A Function for Egf Receptor Signaling in Expanding the Developing Brain in Drosophila

Current Biology, Vol. 13, 474–482, March 18, 2003, 2003 Elsevier Science Ltd. All rights reserved. DOI 10.1016/S 09 60 - 98 22 ( 03 )0 0 09 4- 0

A Function for Egf Receptor Signaling in Expanding the Developing Brain in Drosophila Damon T. Page* MRC Laboratory of Molecular Biology Hills Road Cambridge CB2 2QH United Kingdom

Summary Background: In invertebrates and vertebrates, neural midline cells secrete signals that pattern the central nervous system (CNS). However, an important part of the developing insect brain, involved in functions such as olfaction and feeding behavior, is positioned lateral to the foregut and lacks neural cells at the midline. Could the foregut substitute for neural midline cells and secrete signals that pattern this part of the brain? Results: In Drosophila embryos, the neural midline marker Single-minded is expressed in foregut cells adjacent to the brain, as are members of the Egf receptor signaling pathway. Removing the function of these molecules results in aberrant proliferation and reduced size in the brain lateral to the foregut. Conclusions: Cells of the brain lateral to the foregut receive an Egf signal from the midline and proliferate in response. A likely source of this signal is the foregut. These findings raise the possibility that the brain lateral to the foregut is an evolutionarily recent addition to the arthropod brain, and that the anterior boundary of the brain neural midline is a conserved feature in bilaterally symmetric animals. Introduction In both vertebrates and invertebrates, neural midline cells influence the development of CNS tissue. In vertebrates, signals originating from neural cells located at the midline of the CNS (the floor plate) pattern adjacent neural tissue (see [1] for review). Similarly, the insect Drosophila has neural midline cells in the ventral nerve cord that express a combination of molecules that distinguishes them from other CNS cells [2]. When these neural midline cells are ablated, there is a decrease in the number of neural cells located in the ventral nerve cord [3]. Molecules that are expressed in Drosophila neural midline cells include the bHLH-PAS transcription factor Single-minded (Sim), which regulates neural midline fate [4], Rhomboid, which processes the Egf receptor (Egfr) ligand Spitz [5, 6], and Vein, which is a secreted ligand for Egfr [7]. All of these molecules are involved in Drosophila ventral nerve cord development [8, 9]. Some Egfr pathway molecules also act in the midline in the dorsal-most region of the embryonic brain [10]. However, in the region of the deutocerebral and tritocerebral neuromeres, a proper neural midline is missing, and instead the foregut is located at the midline, with *Correspondence: [email protected]

the brain forming around this tissue (I will refer to this part of the brain as the midbrain) (see Figure 1A). Does the foregut act as a replacement for the neural midline in this region? Here, I show that Sim is strongly expressed in the foregut, where this tissue intersects with the brain, but not in the brain lateral to the foregut. rhomboid and vein are also expressed in the foregut. I examine the function of these genes and find that they are required for the formation of the brain lateral to the foregut cells in which they are expressed.

Results and Discussion Expression of Neural-Midline Markers in the Brain The embryonic Drosophila brain consists of three neuromeres in the supraesophageal ganglion—the protocerebral, deutocerebral, and tritocerebral neuromeres— and three neuromeres in the subesophageal ganglion. The protocerebral neuromere forms dorsal to the foregut, while the deutocerebral and tritocerebral neuromeres are found lateral to the foregut at the end of embryogenesis. Examining the placement of the foregut relative to these neuromeres shows that the foregut is in the position at which neural midline cells would be elsewhere in the CNS (Figure 1A). Does the neural midline bifurcate around the foregut, or does the neural midline stop ventral to the foregut? Do foregut cells express neural midline markers? I examined the expression pattern of sim and saw that this gene is expressed in the foregut (see [11]); it is strongly expressed in the ventral foregut, where this tissue intersects with the brain. This expression is present as the foregut begins to invaginate (Figure 1B) and continues at a high level throughout embryogenesis (Figures 1C– 1E). I did not see expression of sim (as assayed by using reporter genes and antibody) in the midbrain at any developmental stage. Also, at no stage did I observe sim expression in the glia that lie between the foregut and the circumesophageal connectives; instead, these glia express Repo (Figure 1A), which is a marker for lateral glia in the CNS but is not expressed in neural midline cells [12]. This shows that cells expressing the neural midline marker sim do not bifurcate around the foregut, but, rather, it is the foregut cells that express sim. Next, I examined rhomboid expression and found strong expression in the same ventral foregut cells that express sim (Figures 1F, 1G, and 1I). rhomboid is also expressed in the midbrain in a dynamic pattern (Figures 1H and 1J). Coexpression of rhomboid and sim in the foregut was confirmed by using the anti-Sim antibody in embryos carrying the rhomboid reporter gene (data not shown). vein encodes a ligand for Egfr and is also expressed in the foregut adjacent to the brain (Figures 1K and 1L). Also, vein is expressed in a dynamic pattern in the midbrain (Figures 1M and 1N). This shows that foregut cells adjacent to the brain express markers similar to those expressed in neural midline cells, even

