintenance of the expression of decapentaplegic (dpp) and becomes essential for vein differentiation. Together, these temporal and spatial changes in the activity of DER constitute an autoregulatory network that controls the definition of vein and intervein cell types.

The Drosophila EGF receptor (DER) belongs to a large family of transmembrane receptors with intrinsic kinase activity known as receptor tyrosine kinases (RTKs). Following ligand binding, these receptors undergo dimerization, which triggers transphosphorylation of specific tyrosine residues in their cytoplasmic domains. These phosphotyrosines define binding sites for proteins that interact with the receptor through their SH2 domains. One of these, Drk acts as an adapter for the guanine nucleotide releasing factor Son of sevenless (Sos), which facilitates the GDP-to-GTP exchange on Ras. Ras activates the Raf, MEK, MAPK phosphorylation cascade, which ultimately regulates the activities of specific transcription factors (see Perrimon, 1993).

One of the processes in which DER signaling has been extensively studied is cell fate determination in the Drosophila wing. The adult wing is made up of three basic components: the margin structures, the wing veins and the intervein regions. Wing veins are epidermal sclerotizations that arise at stereotypical positions on the wing blade and enclose trachea and nerves (García-Bellido and De Celis, 1992). Animals that carry viable combinations of DER alleles exhibit a partial loss of wing veins (Clifford and Schüpbach, 1989). Similar defects are also observed in somatic clones induced during larval stages of DER mutations and downstream components of RTK signaling (Díaz-Benjumea and García-Bellido, 1990; Díaz-Benjumea and Hafen, 1994). Moreover, ectopic veins develop in wings after overexpression of downstream effectors of DER or from gain-of function alleles (Brunner et al., 1994; Martín-Blanco, 1998). Together, these results suggest that the activation of DER during larval development is responsible for the induction of vein cell fates.

Interestingly, the pattern of expression of DER during wing development is temporally regulated. DER expression, which during the larval period is uniform, is suppressed in vein territories from 8 hours after puparium formation (APF), while it is maintained in intervein cells (Sturtevant et al., 1994). This suggests the possibility that DER, and the Ras/Raf/MAPK cascade, serves a different function(s) during the pupal period.

Other genes involved in DER signaling and vein specification, such as rhomboid, vein or argos, are also dynamically regulated during both larval and pupal periods. Rhomboid (Rho), a seven transmembrane protein, is the earliest marker expressed along veins and rho hypomorphic alleles display a partial loss of wing veins (Sturtevant et al., 1993). Rho has been suggested to be involved in the processing of Spitz (Spi), a stimulatory DER ligand, and to participate in the localized activation of DER (Golembo et al., 1996). It also appears to be involved in DER downregulation from veins during the pupal period (Sturtevant et al., 1994). Vein (Vn) is a Neuregulin-like molecule (Simcox et al., 1996; Schnepp et al., 1996). In the wing disc, Vn is expressed in a stripe delimited by the veins 3 and 4 in larval periods and then it expands to occupy all intervein territories in pupal stages (Simcox et al., 1996). vn mutants have a partial loss of vein phenotype (García-Bellido et al., 1994). Argos (Aos) has been demonstrated to be an inhibitory ligand of DER and appears to act similarly in vein formation (Schweitzer et al., 1995). Aos has been shown to be expressed in vein cells from the third larval instar and aos hypomorphic alleles show weak extra vein phenotypes (Sawamoto et al., 1994).

The differentiation of vein and intervein cells during the pupal period is less well understood. Amongst the genes involved in this process is decapentaplegic (dpp), a TGFβhomologue, which appears to be necessary for vein differentiation. dpp is upregulated in vein cells during pupal period and its expression depends on DER activity (Yu et al., 1996; De Celis, 1997).

In this study, we have found that the level of DER/Ras/Raf/MAPK pathway activity is regulated in time and space during wing development. We show that during the larval period, DER signaling is required for the activation of rho expression in the future vein cells triggering a positive signal amplification loop. This early activation of DER signaling is necessary for the acquisition of ‘vein competence’. We further show that DER signaling (MAPK activity) is downregulated in vein cells and restricted to intervein tissue during pupal development. As a consequence, aos expression is also limited to intervein cells. These changes in the activity of DER signaling control a cell specification switch. DER downregulation in vein cells is necessary for the maintenance of the expression of dpp and the implementation of vein differentiation. Conversely, late DER signaling during pupal stages specifies intervein cells differentiation.

