Delta-1 Is a Regulator of Neurogenesis in the Vertebrate Retina

DEVELOPMENTAL BIOLOGY 185, 92–103 (1997) ARTICLE NO. DB978546

Delta-1 Is a Regulator of Neurogenesis in the Vertebrate Retina Iqbal Ahmad, Constance M. Dooley, and Dorisa L. Polk Department of Cell Biology and Anatomy, and Department of Ophthalmology, University of Nebraska Medical Center, Omaha, Nebraska 68198-6395

In the retina, cell fate determination is thought to be regulated by a series of local cell–cell interactions. Evidence suggests that retinal precursors utilize Notch-mediated intercellular signaling to regulate their fates. However, the identity of the endogenous ligand and its role in the Notch-signaling pathway is not well understood. We have identified C-Delta-1 as the putative endogenous ligand for Notch, in the developing chick retina. C-Delta-1 is coexpressed spatially and temporally with C-Notch-1 and their expression is associated with the temporal aspects of cell birth in the developing retina. This suggests that Delta–Notch signaling is utilized to maintain progenitors in an uncommitted state and that a subtle fluctuation in this signaling helps to sort out competent cells during successive cell-fate determination. We have tested the latter possibility in the specification of the ganglion cells. In early stages of retinal development when ganglion cells are the predominant cells born, decreasing C-Delta-1 expression with antisense oligonucleotides increases the proportion of RA4 antigen-expressing ganglion cells which are recruited predominantly in the periphery. Conversely, use of exogenous Drosophila Delta leads to a decrease in the RA4 antigen-expressing ganglion cells. Our results suggest that C-Delta-1 activation of the Notch pathway regulates the specification of retinal neurons in general and of ganglion cells in particular. q 1997 Academic Press

INTRODUCTION In the vertebrate retina neurons develop from neuroepithelial cells lining the inner wall of the developing optic vesicle (reviewed in Robinson, 1991). Thymidine birthdating analyses in several species have shown that these progenitors give rise to retinal neurons in a temporal order that is evolutionarily conserved (Sidman, 1961; Kahn, 1974; Young, 1985; Prada et al., 1991; LaVail et al., 1991). Retinal ganglion cells (RGC) are the first neurons which are born and the bipolar cells are usually one of the last retinal cell types to become postmitotic. In addition to this temporal pattern, cell fate specification in retina follows a stereotypical spatial pattern; retinal neurons are born first in the central retina and last in the periphery with the gradient of cell differentiation spreading from the center to the periphery (Harman and Beazley, 1989; Robinson, 1991; Prada et al., 1991; Lavail et al., 1991). Evidence from a variety of experimental approaches including cell ablation experiments (Reh and Tully, 1986)) and lineage analyses (Turner and Cepko, 1987; Price et al., 1987; Holt et al., 1988; Wetts and Fraser, 1988; Turner et al., 1990) suggests that the retinal precursors are multipotential and that the decision taken by a

precursor to follow a particular path depends on local cell interactions. The molecular and cellular mechanisms underlying cellfate regulation by cell interactions are not well understood. However, recent evidence suggests that most precursors, both neural and nonneural, utilize a very general and evolutionarily conserved intercellular signaling mechanism, the Notch pathway, to regulate their fates (reviewed in Greenwald and Rubin, 1992; Muskavitch, 1994; ArtavanisTsakonas et al., 1991, 1995). In this pathway, Notch acts as a cellular receptor that participates in cell-fate specification by regulating the competence of a differentiating cell to respond to epigenetic cues. The ligands for Notch belong to a family of membrane-anchored proteins that contain EGF repeats and a cysteine-rich DSL (Delta–Serrate–Lag2) domain in the extracellular region (reviewed in Kopan and Nye, 1996). These putative ligands include Delta and Serrate in Drosophila (Vaessin et al., 1987; Kopczynski et al., 1988; Fleming et al., 1990; Thomas et al., 1991), Apx-1 and Lag-2 in Caenorhabditis elegans (Tax et al., 1994; Henderson et al., 1994; Gao and Kimble, 1995; Fitzgerald and Greenwald, 1995) and Delta-1, Serrate-1, and Jagged in vertebrates (Chitnis et al., 1995; Henrique et al., 1995; Lindsell et al., 1995; Myat et al., 1996). Evidence in Drosophila sug0012-1606/97 $25.00 Copyright q 1997 by Academic Press All rights of reproduction in any form reserved.