Midline Signaling in the Drosophila Brain 475

Figure 1. The Expression of sim, rhomboid, and vein in the Brain Midline (A) Arrangement of the foregut and brain in wild-type Drosophila embryo. Solid lines mark the epithelium of the stomodeum or foregut (fg). Dashed lines mark the ventral midline. The fluorescent image shows the position of Repo-expressing lateral glia (green), and CNS axons are in red (marked with BP102). The schema shows that the ventral neural midline is interrupted by the foregut. fg, foregut; b1, protocerebral neuromere; b2, deutocerebral neuromere; b3, tritocerebral neuromere; sbg, subesophageal ganglion. (B) ES 10 embryo showing expression of sim in the ventral foregut, as it invaginates past cells of the subesophageal ganglion (white dashed lines). (C) ES 14 embryo showing continued sim expression in the foregut (solid white lines) at the foregut-brain intersection. Note that the position has shifted posterior, corresponding with the morphogenetic movement of the brain. (D) ES 15 embryo, viewed frontally to show sim expression in the foregut. (E) ES 15 embryo, viewed dorsally, again showing sim expression in the foregut (arrow). This expression pattern was confirmed by using the anti-Sim antibody; in these embryos, additional staining was seen in the dorsal anterior foregut starting around ES 13 and in a small number of cells adjacent to the preoral brain commissure (data not shown). (F) ES 10 embryo showing rhomboid-lacZ expression in the ventral foregut (arrow). (G) ES 14 embryo showing rhomboid-lacZ expression in the foregut at the foregut-brain intersection (arrow). (H) ES 10 embryo showing rhomboid-lacZ expression in midbrain cells (arrow); expression in the anterior foregut can also be seen here (arrowhead). (I and J) ES 10 wild-type embryos showing the pattern of rhomboid in situ hybridization. (I) Embryo with staining in the ventral foregut (arrow). Note staining in the dorsal foregut that is not seen in rhomboid-lacZ embryos. (J) Staining in midbrain cells (arrow) and in the anterior foregut (arrowhead). (K–N) Wild-type embryos showing the pattern of vein in situ hybridization. (K) ES 10 embryo with staining in the ventral and dorsal foregut (arrow). (L) ES 13 embryo with vein expression in the foregut (arrow). (M) ES 10 embryo showing vein expression in midbrain cells (arrow) and in the anterior foregut (arrowhead). (N) ES 13 embryo with weakening vein expression in midbrain cells.

Current Biology 476

Table 1. Penetrance and Expressivity of Phenotypes Severity of Defects (% Penetrance) Genotype

⫹

⫹⫹

⫹⫹⫹

Average Cell Counts in Whole Population (SD)

Average Cell Counts in ⫹⫹⫹ Embryos (SD)

Wild-type rhomboid7m43 vein⌬P10567 single-minded2 Egfrf1

— 0 5 0 25

— 20 25 30 20

— 80 70 70 55

1807 (205) 1269 (245) 1519 (280) 512 (290) 703 (557)

— 1119 (136) 1429 (178) 212 (67) 175 (75)

The value given for cell counts is the average area in ␮m2 occupied by Elav-expressing nuclei. For Egfrf1 flies, data were collected from embryos that were temperature shifted at ES 9, as described in the text. Statistically significant deviations in the midbrain area from wildtype were seen in all groups of mutant embryos. n ⫽ 20 in all cases. Severity of defects was rated as follows: ⫹⫹⫹ embryos showed the complete phenotype described for the mutation in the text of this paper, ⫹⫹ embryos showed some of this phenotype, and ⫹ embryos did not show this phenotype.

Figure 2. Disrupting sim or Egfr Signaling Results in Brain Patterning Defects All embryos are viewed laterally. (A–E) Embryos are at ES 15/16; neuronal nuclei are in blue (anti-Elav), glia nuclei are in green (anti-Repo), and axons are in red (anti-Fasciclin II). The approximate limits of the midbrain are shown with white dashed lines. (A) Wild-type embryonic brain. (B and C) Embryos mutant for rhomboid, showing (B) diminished midbrain and the (C) circumesophageal connectives that are frequently broken at ES 16 (arrow). (D) Embryo lacking Vein function, showing a slight loss of midbrain. (E) Sim⫺ embryo showing loss of brain lateral to the foregut. (F–I) ES 11 embryos showing the pattern of (F and H) rhomboid or (G and I) vein in situ hybridization in sim mutant embryos. (F) Loss of rhomboid expression in the ventral foregut (arrow). (G) vein expression is also lost in the ventral foregut (arrow). (H and I) In the midbrain, rhomboid and vein expression is decreased (arrow).