Drosophila strains

The scaGAL-4, Gal-4604 and Gal-4MS1096 lines are insertions of a GAL-4 construct (Brand and Perrimon, 1993). Gal-4604 is expressed in the dorsal and ventral wing blade regions. This expression is maintained in pupal stages. Gal-4MS1096 expression starts on the dorsal wing pouch early in third larval instar and expands later to the ventral surface (Capdevila and Guerrero, 1994). The scaGAL-4 is a hypomorphic sca mutation whose expression initiates in second larval instar and progress to late pupal stages (see Fig. 6). The UAS-Sem and UAS-Vn lines have been already described (Martín-Blanco, 1998; Simcox et al., 1996). The UAS-Aos line was a gift of Mathew Freeman. The hs-DN-DER29-29-1 line was provided by Alan Michelson and encodes a kinase-dead DER.

DNA constructs

A truncation of the D-Raf coding sequence in which amino acids 2-431 are removed, which results in a constitutively activated protein (Stanton et al., 1989), was cloned into the vector pUAST (Brand and Perrimon, 1993) (UAS-ΔD-RafF20). The same D-Raf truncation was also subcloned into pCaSpeR-hs vector to make hs-ΔD-RafF22 (Brand and Perrimon, 1994). UAS-KM-Raf3.1 and UAS-KM-Raf2.1 represent different UAS lines expressing a kinase dead protein. This protein acts as a dominant negative molecule (Sprenger et al., 1992).

Antibody staining and in situ hybridization to imaginal discs

Antibody staining with the anti-active MAPK antibody (Sigma) (1:200 dilution) was performed in third instar larval discs and 24-30 hours old pupal wings fixed during 20-30 minutes with 4% paraformaldehyde in PBS. Fluorescein-conjugated anti-mouse antibodies (Jackson) were used at 1:500 dilution and staining visualized using a confocal microscope. dpp, aos and rho in situ hybridization to imaginal discs was performed as described in Sturtevant et al. (1993).

A positive feedback loop between rho and D-Raf mediates vein cell fate specification

In the third instar wing imaginal disc, the activated form of MAPK, detected with antibodies that recognize phosphorylated Rolled (dpERK) (Gabay et al., 1997), shows a prominent localization along the veins in the wing pouch (Fig. 1A; Gabay et al., 1997). This activated MAPK distribution reflects the activity of DER via the Ras signaling cascade.

Based primarily upon the enhanced loss of veins observed in mutant combinations of DER and rho, it has been suggested that Rho acts upstream of DER in the specification of vein cell fates (Sturtevant et al., 1993). Moreover, rho strongly interacts with vein, a putative DER ligand, as the double-mutant combination rhovevn1 results in a complete loss of veins. To directly analyze the role of these genes in DER activation, we examined the expression of activated MAPK in conditions where we altered the expression of Rho and Vn.

In wings from the rho regulatory mutation rhove, which eliminates all rho expression in the larval wing disc (Sturtevant et al., 1993), dpERK expression is eliminated from the distal parts of veins L3 and L4 and the entire L5 (Fig. 1B). These changes in the distribution of activated MAPK correlate with the absence of vein differentiation in equivalent positions in mutant adults (see Díaz-Benjumea and García-Bellido, 1990). In a more extreme condition, rhovevn1, where all veins are eliminated, the expression of dpERK is abolished from all vein territories in the larval disc (Fig. 1C). In contrast, the overexpression of rho (Sturtevant et al., 1993) leads to a broadening of dpERK expression on vein territories (Fig. 1D). In this condition, the adult wing vein tissue is dramatically enlarged (data not shown). All these data suggest that rho acts upstream of MAPK activation and that vn and rho cooperate in the activation of Ras signaling.

In wild-type wing discs, rho is expressed in the presumptive veins beginning in the mid-third instar larval stage and through pupal stages (Fig. 2A; Sturtevant et al., 1993). To examine the effects of reduced RTK signaling on rho expression in the wing, we ectopically expressed a dominant negative UAS-KM-Raf3.1 transgene from second larval instar, under the control of the GAL-4MS1096 line (Capdevila and Guerrero, 1994), and we found that this prevented the expression of rho mRNA in vein territories during third instar larval stages (compare Fig. 2A and B).

These results indicate that the maintenance of rho expression in the wing disc requires D-Raf activity.

In support of these observations, we also found that DER signaling can activate rho expression. rho mRNA is ectopically expressed in intervein territories after the expression of UAS-Sem (an activated form of the rolled MAPK; Brunner et al., 1994; Martín-Blanco, 1998) and UAS-Vn (Schnepp et al., 1996; Simcox et al., 1996). The highest levels of rho accumulate in dorsal territories in the wing pouch where the expression of the GAL-4MS1096 line is stronger (Fig. 2C,D). Under these conditions, the expression of UAS-Sem or UAS-Vn direct the formation of ectopic veins with the same pattern (data not shown).

MAPK activity is temporally and spatially regulated during pupal wing development

During the early pupal period, the expression of DER rapidly disappears from the vein territories as these territories are established (Sturtevant et al., 1994). This suggests that the downregulation of DER signaling during pupal stages could be a critical event in the differentiation of wing cells.