92

AID

DB 8546

/

6x22$$$$$1

04-09-97 20:11:01

dba

93

Delta–Notch Signaling in the Retina

gests that the binding of ligands activates the Notch receptor. For example, null mutation of Delta results in the neurogenic loss of function phenotype that is also observed in a loss of function mutation of Notch (Heitzler and Simpson, 1991; Greenwald and Rubin, 1992). Serrate, which is closely related to Delta structurally and in its interaction with Notch (Rebay et al., 1991; Muskavitch, 1994), is able to rescue the neurogenic phenotype in Delta mutant embryos (Gu et al., 1995). Similar observations that Notch signaling is facilitated by the activation of Notch receptor by ligands have been reported in vertebrates. For example, the suppression of neurogenesis by constitutively activated Notch is mimicked by the overexpression of Delta-1 in Xenopus (Chitnis et al., 1995) and Jagged, the rat homolog of Serrate, suppresses the differentiation of cultured myoblasts expressing Notch-1 (Lindsell et al., 1995). Recent observations indicate that, like in Drosophila (Cagan and Ready, 1989; Fortini et al., 1993), the Notch pathway plays a significant role in the specification of retinal neurons in vertebrates. Analyses of the pattern of Notch-1 expression in rat retina have suggested that Notch signaling regulates temporal as well as spatial aspects of retinal neurogenesis (Ahmad et al., 1995). It is thought that the Notch pathway maintains a population of immature cells in an uncommitted state until proper cues for differentiation become available. For example, expression of constitutively active Notch-1 in Xenopus retina prevents immature cells from differentiating (Dorsky et al., 1995). Conversely, a decrease in Notch-1 signaling promotes differentiation in the retina (Austin et al., 1995). While these observations strongly suggest that Notch signaling plays a significant role in the specification of retinal neurons, the endogenous ligand(s) involved in regulating the signaling have not been identified. In Drosophila, Delta and Serrate participate in Notch signaling to facilitate different developmental decisions. While Delta–Notch signaling mediates lateral inhibition that leads to the segregation of neuroblasts (Muskovitch, 1994; Artavanis-Tsakonas et al., 1995), Serrate– Notch signaling participates in the organization of the wing imaginal disc (Speicher et al., 1994; Diaz-Benjumea and Cohen, 1995; Couso et al., 1995). However, in later developmental processes both Delta and Serrate show overlapping but distinct expression patterns, suggesting that developmental decisions in the same tissue may utilize either Serrate or Delta, and in some cases both ligands (Gu et al., 1995). In retina, where Notch-1 activity is correlated with successive cell-fate decisions (Ahmad et al., 1995) the possibility exists that the Notch pathway utilizes either one or different ligands for the specification of different cell types. The observation that in a coculture study Drosophila Delta was able to suppress RGC specification suggests that Drosophila Delta mimics a vertebrate ligand that activates Notch during retinal development (Austin et al., 1995). That vertebrate ligand (s) could be homologs of either Delta or Serrate based on the following observations: (1) In Drosophila Serrate can substitute for Delta in the activation of the Notch pathway during neurogenesis (Gu et al., 1995).

(2) In developing chick hindbrain and spinal cord C-Serrate1 may perform the same function as C-Delta-1 in complementary domains during neurogenesis (Myat et al., 1996). In this study we have identified and analyzed the role of the putative endogenous ligand that activates Notch signaling during retinal neurogenesis. We show that C-Delta-1 appears to be the dominant endogenous ligand for the Notch pathway in chick retina. Both C-Delta-1 and C-Notch-1 genes are coexpressed spatially and temporally and the levels of their transcripts correspond to stages of cell-fate specification, suggesting the involvement of Delta–Notch signaling in successive cell-fate determination as well as the spatial specification of retinal neurons. We analyzed the effect of a decrease in Delta-1 expression on the distribution of RA4 antigen, an early marker for RGC differentiation (McLoon and Barnes, 1989; Waid and McLoon, 1995). In the developing retina, RA4 immunoreactivity showed a distinct central to peripheral gradient and was largely absent from the periphery at Embryonic Day (E) 4. A decrease in the levels of C-Delta-1 transcripts in E4 retina in response to antisense treatment increased the number of RA4-positive RGCs which appear to be recruited more in the periphery than in the center of the retina. A similar increase in the number of RA4-positive RGCs was also observed when the expression of C-Notch-1 was decreased. Additionally, exposing developing retinal cells to extraneous Delta in explant cultures prevented the recruitment of the RA4-positive cells to the peripheral retina. These results suggest that C-Delta-1 activation of the Notch pathway is required for the temporal and spatial specification of the retinal neurons.