Midline Signaling in the Drosophila Brain 477

Figure 3. Temporal and Spatial Characteristics of Egfr Signaling, and the Effects of Sim on Proliferation and Apoptosis in the Midbrain All embryos are viewed laterally. (A–C) All embryos are at ES 15/16; neuronal nuclei are in blue (anti-Elav), glia nuclei are in green (anti-Repo), and axons are in red (antiFasciclin II). The approximate limits of the midbrain are shown with white dashed lines. (A) Embryo in which Egfr function was removed at ES 9, showing loss of midbrain. (B) Embryo in which Egfr function was removed at ES 10, showing loss of midbrain. (C) Embryo in which Egfr function was removed at ES 12, not showing loss of midbrain, but showing a disruption of circumesophageal connective formation (arrow). (D–G) All embryos are stained with anti-DpERK antibody. (D and E) Wild-type embryos. (D) ES 8 embryo with staining in the head ectoderm. (E) ES 11 embryo showing light staining in the midbrain. (F and G) sim⫺ embryos. (F) ES 8 sim⫺ embryos showing staining in head ectoderm. (G) ES 11 sim⫺ embryo lacking staining in the midbrain. (H) ES 15 embryo of the UAS-Egfr*, UAS-lacZ/1407-GAL4; sim2/sim2 genotype. The loss of the midbrain in a sim⫺ background is rescued by expressing activated Egfr in neural cells. Activation of Egfr in the midbrain of these embryos was verified by staining with anti-DpERK antibody, which showed increased staining throughout the CNS (data not shown). (I–K) All embryos are at ES 12 and are stained with the mitosis marker anti-Phosphohistone-H3. The approximate limits of the midbrain are marked with dashed black lines. (I) Wild-type embryo. (J) In sim⫺ embryos, proliferation in the midbrain is decreased. (K) Overexpressing Rhomboid using sim-GAL4 leads to increased proliferation in the midbrain. (L–O) Embryos marked with TUNEL staining. (L) and (N) are wild-type embryos, and (M) and (O) are sim⫺ embryos. (M) In ES 12 Sim⫺ embryos, apoptosis is not increased in the midbrain (compare with [L]). (O) ES 15 sim⫺ embryo showing an increase in apoptotic cells in the midbrain (arrow) (compare with [N]).

though they originate from a different primordia. The foregut is ectodermal in origin, while the neural midline arises from the mesectoderm [13].

function of rhomboid results in a loss of midbrain (Table 1, Figures 2A–2C). I saw a similar, but less severe, result in vein⫺ embryos (I will refer to loss of function mutants here as “⫺”) (Table 1, Figure 2D). In the ventral nerve cord, Sim influences the expression of Egfr signaling molecules [2], and loss of Sim function results in a disruption of connective development in the brain [14]; thus, sim could be involved in midbrain ganglia develop-

The Function of Neural Midline Genes in Midbrain Development Next, I wanted to know if these neural midline markers function in the patterning of the midbrain. Removing the Table 2. Proliferating Cell Counts

Severity of Defects (% Penetrance) Genotype

⫹

⫹⫹

⫹⫹⫹

Average Cell Counts in Whole Population (SD)

Average Cell Counts in ⫹⫹⫹ Embryos (SD)

Wild-type single-minded2 single-minded-GAL4, UAS-rhomboid

— 15 45

— 15 15

— 70 40

4.8 (0.8) 2.2 (1.6) 6.0 (1.6)

— 1.9 (0.9) 6.3 (1.5)

Assayed with anti-PH3 antibody. Statistically significant deviations in mitotic cell counts from wild-type were seen in both groups of embryos. n ⫽ 20 in all cases.

Current Biology 478

Table 3. Empty Spiracles-Expressing Cell Counts Severity of Defects (% Penetrance)

Wild-type single-minded2 Egfrf1 forkhead6

⫹

⫹⫹

⫹⫹⫹

Average Cell Counts in Whole Population (SD)

Average Cell Counts in ⫹⫹⫹ Embryos (SD)

— 10 10 15

— 25 30 20

— 65 60 65

3.8 1.8 2.2 2.1

— 1.6 (1.0) 2.0 (1.3) 1.8 (1.1)

(0.8) (1.4) (1.6) (1.3)

Assayed with anti-Empty spiracles antibody. Statistically significant deviations in counts of Empty spiracles-expressing midbrain cells from wild-type were seen in all groups of embryos. n ⫽ 20 in all cases.

ment. In sim⫺ embryos, the midbrain is mostly deleted (Table 1, Figure 2E). This phenotype includes, but is stronger than, the rhomboid⫺ phenotype or the vein⫺ phenotype, suggesting that these Egfr signaling molecules act cooperatively downstream of Sim. To test this, I examined the expression of rhomboid and vein in a sim⫺ background, and I found that the expression of these two genes is missing in the ventral foregut at the intersection with the brain (Figures 2F and 2G) and that expression in the midbrain is diminished (Figures 2H and 2I). The midbrain phenotype of rhomboid mutants is stronger than that of vein mutants, suggesting that Rhomboid, and thus Spitz, are more important effectors of Egfr activation in this region. Vein may act to boost the levels of Egfr activity in a manner similar to what happens in the embryonic ventral ectoderm [15]. Temporal and Spatial Characteristics of Egfr Activation in Brain Development A prediction would be that the midbrain phenotype for Egfr⫺ embryos should resemble that of sim⫺ embryos. I used a temperature-sensitive mutation in Egfr to circumvent the widespread patterning defects associated with early loss of Egfr and found that, when Egfr function is removed at ES 9, a midbrain phenotype results at ES 15 that is as strong as that of sim⫺ (Table 1, Figure 3A). A shift at late ES 10 also results in a loss of midbrain ganglia (Figure 3B). Importantly, removing Egfr function at late ES 12 does not result in a gross loss of midbrain; however, in these embryos, the circumesophageal connectives are sometimes disrupted (Figure 3C). To visualize where and when Egfr is activated in the brain, I used an anti-DpERK antibody, which marks cells in which Egfr has been activated by recognizing MAPK phosphorylation [16]. Egfr is strongly activated in the surface head ectoderm around ES 8 (Figure 3D), and activation is weak in the midbrain at early ES 10. There is a brief period of stronger activation in the midbrain around ES 11 (Figure 3E), which fades through ES 12. Thus, based on the temperature shift series and the DpERK assay, it appears that a critical period for Egfr activation in the developing midbrain is around ES 11. Sim Is Required for Egfr Activation in the Brain, and Egfr Activation in the Brain Can Rescue the sim Mutant Phenotype To see if Sim acts upstream of Egfr activation, I used two approaches. First, I examined the pattern of DpERK in a sim⫺ background. At ES 8, I saw DpERK in cells on the surface of the head ectoderm (Figure 3F). However,