To determine the activity of the DER signaling cascade at different times during pupal development, we again took advantage of the anti-dpERK antibodies. As described above, activated MAPK was detected along the veins of third instar larval wing discs (Fig. 1A). From 20-24 hours after puparium formation (APF), high levels of activated MAPK expression in veins are accompanied by expression in intervein territories (Fig. 3A). At late stages, activated MAPK expression is eliminated from veins (Fig. 3B), which correlates with the downregulation of DER expression. By confocal analysis, we found that activated MAPK is maintained in the intervein territories, both in the dorsal and the ventral surfaces of the wing blade and in haemocytes that colonize the vein cavities (Fig. 3C). These data show that DER mRNA downregulation is followed by the downregulation of MAPK activity in the veins, while differentiating intervein cells have high levels of MAPK activity.

In pupal stages D-Raf activity represses wing veins, whereas suppression of DER and D-Raf signaling leads to ectopic wing vein differentiation

To test if DER activation has an instructive role during pupal stages in vein and/or intervein differentiation, we crossed flies that carried a

UAS-ΔD-RafF20 transgene, which produces a constitutively activated form of D-Raf, to flies with the GAL-4604 insertion. Expression in this GAL4 line is very strong in pupal stages. Transheterozygous animals had reduced viability and displayed dominant wing phenotypes: ectopic veins and sensilla and a loss of endogenous veins (Fig. 4A). Several other GAL-4 lines active during pupal stages in combination with UAS-ΔD-Raf produce similar defects (data not shown). Examination of MAPK activation during pupal stages in these animals correlates with the suppression of vein fates. MAPK activity in UAS-ΔD-Raf animals was high all over the wing blade during pupal stages (Fig. 4B), ruling out possible antimorphic effects of the activated Raf construct.

We confirmed these results by analyzing the effects of elevation of D-Raf activity at different times. When 30 minute heat shocks were given to animals carrying a constitutively active D-Raf protein under the control of a heat-shock promoter (hs-ΔD-RafF22) during late third instar larval stages, they displayed a significant amount of vein loss (Fig. 4C). Flies of the same genotype, which were heat shocked at early-to-mid second instar stages, developed ectopic vein tissue, in the majority of cases in close proximity to a vein (Fig. 4D). We further verified these effects by using an activated form of Ras1. As described for hs-ΔD-RafF22, heat-shock induction of hs-Ras1V12 late in larval development also results in a loss of veins (data not shown). The observed inhibition of vein differentiation after ectopic expression of activated D-Raf or D-Ras at late developmental periods strongly suggests that the downregulation of DER RTK signaling is crucial for the differentiation of wing vein cells.

If vein specification can be reversed as a consequence of high levels of RTK signaling during late larval and pupal stages, then late downregulation of DER activity should be sufficient to trigger vein specification and differentiation. We analyzed this possibility by using a dominant negative receptor. Heat shocks were given very late in third instar and into pupal stages to animals carrying hs-DN-DER29-29-1. These adults consistently displayed ectopic veins, often in close proximity to normal veins (see Fig. 5A), showing that late suppression of DER activity promotes vein differentiation.

Conversely, hs-DN-DER29-29-1 animals, heat shocked during early-to-mid second instar, displayed substantial vein losses (Fig. 5B).

To verify that the downregulation of RTK signaling during pupal stages is sufficient to induce vein tissue, a dominant negative Raf transgene UAS-KM-Raf2.1 was expressed using the scabrous-GAL-4 line (sca-GAL-4). This line directs expression in third instar wing discs in most of the proneural clusters and along the anteroposterior boundary (Fig. 5C). At 24 hours APF, high levels of expression are detected near the wing margin (Fig. 5D). In pharate adults, GAL-4 activity is maintained in the area around the wing margin, long after expression in the A/P border has disappeared (Fig. 5E). Animals that expressed the dominant negative UAS-KM-Raf2.1 along with sca-GAL-4 lack pieces of L3 and L4 veins (Fig. 5F), which correlates well with the early GAL-4 expression in third instar (Fig. 5C, arrow). Further, they show ectopic veins in positions that associate with the late pupal sca-GAL-4 expression (Fig. 5F). Thus it appears that loss of RTK activity early in third instar results in loss of veins, while loss of RTK activity late in pupal development induces extra vein tissue.

The suppression of vein differentiation by Raf does not depend on a negative feedback loop mediated by argos

Aos is a secreted molecule that functions as an inhibitor of the signaling triggered by DER (Freeman et al., 1992). In the embryonic ventral ectoderm, aos is expressed in the ventralmost row of cells where it is induced by the DER pathway. In this way, the activity of the DER pathway is restricted through an inhibitory loop (Golembo et al., 1996).