MATERIALS AND METHODS Embryos. Fertilized hens’ eggs were incubated in a humidified chamber at 387C. Embryos were staged according to Hamburger and Hamilton (1951). RT–PCR. DNase-digested total RNA (5 mg) from each developmental stage was used to synthesize cDNA, using a random hexamer as a primer (Grillo and Margolis, 1991; Ahmad, 1995). Onetenth (5 ml) of the cDNA reaction/developmental stage was amplified using either Delta-1 specific (5*-CACCAGCCCGAGGCCTGC-3*; 5*-GATGCACTCATCTTTCTC-3*) (Henrique et al., 1995) or Notch-1 specific (Ahmad et al., 1995; Myat et al., 1996) primers for 25-step cycles (947C, 1 min; 487C, 1 min; 727C, 1.5 min). The amplification after 25-step cycles was linear as ascertained by amplification of serially diluted cDNA. PCR products were resolved on 1.5% agarose gel and Southern analyses were carried out using C-Delta-1 or C-Notch-1 cDNA as probes (Henrique et al., 1995; Myat et al., 1996). A 548-bp b-actin cDNA was amplified using 5 ml of cDNA/developmental stage as a measure of the constitutive expression and RNA integrity (Ahmad et al., 1995). To ascertain the effect of antisense oligonucleotides corresponding to CDelta-1 and C-Notch-1 transcripts cDNA was synthesized from equal amounts of total RNA isolated from explant culture. Because of the small amount of total RNA extracted (approximately 5 mg) the possible variation between the samples was further controlled by monitoring the amount of cDNA synthesized using 32P-labeled dCTP in an aliquot of cDNA reaction. RT–PCR was carried out

Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.