I did not see DpERK expression in the midbrain at ES 11 (Figure 3G). Thus, it appears that the early activation of Egfr in the surface head ectoderm occurs independently of Sim, whereas later Egfr activation in midbrain cells depends on Sim. This later Egfr activation period (ES 11) occurs within an hour of the critical time at which Egfr function is required for normal midbrain development based on Egfr temperature shift experiments (late ES 10). My second approach was to express a constitutively active form of Egfr by using a nervous system-specific driver in sim⫺ embryos to see if this is sufficient to rescue the sim mutant phenotype. These embryos showed restoration of the midbrain, including ganglia and axon tracts (Figure 3H), in 55% of embryos (n ⫽ 20). The average midbrain area, as assayed by measuring the area occupied by Elav-expressing cells, in rescued embryos was 1911 ␮m2 (⫾267 ␮m2) (n ⫽ 20). These results argue that Sim acts upstream of Egfr activation in the midbrain and that the brain phenotype resulting from disrupting Egfr signaling can be accounted for by a loss of Egfr activation in the midbrain. This cooperative influence of Sim and Egfr signaling originating from the midline may be a situation unique within the Drosophila CNS; in the ventral nerve cord, midline cells influence the development of lateral CNS tissue, but they do so in a manner that is not directly mediated by the Egfr pathway [3]. The observation that both rhomboid and vein mutant embryos have weaker midbrain phenotypes than single-minded or Egfr mutants means that Egfr ligands other than Spitz and Vein may also function in midbrain development.

Perturbations in Egfr Signaling Changes the Pattern of Proliferation in the Brain What is the cause of the reduced midbrain in the absence of Sim and Egfr signaling? One possibility is that the pattern of proliferation in midbrain neuroblasts is disrupted. In wild-type embryos at early ES 12, cells undergoing mitosis are visible among the neuroblasts in the midbrain (Table 2, Figure 3I). Importantly, in sim⫺ embryos, the number of midbrain cells undergoing mitosis is reduced (Table 2, Figure 3J). Also, increasing midline production of Egfr ligand results in an increase in proliferating midbrain cells (Table 2, Figure 3K). These results demonstrate a connection between proliferation and Sim function and argue that reduced proliferation contributes to the loss of midbrain tissue in sim mutants.

Midline Signaling in the Drosophila Brain 479

Apoptosis Does Not Increase until Late Embryogenesis, When Egfr Signaling Is Disrupted Does Egfr signaling function to maintain cell survival during brain development, as happens in the ventral nerve cord [16, 17] and in the dorsal-most region of the brain [10]? I looked at the pattern of apoptotic cells in the midbrains of wild-type and sim⫺ embryos by using the TUNEL method, and I found no difference until late embryogenesis, when there is a slight increase (Figures 3L–3O). The cells dying may be glia associated with the circumesophageal (longitudinal) connectives that need Egfr signaling to survive; a similar mechanism has been demonstrated in the ventral nerve cord [17]. Influence of sim and Egfr Mutants on Populations of Brain Cells The midbrain normally expresses the gap gene empty spiracles and the homeotic gene labial, both of which are required for normal midbrain development [18–20]. Thus, I examined the expression of Empty spiracles and Labial in the brains of sim⫺ embryos. Empty spiracles is normally expressed in the deutocerebral and tritocerebral neuromeres [19]. In sim⫺ embryos, expression in the midbrain is normal early, but at later stages, fewer cells express this protein (Table 3, Figure 4A). Removing Egfr function results in a similar phenotype (Table 3, Figure 4A). Labial is normally expressed in the tritocerebral neuromere [21], and, in sim⫺ embryos, the number of Labial-expressing midbrain cells appears to be normal (Figure 4A). A similar result was seen in Egfr⫺ embryos (Figure 4A). Loss of forkhead Function Affects the Number of Empty Spiracles Expressing Brain Cells What happens to Empty spiracles expression when the foregut source of Sim is removed? When the foregut is disrupted genetically (by using Forkhead⫺ embryos [22]), the brain is diminished as compared to wild-type [23]. In forkhead⫺ embryos, the number of midbrain Empty spiracles-expressing cells is reduced (Table 3, Figure 4A). Also, Labial expression in the midbrain appears normal (Figure 4A). Forkhead is expressed in the neural midline [22], so I wanted to see if forkhead mutants lack brain midline glia, which could signal to the midbrain (their position just ventral to the foregut is shown in Figures 1B–1D). In forkhead mutant embryos, the midline glia ventral to the foregut are still present (Figures 4B and 4C). These results show that the foregut is a likely source of signals for the developing midbrain. Conclusions These data lead to a model for how Egfr signaling helps pattern the midbrain in Drosophila (summarized in Figure 5A). Sim expression in the foregut leads to production of an Egf signal that is received by midbrain cells and expands a population of neuroblasts, some of which express Empty spiracles, by promoting proliferation. These findings may help us understand how the brain arose in evolution. Snodgrass [24] argued that the archicerebrum (the primitive arthropod brain) was located dorsal to the foregut, with minimal brain ganglia lateral