In the aossty1 enhancer trap line, X-Gal staining appears to be restricted to the presumptive vein primordia throughout wing development (Sawamoto et al., 1994). Nonetheless, by in situ hybridization, we found that the expression of aos mRNA precisely follows the pattern of MAPK activation (Fig. 3). During pupal development, aos is first expressed in vein territories until 24 hours APF (Fig. 6A). From this time onwards, expression of aos in veins fades away while strong levels accumulate in the intervein tissue (Fig. 6B). To test if aos expression depends on MAPK activity, we overexpressed the UAS-ΔD-RafF20 transgene with the GAL-4604 insertion. In this condition, aos at 30 hours APF is expressed in all vein territories that are later transformed to interveins (Fig. 6C). This strongly suggests that the expression of aos is activated and maintained through the action of the Ras signaling.

We next overexpressed Aos with a UAS-Aos transgene under the control of the GAL-4604 insertion. In this condition, wing veins developed normally and, in some cases, we found some extra vein tissue (Fig. 6D). This is reminiscent of those ectopic veins observed after the overexpression of dominant negative DER or Raf (see Fig. 5A,F). The overexpression of Aos during the larval period using the same transgene resulted in suppression of veins (data not shown). These results suggest that a negative feedback loop mediated by aos is not responsible for the suppression of vein differentiation by Raf signaling.

D-Raf signaling represses decapentaplegic expression during pupal development

The mechanisms involved in the differentiation of wing cells are largely unknown. Detailed genetic analysis has demonstrated that dpp is essential for vein differentiation (Yu et al., 1996; De Celis, 1997). In the absence of dpp signaling during pupal stages, the vein cells do not differentiate, but rather form intervein structures. The activation of dpp expression, which initiates in wing vein territories from 18 hours APF appears to be mediated by the activity of DER (De Celis, 1997). Therefore, we have monitored the expression of dpp subsequent to altering Raf activity during pupal period. We have found that the overexpression of the UAS-ΔD-RafF20 transgene with the GAL-4604 insertion strongly reduces the expression of dpp at 20-24 hours APF (compare Fig. 7B to A). This reduction correlates with the suppression of vein territories shown above (Fig. 4A). Conversely, the downregulation of RTK signaling during pupal stages (using the UAS-KM-Raf2.1/sca-GAL-4 combination) triggers ectopic dpp expression (Fig. 7C). These results suggest that the repression of vein differentiation by Raf activity during pupal stages is likely mediated by the downregulation of dpp signaling.

Vein cell fate specification

A very precise regulation of DER activity must be achieved to control its multiple roles triggering the development of different cell types. This regulation can be controlled at two levels: by transcription and by localized expression of its ligands. In Drosophila, one activating ligand for DER is Spitz, a transmembrane protein, that may be cleaved and act as a diffusible ligand (Schweitzer et al., 1995; Golembo et al., 1996). Based on the results of genetic epistasis, it has been proposed that the processing of Spitz requires the action of Rho. Vn, an additional ligand for DER, is a secreted growth factor. Genetic evidences suggest a functional link between both DER ligands; e.g. a reduction in spitz levels significantly enhances the lack of muscle precursors in vn mutant embryos (Yarnitzky et al., 1998). The levels of activated MAPK induced by the addition of secreted Spitz to DER-transfected S2 cells are significantly higher compared to those induced by Vn (Schnepp et al., 1998). This lower activation of DER by Vn may convey qualitative differences in terms of gene expression. Accordingly, it has been shown that unlike Spitz, Vn overexpression does not induce Aos expression in muscle founders (Yarnitzky et al., 1998). Furthermore, in some systems Vn appears to act after Spitz; e.g. vn mRNA in the somatic mesodermal cells follows Rho expression in the embryonic muscle progenitors (Buff et al., 1998). In addition, secreted Spitz emanating from the midline triggers the expression of vn in the ventral-most cells of the embryonic ectoderm (Golembo et al., 1999).