AID

DB 8546

/

6x22$$$$$1

04-09-97 20:11:01

dba

94

Ahmad, Dooley, and Polk

as described and stage-specific normalization of the Delta-1- and Notch-1-specific transcript was carried out using b-actin-amplified product. In situ hybridization. Probes labeled with 35S were transcribed using C-Delta-1 or C-Notch-1 cDNA as template. In situ hybridization was carried out on fresh frozen retinal sections (12 mM) as previously described (Ahmad et al., 1994, 1995). Northern analyses. Northern analyses were performed on 20 mg of total RNA isolated from E3.5 retina as previously described (Ahmad, 1995) using 32P-labeled C-Delta-1 or C-Notch-1 cDNA probes. The final wash was carried out at 657C for 1 hr in 11 SSC and 0.1% SDS. Explant culture and antisense treatment. Retinae were harvested from stage-specific embryos in Hanks’ balanced salt solution (HBSS). The explants were cultured in 96-well plates in DMEM:F12 medium containing 11 N2 supplement (Gibco), 1% fetal bovine serum (FBS), 2 mM L-glutamine, 100 U/ml penicillin, and 100 mg/ ml streptomycin at 377C in 95% humidity and 5% CO2 . Following 24 hr of incubation with and without antisense, sense or missense oligonucleotides (50 mM), the explants were fixed in Streck Tissue Fixative (Streck Laboratories) and cryoprotected in 30% sucrose for immunocytochemistry. For dissociation into single cell suspension explants were incubated in HBSS (Ca2/ and Mg2/ free) containing 0.25% trypsin, 1 mM EDTA, and 20 mg/ml DNase1 at 377C for 20 min. Trypsin was neutralized by washing the tissue in HBSS containing 20% FBS. Cells were dissociated by trituration (10–15 times) in the culture medium and plated at a density of 103 –104 cells/glass coverslip treated with 50–100 mg/ml of poly-D-lysine. Cells were allowed to adhere to the coverslips for 2–3 hr at 377C before fixation in 4% paraformaldehyde. Each explant culture was analyzed for cell viability using the trypan blue dye-exclusion method. Experiments were performed at least three times and repeated in triplicate. On average, 35 cells were counted in four different fields per treatment. The antisense, sense, and missense oligonucleotides were synthesized at Yale University (Department of Pathology) and University of Nebraska Medical Center (Eppley Cancer Research Center) and purified by HPLC. Sequence were as follows: for b-globin antisense, 5*-AGTCCAGTGCACCAT-3* (Represa et al., 1991); Notch-1 antisense, 5*-CCTCCGCTGCAGGAGGCAATCAT-3* (Weinmaster et al., 1991; Austin et al., 1995); Notch-1 sense, 5*-ATGATTGCCTCCTGCAGCGCAGG-3*; Notch-1 missense, 5*-CGTAGTGACTACAGAGCGCTCCC-3*; Delta-1 antisense, 5*-GACCTTCTCGCCACGCTC-3* (Henrique et al., 1995); Delta-1 sense, 5*-GAGCGTGGCGAGAAGGTC-3*; Delta-1 missense, 5*-TCGTCCACCGATCCGTCC-3*. For the eyecup culture, eyes were removed from E4 embryos, placed in HBSS, and hemisected removing the lens to ensure that the peripheral tips of the retina which remain devoid of RA4 immunoreactivity (McLoon and Barnes, 1989) were not included in the culture. The eye-cups were incubated for 24 hr in culture medium containing 50 mM C-Delta-1 antisense oligonucleotide in a 96-well plate. The medium was changed once halfway through the culture. The eye cups were fixed in STF fixative for 10 minutes, cryoprotected overnight in 30% sucrose, sectioned, and analyzed for RA4 immunoreactivity. To distinguish the central from the peripheral retina we scored for retinal sections in which a labeled region was flanked by an unlabeled region for RA4 immunoreactivity. Because the unlabeled tips were removed prior to culturing, any region devoid of RA4 immunoreactivity represented true periphery. In two of four repeat experiments sections were obtained in both treated and untreated groups in which the central and peripheral retina can be distinguished by the above-mentioned criteria.

Drosophila S2 cell culture. Drosophila Delta cDNA-transfected and untransfected S2 cells were cultured as previously described (Fehon et al., 1990). To determine the effects of exogenous Delta on retinal development the lens was removed from E3.5 eye cups giving S2 cells access to the vitreal surface of the retina. The eye cup was filled with excess S2 cells and incubated in culture medium containing additional S2 cells for 24 hr. After culturing, the eye cups were fixed in STF fixative for 10 min, cryoprotected overnight in 30% sucrose, sectioned, and analyzed for RA4 immunocytochemistry. Immunocytochemistry. Immunocytochemistry was carried out as previously described (McLoon and Barnes, 1989; Ahmad et al., 1995). Briefly, whole sections or dissociated cells were incubated in PBS containing 5% normal goat serum and 0.2% Triton X-100 followed by an overnight incubation in RA4 antibody (1:1000) or Notch-1 antibody (20F) (1:100) at 47C. The sections were examined for epifluorescence using Leica DMR microscope following incubation in anti-mouse (RA4) or anti-rat (Notch-1) IgG conjugated to CY3.

RESULTS C-Delta-1 and C-Notch-1 Transcripts Are Coexpressed Temporally The identification of C-Delta-1 and subsequent analyses of the temporal pattern of expression of C-Delta-1 and CNotch-1 in the developing retina were carried out by semiquantitative RT-PCR analysis. cDNA was synthesized from equal amounts of total RNA isolated from retina from various developmental stages and used in PCR to amplify a 418and 298-bp region in C-Delta-1 and C-Notch-1 transcripts, respectively. The specificity of amplification was further confirmed by Southern analyses using C-Delta-1 and CNotch-1 cDNA as probes. The earliest stage that was analyzed for C-Delta-1 expression was E2.5, at which approximately 15% of ganglion cells have been generated (Prada et al., 1991). C-Delta-1 expression was detected at this stage and its levels increased until reaching a peak of expression at E4 and E7.5 (Fig. 1A). This is the period when all the retinal neurons and Mu¨ller glia are in various stages of differentiation (Fig. 1C). A decrease in the levels of C-Delta-1 transcripts was observed beginning at E10. Since one of the requisites for C-Delta-1 action is the presence of C-Notch1 receptor we wanted to know if C-Delta-1 is coexpressed temporally with C-Notch-1. Therefore, we analyzed the levels of C-Notch-1 transcript in embryonic stages in which the expression of Delta-1 was measured. The levels of CNotch-1 transcripts showed a pattern of temporal expression similar to that of C-Delta-1 transcripts with their highest levels detected at E4 and E7.5. (Fig. 1B). Therefore, the levels of C-Delta-1 expression appeared to be tightly correlated with that of C-Notch-1 expression and, together, their expression was associated with cell birth in the developing retina. To corroborate that transcripts amplified by RT–PCR were the products of C-Delta-1 genes expressed in the developing retina, we carried out Northern analyses on E3.5 total

Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.

AID

DB 8546

/

6x22$$$$$1

04-09-97 20:11:01

dba

95

Delta–Notch Signaling in the Retina

FIG. 2. Northern analyses of embryonic retinal mRNA. Twenty micrograms of total RNA obtained from Embryonic Day 4 retina was resolved on 1.4% denaturing agarose gel, transferred onto a nylon membrane, and hybridized with C-Delta-1 cDNA probe (A). The membrane was stripped of the previous probe and hybridized with C-Notch-1 cDNA probe (B). Molecular size (kb) markers correspond to RNA Ladder (BRL).

retinal RNA. A single hybridization band of approximately 3 kb corresponding to C-Delta-1 mRNA was observed when C-Delta-1 cDNA was used as a probe (Fig. 2A). Additionally, a slow-moving hybridization band was also detected which may correspond to cross-hybridizing Notch transcripts. Rehybridization of the stripped membrane with C-Notch-1 probe revealed an 11-kb hybridization band corresponding to the full-length C-Notch-1 mRNA (Fig. 2B).

C-Delta-1 and C-Notch-1 Transcripts Are Coexpressed Spatially

FIG. 1. Temporal expression of Delta-1 and Notch-1 transcripts in the developing retina. Five microliters of cDNA prepared from 10 m g of DNase-digested total RNA/developmental stage was amplified by PCR using primers specific for Delta-1 and Notch-1 genes. The PCR products were resolved on 1.5% agarose gels and analyzed by Southern analysis. The hybridization signals were densitometrically scanned and normalized with the stage-specific levels of b-actin transcripts. Relative levels of Delta-1 (A) and Notch-1 (B) transcripts at various stages of retinal development. Insets show hybridization signals corresponding to 418- and 298-bp PCR-amplified Delta-1 (A) and Notch-1 (B) transcripts. The Y axes represent arbitrary units of densitometric scanning. Schematic representation of neurogenesis in chick retina based on a [3H]thymidine birth dating study (Prada et al., 1991) (C). GC, ganglion cell; AC, amacrine cell; HC, horizontal cell; PC, photoreceptor cell; MC, Muller cell; BC, bipolar cell. The tips of the triangle indicate the stage at which 50% of neurogenesis for a specific cell-type is complete.

The fact that C-Delta-1 is a membrane protein suggests that in order to interact with the C-Notch-1 receptor and participate in the Notch-signaling pathway, it is expected to be expressed in close proximity to C-Notch-1. Therefore we studied the spatial distribution of C-Delta-1 and C-Notch-1 transcripts in E3.5 retina by in situ hybridization using 35S-labeled C-Delta-1 and C-Notch-1 probes. CDelta-1 transcripts were detected throughout the neural retina. However, the levels of transcripts as judged by the density of silver grains were relatively higher in the peripheral than in the central retina (Fig. 3A). Additionally, like in the developing chick brain (Henrique et al., 1995; Myat et al., 1996), the majority of C-Delta-1 transcripts were localized toward the vitreal surface, the surface that faces the lens and is analogous to the basal surface in the developing brain vesicles (Fig. 3B). In the same age embryo C-Notch-1 transcripts were detected in the retina and in adjoining structures such as the lens and retinal pigment epithelium (Fig. 3D). Consistent with the

Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.