Figure 4. Effects of Sim and Egfr Mutants on Empty Spiracles and Labial Expression in the Midbrain (A) All embryos are viewed laterally and are at ES 12. The approximate limits of the midbrain are marked with dashed black lines. This panel shows that the expression of Empty spiracles in the midbrain is diminished in sim⫺ embryos, or when Egfr function is removed immediately after collection, or in forkhead (fkh)⫺ embryos. Labial expression does not appear to be effected. (B and C) ES 15 wild-type and fkh⫺ embryos, respectively, with midline glia marked with anti-Wrapper antibody. There does not appear to be a loss of midline glia in fkh⫺ embryos. Note that Wrapper is not expressed in the midbrain at any developmental stage in wild-type or fkh⫺ embryos.

to the foregut, and that there were connectives that connected the archicerebrum to the ventral neural mass (subesophageal ganglia). We can imagine a situation in

Current Biology 480

Figure 5. Summary Schemes (A) A hypothesis regarding the action of Egfr signaling in Drosophila midbrain development. Sim-expressing cells of the foregut produce an Egf signal (red with white arrows) that is received by midbrain neuroblasts (blue), leading to proliferation of midbrain neuroblasts and ultimately to the filling in of the brain lateral to the foregut with ganglia. (B) A hypothesis regarding the evolution of the arthropod brain. The ancestral arthropod brain consisted of an archicerebrum, a ventral (subesophageal) neural mass, and connectives that connected these two structures around the foregut, which had minimal brain ganglia lateral to it. Among the patterning molecules expressed in this ancestral brain were foci of Empty spiracles (blue) and Labial (green) expression. As arthropods evolved, ganglia adjacent to the foregut expanded with the help Egfr to fill the area lateral to the foregut with brain ganglia.

which sim expression extended anterior from the anlage of the neural midline into the foregut anlage, thus allowing strong production of an Egf signal that facilitated increased proliferation of neuroblasts in the adjacent brain (Figure 5B). Ultimately, this led to the expansion of brain ganglia so that the area lateral to the foregut was filled in with ganglia. Considering the view that developmental timing may reflect phylogeny, it is worth keeping in mind that the ganglia lateral to the foregut form last in the Drosophila brain. This provides a basis for speculation that the brain ganglia lateral to the foregut are an evolutionarily recent addition in embryonic arthropods. We can also use these findings to revisit a hypothesis put forward by Anton Dohrn in the late 19th century [25] and subsequently reiterated and elaborated upon by others, including Richard Owen [26], and more recently by Arendt and Nu¨bler-Jung [27–29]: the region of the diencephalon that includes the pituitary and infundibulum in vertebrates represents the site where the ancestral mouth was located. According to this hypothesis, a common ancestor of bilaterally symmetric animals had a brain that formed around the foregut. As protostomes and deuterostomes diverged, the old mouth was lost, and a new mouth formed on what was the dorsal side of the animal. A prediction can be made from this hypothesis: if the ancestral arrangement of the foregut and brain in Bilateria was like the one seen in protostomes (the foregut passing through the brain), and if the pituitary marks the spot where the ancestral mouth was located in deuterostomes, then the neural midline in deuterostomes might stop in the diencephalon where the pituitary is located. Importantly, in vertebrates, the floor plate cells of the CNS extend from the caudal CNS and stop at the diencephalon, the region where the pituitary is located. In the forebrain anterior to the floor plate are rostal ventral midline cells that express a different set of markers than those expressed in the floor plate,