It has been previously demonstrated that signaling from DER is required for the establishment of ‘vein competency’ and the specification of wing vein cells (Clifford and Schüpbach, 1989; Díaz-Benjumea and García-Bellido, 1990; Sturtevant et al., 1993; Díaz-Benjumea and Hafen, 1994). However, DER is widely expressed throughout the wing and the regulation of the activity of the signaling cascade is thought to be accomplished by localized release of its ligands. While spitz is ubiquitously expressed all over the wing pouch, rho is expressed in narrow vein stripes in all vein primordia and vn is expressed first in a broad central region and then in most intervein cells. A direct role for Spitz in wing development has not yet been revealed and, indeed, clones of null alleles of spitz do not have wing defects (Nagaraj et al., 1999). rho, however, is necessary in conjunction with DER for the specification of vein cells (Sturtevant et al., 1993) and directs the local activation of the DER signaling pathway (Fig. 1). Nonetheless, the complete suppression of rho expression in wing discs (in a rhove mutant) does not result in a full repression of MAPK activity from presumptive vein territories. This suggests that the activation of MAPK depends on an additional input. Indeed, MAPK activation in wing veins also requires the presence of Vn (Fig. 1). In all, these results lead us to suggest a model in which Vn is required to supplement Rho activities in the specification of vein cells. Vn could act synergistically or sequentially to Rho to provide a low, but continuous level of DER activation. Beyond reinforcing the DER pathway activity in neighbouring vein cells, Vn expression could provide an autocrine function in interveins where it will direct low level activation of DER and be necessary for intervein cell specification.

An early positive feedback loop

Signals passing between adjacent sectors in the wing primordia activate the expression of ‘vein-organizing genes’ (as rho) in sharp vein stripes. These genes would be involved in the activation of secondary signals and the development of provein territories (Biehs et al., 1998). The crossregulation between ‘vein-organizing genes’ and late effectors is thought to establish and refine the vein pattern (e.g. Roch et al., 1998).

rho expression and the specification of vein cells can be compromised by different mutations affecting distinct vein-promoting activities (Sturtevant and Bier, 1995). Among the genes involved in the activation of rho expression is vn. vn allelic combinations reduce the expression of rho in the wing. Moreover, insufficient levels of D-Raf expression strongly correlate with a general failure to express rho (Fig. 2). These results indicate that rho expression is downstream of RTK signaling, and is therefore likely to be downstream of DER signaling activity. These findings are consistent with the regulation of rho expression in the ovary, where the activation of DER by Gürken, an ovary-specific DER ligand, leads to rho expression activation and the determination of the dorsal follicle cells (Sapir et al., 1998).

rho expression is ectopically induced following general DER signaling activation. Interestingly, rho is upregulated only in areas in proximity to the veins (Fig. 2C,D). This suggests that there may be a convergent signaling pathway providing spatial and temporal information. An elevation of vein ‘competence’ via increased DER signaling would permit the development of extra vein cells upon input of a localized second signaling pathway. In accordance with this model, one would predict that the ectopic vein tissue produced by artificially elevated RTK signaling during the second instar would be found in close proximity to normal veins, as it is observed, rather than randomly distributed in the wing blade. We suggest that the function of a positive regulatory loop between DER and rho will be the reinforcement of the activation of the pathway that will eventually result in the establishment of vein competent territories.

A biphasic response to DER/D-Raf signaling

It has been recently demonstrated the reiterated use of DER as a common effector of differentiation. In the Drosophila eye, DER is required for the determination of all cell types. In this system, cell fate depends on the developmental stage when the receptor is activated (Freeman, 1996).

By interfering with DER signaling activity, it can be found that the specification of veins respond to the activation of RTK signaling during larval stages, but that continued activation of RTK signaling results in a failure of vein cells to differentiate (see Fig. 4). One explanation for these opposite effects could be that early activation of RTK signaling would specify vein cells, while late RTK signaling would implement intervein cell fates. Several observations provide support for this model.

In pupae, MAPK is repressed in veins and activated in intervein cells. This activation of MAPK (and the expression of downstream genes, such as argos) responds to Ras signaling activity (Figs 4B, 6C), and appears to be involved in the suppression of vein cell fates. Indeed, after ectopical activation of D-Raf during the pupal period, promoting intervein cell fates, the MAPK activity remains stimulated all over the wing blade (Fig. 4B).

It seems that DER is the only receptor tyrosine kinase at work in the wing, able to activate Ras and Raf. While DER is ubiquitously expressed during larval imaginal disc development, DER mRNA levels are downregulated in the pupal period in presumptive vein cells (Sturtevant et al., 1994). This downregulation of DER could be involved in the supression of MAPK activity in vein territories (Fig. 3). Furthermore, when a DN-DER molecule is overexpressed, titrating the endogenous DER, in pupae, extra vein tissue is induced (Fig. 5A). MAPK dephosphorylation in veins could also be induced by other mechanisms; for instance, the early expression of the inhibitor ligand Argos in veins up to 24 hours APF (Fig. 6A) could cooperate in the inactivation of MAPK in these territories.