AID

DB 8546

/

6x22$$$$$1

04-09-97 20:11:01

dba

96

Ahmad, Dooley, and Polk

distributed toward the ventricular surface (Fig. 3F). This pattern of distribution of Notch-1 immunoreactivity is similar to that observed in the developing brain vesicle where Notch-1 immunoreactivity is also detected in the ventricular zone (Ahmad et al., 1995) and suggests that the expression of C-Notch-1 is related to a population of cells that are proliferating.

The Ganglion Cell-Specific Antigen RA4 Shows Central to Peripheral Gradient

FIG. 3. Spatial expression of C-Delta-1 and C-Notch-1 genes. In situ hybridization was carried out on 15 mM E3.5 retinal sections using 35S-labeled probes. Antisense probe corresponding to C-Delta1 cDNA localizes C-Delta-1 transcripts above background in the neural retina (A). Note that C-Delta-1 transcripts are predominantly localized away from the ventricular surface (B). Hybridization with C-Delta-1 sense probe does not reveal silver grains above background (C). Antisense probe corresponding to C-Notch-1 cDNA localizes C-Notch-1 transcripts above background in the neural retina (D). Note that the C-Notch-1 transcripts are predominantly localized near the ventricular surface (arrow). Hybridization with C-Notch-1 sense probe does not reveal silver grains above background (E). Immunohistochemical analysis using Notch-1 antibody (20F) by indirect immunofluorescence reveals Notch-1 immunoreactivity near the ventricular zone (arrow) (F). Bars, 50 mm.

earlier observations of Notch-1 distribution (Coffman et al., 1990; Weinmaster et al., 1991; Reaume et al., 1992; Ahmad et al., 1995; Myat et al., 1996), C-Notch-1 transcripts were largely localized to the region corresponding to the ventricular zone in the neural retina. This was further ascertained by studying the distribution of CNotch-1 immunoreactivity using an antibody raised against human NOTCH-1 which has shown cross-reactivity across species (Ahmad et al., 1995; Chenn and McConnel, 1995). In E4 retina Notch-1 immunoreactivity, like C-Notch-1 transcripts in E3.5 retina, was predominantly

The temporal and spatial pattern of C-Delta-1 and CNotch-1 expression suggested that Delta–Notch signaling is involved in the specification of retinal neurons such as the RGCs. To test this hypothesis we first analyzed the expression of RA4, an antigen which is an early marker for RGC differentiation in chick retina and is expressed by RGCs soon after they become postmitotic (McLoon and Barnes, 1989). The specificity of this antigen as a marker for RGC differentiation has been confirmed by temporal and spatial analyses of RA4 expression (McLoon and Barnes, 1989), analysis of RGC generation by BrdU incorporation and RA4 immunohistochemistry (Waid and McLoon, 1995), and by analysis of RA4 specificity by other antibodies that recognize RGCs (Austin et al., 1995). Immunohistochemical analyses showed that RA4 antigen is detected in the retina as early as at E3 (data not shown). At E3.5 the few RA4-positive RGCs that were detected were localized in the center of the retina (Fig. 4A). Approximately 12 hr later at E4 a large population of RA4-positive RGCs was observed predominantly in the central retina (Fig. 4B). The pattern of distribution of RA4-positive ganglion cells changed in later stages. At E6 a decrease in RA4 immunoreactivity was observed in the central retina, whereas immunoreactivity was accentuated in the periphery (Fig. 4C). In the central retina RA4 immunoreactivity was confined to the optic fiber layer as previously observed (McLoon and Barnes, 1989). The appearance of RA4 immunoreactivity in the central retina and its progressive localization in the peripheral retina correlates with the central to peripheral gradient in the generation of RGCs (Prada et al., 1991). The change in the spatial distribution of RA4 immunoreactivity was accompanied by a change in the relative number of RA4-positive RGCs (Fig. 5) such that the proportion of RA4-expressing ganglion cells was higher in E6 than in E4 retina (42.6 { 1.6% vs 18.9 { 2%, P õ 0.005). A similar temporal increase in the relative number of RGCs has been observed by thymidine birthdate analysis in the developing chick retina (Prada et al., 1991).

Decrease in Delta-1 Expression Increases the Relative Number of RA4-Positive Ganglion Cells To evaluate the effect of Delta–Notch signaling on RA4 immunoreactivity we treated E4 retinal explant cultures with an antisense oligonucleotide to decrease C-Delta-1 expression. The sequence of the antisense oligonucleotide cor-

Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.