although there is some overlap [30]. Furthermore, the anterior pituitary itself derives from an outpocketing of stomodeum (mouth) ectoderm called Rathke’s pouch. It is worth considering that the anterior limit of the floor plate in the vertebrate diencephalon, adjacent to where the pituitary is located, represents the vestige of an arrangement present in an ancestor of protostomes and deuterostomes where the neural midline extended into the primitive brain and stopped at the mouth, which then pulled out of the brain as deuterostomes evolved. This removal of the mouth from the brain midline would have created additional space for neural tissue in the brain, possibly corresponding with the development of complex behaviors in deuterostomes. The degree of complexity of the brain in the last common ancestor of protostomes and deuterostomes is something that is not clear at present. While molecules such as Hedgehog/Sonic hedgehog [31, 32], Sim/Sim1 [11, 33], and Forkhead/HNF3-␤ [22, 34] are expressed in the neural midlines of protostomes and deuterostomes, their function in the development of these tissues appears to vary. For example, in Drosophila forkhead⫺ embryos, midline glia in the subesophageal ganglion are not grossly disrupted, whereas HNF3-␤ appears to have a more critical function during floor plate development in the vertebrate CNS [35]. Thus, it is possible that in the last common ancestor of Bilateria there was a primitive collection of neural cells at the midline that expressed these markers, and as protostomes and deuterostomes diverged, these molecules were differentially recruited to make more elaborate neural midlines. Experimental Procedures Fly Stocks As a wild-type fly stock, Oregon-R flies were used. The following strains were also used (all strains are described in Flybase): 1407GAL4, UAS-Egfr* (constitutively active Egfr, Egfr::toract.Scer\UAS), UASlacZ, UAS-tau-GFP (Avic\GFPScer\UAS.T:Btau\MAPT), UAS-ve32 (UAS-rhom-

Midline Signaling in the Drosophila Brain 481

boid), Egfrf1 (temperature-sensitive allele), forkhead6, rhomboid7M43 (this stock also contains roughoid1, which is a viable hypomorphic allele of a gene involved in Egfr signaling [36]; however, this allele on its own does not appear to have a gross effect on brain development [data not shown]), rhomboidX81 (rhomboid-lacZ), sim2, and sim-GAL4 (Scer\GAL4sim.PS). vein⌬P10567 [37] was also used. For identifying mutant embryos, balancers expressing GFP or ␤-Galactosidase were used. Temperature Shift Experiments For Egfr temperature shift experiments, Egfrf1 embryos were collected on plates at the permissive temperature (18⬚C) for 1 hr, and the midpoint of the collection period was taken as the time of egg laying. Embryos were temperature shifted from the permissive temperature to the restrictive temperature (29⬚C) at the following stages: immediately after collection, ES 9, ES 10, and ES 12. Staging was done according to [13], with the following adjustments: at the permissive temperature, embryos took approximately twice as long to develop as at 25⬚C, and, at the restrictive temperature, embryos took approximately 0.8 times as long to develop as at 25⬚C. All staging was confirmed with a microscope. Egfrf1 embryos raised entirely at the permissive temperature did not show any loss of midbrain (data not shown). In Situ Hybridization rhomboid RNA probe (synthesized from cDNA from E. Bier) was generously provided by A. Baonza, and vein RNA probe (synthesized from cDNA from A. Simcox) was generously provided by B. Olofsson. Whole mount in situ hybridization with these probes was done as described in [38], with minor modifications. For whole mount TUNEL staining, an in situ Cell Death Detection Kit (Roche) was used as previously described [39], with the following modifications: permeabilization of embryos was done in 100 mM sodium citrate and 0.5% Triton X-100 at 70⬚C for 30 min. Embryos were then rinsed in TUNEL Dilution Buffer (Roche) and incubated with the TUNEL reaction mixture for 3 hr at 37⬚C. Immunohistochemistry The primary antibodies used in this study were anti-␤-Galactosidase (Cappel) (Rabbit, 1:1500), anti-DpERK (Sigma) (mouse, 1:1000), antiElav ([40], Developmental Studies Hybridoma Bank [DSHB]) (rat, 1:100), anti-Empty spiracles [41] (rat, 1:500), anti-Fasciclin II [42] (mouse, 1:20), anti-GFP (rabbit, 1:100) (Molecular Probes), antiLabial [21] (rabbit, 1:100), anti-Phosphohistone-H3 (Upstate Biotechnology) (rabbit, 1:2000), anti-Repo [43] (rabbit, 1:1000), anti-Sim (a gift from S. Crews) (mouse, 1:20), anti-Wrapper (DSHB) (mouse, 1:50), and BP102 (DSHB) (mouse, 1:100). The secondary antibodies used were anti-mouse, Cy3 conjugated (Jackson, 1:200); anti-rabbit, Cy2 conjugated (Jackson, 1:200); anti-rat, Cy5 conjugated (Jackson, 1:200); anti-rabbit, biotinylated (Jackson, 1:200); anti-rat, biotinylated (Jackson, 1:200); and anti-mouse, biotinylated (Jackson, 1:200). Embryos were imaged with a BioRad 1024 or Radiance laser confocal microscope or a Zeiss Axiophot light microscope. Quantification of Brain Phenotypes For this study, the midbrain is defined as the brain ganglia that lie lateral to the foregut, with positions confirmed by using dorsoventral placement and landmarks in the head epidermis. Estimates of brain size in mutant backgrounds (Table 1) were done as follows: optical sectioning was done with a 60⫻ objective on a BioRad Radiance confocal microscope. Sections were taken at 2.5 ␮m increments from the lateral edge of the foregut to 25 ␮m into the brain. Images were then projected into one plane by using BioRad software, and Adobe Photoshop was then used to trace the Elav-marked midbrain. The pixel area was then used to estimate the area of the brain occupied by neuronal nuclei. Counts of anti-Phosophohistone-H3 cells and anti-Empty spiracles-expressing cells were done by eye with a Zeiss Axiophot microscope. Statistical analyses of differences in wild-type and mutant brains were done by using the Mann-Whitney test, with the significance level set at 0.05. Embryos were selected for statistical analysis from the population being measured at random, without bias toward phenotype. Embryos showing full mutant phenotypes described herein were also measured to give