What is the function of this change of expression? The first effect of this developmental switch is a modification in the expression of downstream targets. As a consequence of the reduction in MAPK activity from vein cells, aos is eliminated from veins between 24 and 30 hours APF. Conversely, it is upregulated in intervein territories (see Fig. 6). This scenario is reminiscent of the induction of DER ligands in the ventral ectoderm. Here, the primary signal, Spitz induces a relay mechanism by triggering the expression of Vn (and Aos) in adjacent cells. Aos reduces the overall level of DER signaling, whereas Vn provides a lower level of activation, capable of inducing only the lateral cell fates (Golembo et al., 1999). In the larval wing, high levels of DER signaling are achieved in veins through a positive feedback loop (see above). Here, DER activity promotes the expression of Aos. We suggest that Aos diffusion from veins could prevent adjacent cells from responding to the vein inductive signals and producing high levels of DER activity (‘remote inhibition’ – Freeman, 1996). Consistently, aos mutant flies display small deltas and extra veins clustered around vein territories. On the contrary, Aos overexpression in larval stages induces the suppression of veins (Sawamoto et al., 1994). We also propose that, in pupae, while DER activity (and Aos) in veins are lost, Vn and Aos expression in intervein cells will reach a competitive balance leading to the activation of DER and MAPK, and intervein cell specification (see above and Fig. 8).

In an alternative model, different threshold requirements for DER signaling might be necessary at different stages during the development of the wing. DER activity will promote vein specification and induce a negative feedback loop at all times (i.e. by inducing the overexpression of Aos). In this scenario, ectopic veins resulting from the downregulation of the pathway (see Fig. 5) would be a consequence a reduction of DER activity by just the right amount to eliminate negative feedback loops without disrupting positive effector functions. These positive functions would eventually be implemented by a signaling event dependent on DER, but independent of Raf/MAPK activity. Is worth noting that this model is not supported by the consequences of Aos overexpression in pupal stages. At these stages, Aos ectopic expression does not result in the suppression of vein fates (Fig. 6D).

Cross talk between dpp and DER signaling during pupal development

Several types of cell-cell communication have been proposed to be required during the latter stages of pupal wing development. The dpp gene encodes a member of the TGFβ superfamily and is expressed during early pupal development in vein primordia (Yu et al., 1996). A class of loss-of-function dpp alleles and certain combinations of Dpp receptor mutants lead to vein-loss phenotypes (Burke and Basler, 1996). Mosaic analysis of dpp s allele show that mitotic clones affect the differentiation of veins. Meanwhile, the effects of overexpression of dpp or an active form of its receptor thick veins (tkv) indicate that Dpp directs vein differentiation through activation of Tkv in pupal stages (De Celis, 1997).

The initiation of dpp expression in pupal stages depends on the activity of early acting genes, and in particular DER activity (Yu et al., 1996; De Celis, 1997). However, although DER signaling is downregulated in vein territories during pupariation, dpp expression is maintained through an autoregulatory loop and remains high in vein cells until their final differentiation. Interestingly, in intervein cells, dpp expression is not activated in response to the DER activity described above. On the contrary, these cells express short gastrulation (sog), a gene that exerts an opposing effect to dpp (François et al., 1994). sog plays a role restricting vein formation to the center of the provein regions. dpp and sog interact antagonistically during vein differentiation (Yu et al., 1996). Ectopic activation of DER signaling in pupal stages abolishes dpp expression from veins (Fig. 7). This suppression of dpp correlates with the loss of veins observed in this condition and it is reminiscent of the effect of Sog overexpression in pupal wings (Yu et al., 1996). Moreover, vein plexates induced by compromising DER activity in pupal wings, associate with a broadening of dpp-expressing areas. We suggest that DER signaling downregulation from vein territories allows dpp to autoregulate its expression (Fig. 8). It remains to be determined whether sog expression depends on DER in intervein territories, or is a consequence of the activity of intervein-specific genes such as blistered (Roch et al., 1998). The model presented here on how a single receptor (DER), triggering a conserved signal transduction pathway, is used reiteratively to implement two different cell fates in the development of the fly wing serves to reconcile many observations that have been made regarding cell fate specification in the wing. This may well provide a paradigm for the regulation of DER signal transduction in other developmental events.

We thank A. Martínez-Arias for his support during early stages of this work. M. Freeman, I. Guerrero, T. Kline, A. Simcox and J. Urban for generously providing GAL-4 and UAS lines, A. Michelson for the hs-DN-DER flies. We thank M. Freeman, B. Shilo, E. Bier, S. Blair, A. Brand, and members of the Perrimon and Martínez-Arias laboratories for helpful discussions, and J. Modolell for comments on the manuscript. E. M.-B. was a Marie Curie fellow supported by the European Union. MEC and CAM Studentships supported A. B. and F. R. respectively. J. B. D. is supported by the Damon Runyon-Walter Winchell Cancer Research Fund. N. P is an Investigator of the Howard Hughes Medical Institute.

Biehs
,
B.
,
Sturtevant
,
M. A.
and
Bier
,
E.
(
1998
).
Boundaries in the Drosophila wing imaginal disc organize vein-specific genetic programs
.
Development
125
,
4245
4257
.
Brand
,
A. H.
and
Perrimon
,
N.
(
1993
).
Targeted gene expression as a means of altering cell fates and generating dominant phenotypes
.
Development
118
,
401
415
.
Brand
,
A. H.
and
Perrimon
,
N.
(
1994
).
Raf acts downstream of the EGF receptor to determine dorsoventral polarity during Drosophila oogenesis
.
Genes Dev
.
8
,
629
639
.
Brunner
,
D.
,
Oellers
,
N.
,
Szabad
,
J.
,
Biggs
,
W. H.
,
Zipursky
,
S. L.
and
Hafen
,
E.
(
1994
).
A gain-of-function mutation in Drosophila MAP kinase activates multiple receptor tyrosine kinase signaling pathways
.
Cell
76
,
875
888
.
Buff
,
E.
,
Carmena
,
A.
,
Gisselbrecht
,
S.
,
Jimenez
,
F.
and
Michelson
,
A. M.
(
1998
).
Signalling by the Drosophila epidermal growth factor receptor is required for the specification and diversification of embryonic muscle progenitors
.
Development
125
,
2075
2086
.
Burke
,
R.
and
Basler
,
K.
(
1996
).
Dpp receptors are autonomously required for cell proliferation in the entire developing Drosophila wing
.
Development
122
,
2261
2269
.
Capdevila
,
J.
and
Guerrero
,
I.
(
1994
).
Targeted expression of the signaling molecule Decapentaplegic induces pattern duplications and growth alterations in Drosophila wings
.
EMBO J
.
13
,
4459
4468
.
Clifford
,
R. C.
and
Schüpbach
,
T.
(
1989
).
Coordinately and differentially mutable activities of torpedo, the Drosophila melanogaster homologue of the vertebrate EGF receptor gene
.
Genetics
122
,
771
787
.
De Celis
,
J. F.
(
1997
).
Expression and function of decapentaplegic and thick veins during the differentiation of the veins in the Drosophila wing
.
Development
124
,
1007
1018
.
Díaz-Benjumea
,
F. J.
and
García-Bellido
,
A.
(
1990
).
Behaviour of cells mutant for an EGF receptor homologue of Drosophila in genetic mosaics
.
Proc. Natl. Acad. Sci. USA
242
,
36
44
.
Díaz-Benjumea
,
F. J.
and
Hafen
,
E.
(
1994
).
The sevenless signaling cassette mediates Drosophila EGF receptor function during epidermal development
.
Development
120
,
569
578
.
François
,
V.
,
Solloway
,
M.
,
O’Neill
,
J. W.
,
Emery
,
J.
and
Bier
,
E.
(
1994
).
Dorsal-ventral patterning of the Drosophila embryo depends on a putative negative growth factor encoded by the short gastrulation gene
.
Genes Dev
.
8
,
2602
2616
.
Freeman
,
M.
,
Klambt
,
C.
,
Goodman
C. S.
and
Rubin
,
G.M.
(
1992
).
The argos gene encodes a diffusible factor that regulates cell fate decisions in the Drosophila eye
.
Cell
69
,
963
975
.
Freeman
,
M.
(
1996
).
Reiterative use of the EGF receptor triggers differentiation of all cell types in the Drosophila eye
.
Cell
87
,
651
660
.
Gabay
,
L.
,
Seger
,
R.
and
Shilo
,
B.-Z.
(
1997
).
In situ activation pattern of Drosophila EGF receptor pathway during development
.
Science
277
,
1103
1106
.
García-Bellido
,
A.
and
De Celis
,
J. F.
(
1992
).
Developmental genetics of the venation pattern of Drosophila
.
Ann. Rev. Genetics
26
,
277
304
.
García-Bellido
,
A.
,
Cortés
,
F.
and
Milán
,
M.
(
1994
).
Cell interactions in the control of size in Drosophila wings
.
Proc. Natl. Acad. Sci. USA
91
,
10222
10226
.
Golembo
,
M.
,
Schweitzer
,
R.
,
Freeman
,
M.
and
Shilo
,
B.-Z.
(
1996
).
argos transcription is induced by the Drosophila EGF receptor pathway to form an inhibitory feedback loop
.
Development
122
,
223
230
.
Golembo
,
M.
,
Yarnitzky
,
T.
,
Volk
,
T.
and
Shilo
,
B.-Z.
(
1999
).
vein expression is induced by the EGF receptor pathway to provide a positive feedback loop in patterning the Drosophila embryonic ventral ectoderm