an estimate of the effects of mutation on midbrain development. For all groups of embryos measured, n ⫽ 20. The data presented here are not intended to represent the absolute size of the brain, but rather to give a comparison of approximate brain area between groups of embryos. Acknowledgments To Peter Lawrence, I am grateful for laboratory space and support. For useful discussions regarding this work, I am grateful to Jonathan Bacon, Michael Bate, Matthew Freeman, and Klaus Sander. Thanks to Yao Chen for developing the protocol for TUNEL staining in the brain. For stocks and/or antibodies, I am grateful to Manuel Calleja, Jose´ Casal, Steve Crews, Matthew Freeman, Corey Goodman, Alicia Hidalgo, Thomas Kaufman, Andrew Travers, the Bloomington Stock Center, and The Developmental Studies Hybridoma Bank. During the course of this work, I was supported by a Medical Research Council Laboratory of Molecular Biology/Peterhouse Research Studentship. Received: July 15, 2002 Revised: December 17, 2002 Accepted: January 6, 2003 Published: March 18, 2003 References 1. Sasai, Y., and De Robertis, E.M. (1997). Ectodermal patterning in vertebrate embryos. Dev. Biol. 182, 5–20. 2. Nambu, J.R., Franks, R.G., Hu, S., and Crews, S.T. (1990). The single-minded gene of Drosophila is required for the expression of genes important for the development of CNS midline cells. Cell 63, 63–75. 3. Menne, T.V., Luer, K., Technau, G.M., and Klambt, C. (1997). CNS midline cells in Drosophila induce the differentiation of lateral neural cells. Development 124, 4949–4958. 4. Nambu, J.R., Lewis, J.O., Wharton, K.A., Jr., and Crews, S.T. (1991). The Drosophila single-minded gene encodes a helixloop-helix protein that acts as a master regulator of CNS midline development. Cell 67, 1157–1167. 5. Golembo, M., Raz, E., and Shilo, B.Z. (1996). The Drosophila embryonic midline is the site of Spitz processing, and induces activation of the EGF receptor in the ventral ectoderm. Development 122, 3363–3370. 6. Lee, J.R., Urban, S., Garvey, C.F., and Freeman, M. (2001). Regulated intracellular ligand transport and proteolysis control EGF signal activation in Drosophila. Cell 107, 161–171. 7. 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. 8. Lanoue, B.R., Gordon, M.D., Battye, R., and Jacobs, J.R. (2000). Genetic analysis of vein function in the Drosophila embryonic nervous system. Genome 43, 564–573. 9. Mayer, U., and Nusslein-Volhard, C. (1988). A group of genes required for pattern formation in the ventral ectoderm of the Drosophila embryo. Genes Dev. 2, 1496–1511. 10. Dumstrei, K., Nassif, C., Abboud, G., Aryai, A., and Hartenstein, V. (1998). EGFR signaling is required for the differentiation and maintenance of neural progenitors along the dorsal midline of the Drosophila embryonic head. Development 125, 3417–3426. 11. Crews, S.T., Thomas, J.B., and Goodman, C.S. (1988). The Drosophila single-minded gene encodes a nuclear protein with sequence similarity to the per gene product. Cell 52, 143–151. 12. Xiong, W.C., Okano, H., Patel, N.H., Blendy, J.A., and Montell, C. (1994). repo encodes a glial-specific homeo domain protein required in the Drosophila nervous system. Genes Dev. 8, 981–994. 13. Campos-Ortega, J., and Hartenstein, V. (1997). The Embryonic Development of Drosophila melanogaster, Second Edition (Berlin: Springer). 14. Therianos, S., Leuzinger, S., Hirth, F., Goodman, C.S., and Reichert, H. (1995). Embryonic development of the Drosophila

Current Biology 482

15.

16.

17.

18.

19.

20.

21.

22.

23. 24. 25.

26.

27.

28.

29. 30.

31.

32.

33.

34.

35. 36.