.
Genes Dev
.
13
,
158
162
.
Martín-Blanco
,
E.
(
1998
).
Regulatory control of signal transduction during morphogenesis in Drosophila
.
Int. J. Dev. Biol
.
42
,
363
368
.
Nagaraj
,
R.
,
Pickup
,
A. T.
,
Howes
,
R.
,
Moses
,
K.
,
Freeman
,
M.
and
Banerjee
,
U.
(
1999
).
Role of the EGF receptor pathway in growth and patterning of the Drosophila wing through the regulation of vestigial
.
Development
126
,
975
985
.
Perrimon
,
N.
(
1993
).
The torso receptor protein-tyrosine kinase signaling pathway: an endless story
.
Cell
74
,
219
222
.
Roch
,
F.
,
Baonza
,
A.
,
Martín-Blanco
,
E.
and
García-Bellido
,
A.
(
1998
).
Genetic interactions and cell behaviour in blistered mutants during proliferation and differentiation of the Drosophila wing
.
Development
125
,
1823
1832
.
Sapir
,
A.
,
Schweitzer
,
R.
and
Shilo
,
B.-Z.
(
1998
).
Sequential activation of the EGF receptor pathway during Drosophila oogenesis establishes the dorsoventral axis
.
Development
125
,
191
200
.
Sawamoto
,
K.
,
Okano
,
H.
,
Kobayakawa
,
Y.
,
Hayashi
,
S.
,
Mikoshiba
,
K.
and
Tanimura
,
Y.
(
1994
).
The function of argos in regulating cell fate decisions during Drosophila eye and wing vein development
.
Dev. Biol
.
164
,
267
276
.
Schnepp
,
B.
,
Grumbling
,
G.
,
Donaldson
,
T.
and
Simcox
,
A.
(
1996
).
Vein is a novel component in the Drosophila epidermal growth-factor receptor pathway with similarity to the neuregulins
.
Genes Dev
.
10
,
2302
2313
.
Schnepp
,
B.
,
Donaldson
,
T.
,
Grumbling
,
G.
,
Ostrowski
,
S.
,
Schweitzer
,
R.
,
Shilo
,
B.-Z.
and
Simcox
,
A.
(
1998
).
EGF domain swap converts a Drosophila EGF receptor activator into an inhibitor
.
Genes Dev
.
12
,
908
913
.
Schweitzer
,
R.
,
Howes
,
R.
,
Smith
,
R.
,
Shilo
B.-Z.
and
Freeman
,
M.
(
1995
).
Inhibition of Drosophila EGF receptor activation by the secreted protein Argos
.
Nature
376
,
699
702
.
Simcox
,
A. A.
,
Grumbling
,
G.
,
Schnepp
,
B.
,
Bennington-Mathias
,
C.
,
Hersperger
,
E.
and
Shearn
,
A.
(
1996
).
Molecular, phenotypic, and expression analysis of vein, a gene required for growth of the Drosophila wing disc
.
Dev. Biol
.
177
,
475
489
.
Sprenger
,
F.
,
Trosclair
,
M. M.
and
Morrison
,
D. K.
(
1992
).
Biochemical analysis of Torso and D-Raf during Drosophila embryogenesis – implications for terminal signal transduction
.
Mol. Cell. Biol
.
13
,
1163
1172
.
Stanton
,
V. J.
,
Nichols
,
D. W.
,
Laudano
,
A. P.
and
Cooper
,
G. M.
(
1989
).
Definition of the human Raf amino-terminal regulatory region by deletion mutagenesis
.
Mol. Cell. Biol
.
9
,
639
647
.
Sturtevant
,
M. A.
,
Roark
,
M.
and
Bier
,
E.
(
1993
).
The Drosophila rhomboid gene mediates the localized formation of wing veins and interacts genetically with components of the EGF-R signaling pathway
.
Genes Dev
.
7
,
961
973
.
Sturtevant
,
M. A.
,
O’Neill
,
J. W.
and
Bier
,
E.
(
1994
).
Down-regulation of Drosophila EGF-R mRNA levels following hyperactivated receptor signaling
.
Development
120
,
2593
2600
.
Sturtevant
,
M. A.
and
Bier
,
E.
(
1995
).
Analysis of the genetic hierarchy guiding wing vein development in Drosophila
.
Development
121
,
785
801
.
Yarnitzky
,
T.
,
Min
,
L.
and
Volk
,
T.
(
1998
).
An interplay between two EGF-receptor ligands, Vein and Spitz, is required for the formation of a subset of muscle precursors in Drosophila
.
Mech. Dev
.
79
,
73
82
.
Yu
,
K.
,
Sturtevant
,
M. A.
,
Biehs
,
B.
,
François
,
V.
,
Padgett
,
R. W.
,
Blackman
R. K.
and
Bier
,
E.
(
1996
).
The Drosophila decapentaplegic and short gastrulation genes function antagonistically during adult wing vein development
.
Development
122
,
4033
4044
.