brain: formation of commissural and descending pathways. Development 121, 3849–3860. 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. Gabay, L., Seger, R., and Shilo, B.Z. (1997). In situ activation pattern of Drosophila EGF receptor pathway during development. Science 277, 1103–1106. Hidalgo, A., Kinrade, E.F., and Georgiou, M. (2001). The Drosophila neuregulin vein maintains glial survival during axon guidance in the CNS. Dev. Cell 1, 679–690. Hirth, F., Hartmann, B., and Reichert, H. (1998). Homeotic gene action in embryonic brain development of Drosophila. Development 125, 1579–1589. Hirth, F., Therianos, S., Loop, T., Gehring, W.J., Reichert, H., and Furukubo-Tokunaga, K. (1995). Developmental defects in brain segmentation caused by mutations of the homeobox genes orthodenticle and empty spiracles in Drosophila. Neuron 15, 769–778. Page, D.T. (2000). labial acts to initiate neuronal fate specification, but not axon pathfinding, in the embryonic brain of Drosophila. Dev. Genes Evol. 210, 559–563. Diederich, R.J., Merrill, V.K., Pultz, M.A., and Kaufman, T.C. (1989). Isolation, structure, and expression of labial, a homeotic gene of the Antennapedia Complex involved in Drosophila head development. Genes Dev. 3, 399–414. Weigel, D., Jurgens, G., Kuttner, F., Seifert, E., and Jackle, H. (1989). The homeotic gene fork head encodes a nuclear protein and is expressed in the terminal regions of the Drosophila embryo. Cell 57, 645–658. Page, D.T. (2002). Inductive patterning of the embryonic brain in Drosophila. Development 129, 2121–2128. Snodgrass, R. (1935). Principles of Insect Morphology (New York: McGraw-Hill). Dohrn, A. (1875). Der Ursprung der Wirbelthiere und das Princip des Functionswechsels. (Leipzig: Verlag von Wilhelm Engelmann). Owen, R. (1883). Essays on the Conario-Hypophysial Tract and on the Sspects of the Body in Vertebrate and Invertebrate Animals (London: Taylor & Francis). Arendt, D., and Nubler-Jung, K. (1996). Common ground plans in early brain development in mice and flies. Bioessays 18, 255–259. Arendt, D., and Nubler-Jung, K. (1999). Comparison of early nerve cord development in insects and vertebrates. Development 126, 2309–2325. Nubler-Jung, K., and Arendt, D. (1994). Is ventral in insects dorsal in vertebrates? Roux’s Arch. Dev. Biol. 203, 357–366. Dale, J.K., Vesque, C., Lints, T.J., Sampath, T.K., Furley, A., Dodd, J., and Placzek, M. (1997). Cooperation of BMP7 and SHH in the induction of forebrain ventral midline cells by prechordal mesoderm. Cell 90, 257–269. Echelard, Y., Epstein, D.J., St-Jacques, B., Shen, L., Mohler, J., McMahon, J.A., and McMahon, A.P. (1993). Sonic hedgehog, a member of a family of putative signaling molecules, is implicated in the regulation of CNS polarity. Cell 75, 1417–1430. Lee, J.J., von Kessler, D.P., Parks, S., and Beachy, P.A. (1992). Secretion and localized transcription suggest a role in positional signaling for products of the segmentation gene hedgehog. Cell 71, 33–50. Fan, C.M., Kuwana, E., Bulfone, A., Fletcher, C.F., Copeland, N.G., Jenkins, N.A., Crews, S., Martinez, S., Puelles, L., Rubenstein, J.L., et al. (1996). Expression patterns of two murine homologs of Drosophila single-minded suggest possible roles in embryonic patterning and in the pathogenesis of Down syndrome. Mol. Cell. Neurosci. 7, 1–16. Ang, S.L., Wierda, A., Wong, D., Stevens, K.A., Cascio, S., Rossant, J., and Zaret, K.S. (1993). The formation and maintenance of the definitive endoderm lineage in the mouse: involvement of HNF3/forkhead proteins. Development 119, 1301–1315. Sasaki, H., and Hogan, B.L. (1994). HNF-3 beta as a regulator of floor plate development. Cell 76, 103–115. Wasserman, J.D., Urban, S., and Freeman, M. (2000). A family

37.

38. 39.

40.

41.

42.

43.

of rhomboid-like genes: Drosophila rhomboid-1 and roughoid/ rhomboid-3 cooperate to activate EGF receptor signaling. Genes Dev. 14, 1651–1663. Yarnitzky, T., Min, L., and Volk, T. (1997). The Drosophila neuregulin homolog Vein mediates inductive interactions between myotubes and their epidermal attachment cells. Genes Dev. 11, 2691–2700. Lehmann, R., and Tautz, D. (1994). In situ hybridization to RNA. Methods Cell Biol. 44, 575–598. Wang, S.L., Hawkins, C.J., Yoo, S.J., Muller, H.A., and Hay, B.A. (1999). The Drosophila caspase inhibitor DIAP1 is essential for cell survival and is negatively regulated by HID. Cell 98, 453–463. O’Neill, E.M., Rebay, I., Tjian, R., and Rubin, G.M. (1994). The activities of two Ets-related transcription factors required for Drosophila eye development are modulated by the Ras/MAPK pathway. Cell 78, 137–147. Macias, A., and Morata, G. (1996). Functional hierarchy and phenotypic suppression among Drosophila homeotic genes: the labial and empty spiracles genes. EMBO J. 15, 334–343. Bastiani, M.J., Harrelson, A.L., Snow, P.M., and Goodman, C.S. (1987). Expression of fasciclin I and II glycoproteins on subsets of axon pathways during neuronal development in the grasshopper. Cell 48, 745–755. Halter, D.A., Urban, J., Rickert, C., Ner, S.S., Ito, K., Travers, A.A., and Technau, G.M. (1995). The homeobox gene repo is required for the differentiation and maintenance of glia function in the embryonic nervous system of Drosophila melanogaster. Development 121, 317–332.