AID

DB 8546

/

6x22$$$$$1

04-09-97 20:11:01

dba

97

Delta–Notch Signaling in the Retina

FIG. 4. Spatial expression of RA4 immunoreactivity in the embryonic retina. Immunohistochemical analyses was carried out using RA4 antibody on 15-mm sections obtained from fresh-frozen retina. Indirect immunofluorescence shows the development of central to peripheral gradient in the appearance of RA4 immunoreactivity in E3.5 (A), E4 (B), and E6 (C) retina. Arrow, central retina; arrowhead, peripheral retina. Bars (A and B), 50 mm; C, 100 mm.

responded to the DSL domain of the C-Delta-1 transcript (Henrique et al., 1995). Incubation of the explant cultures with C-Delta-1 antisense oligonucleotide (50 mM) for 24 hr doubled the proportion of RA4-positive cells compared to those in untreated controls (43.2 { 1% vs 19.7 { 1.2%, P õ 0.005) (Fig. 6A). No significant change in the relative number of RA4-positive cells was observed when sense and missense oligonucleotides were included in the explant cultures or when we used an antisense oligonucleotide to decrease b-globin expression (Represa et al., 1991). Also, the numbers of cells as well as the viability of cells in treated and untreated groups were comparable (data not shown). Both in treated and untreated groups RA4 positive RGCs

FIG. 5. Accumulation of RA4-positive cells in the embryonic retina. Cells dissociated from E4 and E6 retina were analyzed for RA4 immunoreactivity and counted. E6 retina harbors more RA4-positive cells than E4 retina.

were counted in the explants obtained from entire retinae. This was to avoid any regional selectivity (peripheral vs central) to ensure that the number of RA4-positive cells is not influenced by sampling error due to the gradient of spatial distribution of RA4-positive RGCs. We hypothesized that if the effect of C-Delta-1 antisense was due to the attenuation of Delta-Notch signaling then a decrease in C-Notch-1 should have a similar effect. We tested this possibility by decreasing C-Notch-1 using a CNotch-1 antisense oligonucleotide (Weinmaster et al., 1991; Austin et al., 1995) in the explant cultures. Treatment of the E4 retinal explant culture with Notch-1 antisense oligonucleotide (50 mM) had essentially the same effect as the C-Delta-1 antisense oligonucleotide. The number of RA4positive cells increased twofold above untreated controls (40.5 { 0.9% vs 19.7 { 1.2%, P õ 0.005) (Fig. 6B). However, this increase in RA4 immunoreactivity was lower than that observed by Austin et al. (1995) and the difference could be due to different culture conditions. No significant change in the relative number of RA4-positive cells was observed when explants were treated with Notch-1 sense and missense oligonucleotides or with a b-globin antisense oligonucleotide. The specificity of the antisense treatment was further ascertained by analyzing the levels of C-Delta-1 and CNotch-1 transcripts by PCR amplification in response to antisense oligonucleotides (Fig. 7). The amplification primers amplified regions in C-Delta-1 and C-Notch-1 transcripts that did not include sequences corresponding to the respective antisense oligonucleotides. The levels of CDelta-1 and C-Notch-1 transcripts decreased relative to sense and b-globin-treated groups when treated with CDelta-1 and C-Notch-1 antisense oligonucleotide, respec-

Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.

AID

DB 8546

/

6x22$$$$$1

04-09-97 20:11:01

dba

98

Ahmad, Dooley, and Polk

FIG. 6. Effect of antisense oligonucleotide on the relative number of RA4-positive ganglion cells. E4 retinal explants were cultured for 24 hr in 50 mM oligonucleotide, followed by dissociation and immunocytochemical analysis and cell counting. The percentage of RA4-positive cells increased in cultures treated with Delta-1 (A) and Notch-1 (B) antisense oligonucleotides [1] in comparison with those treated with respective sense [2], missense [3], and b-globin antisense [4] oligonucleotides or untreated controls [5]. Means and SEM were collected from three separate experiments with a minimum of four different fields counted/treatment.

central to peripheral distribution of RA4-positive RGCs. In the retina treated with the C-Delta-1 antisense oligonucleotide we observed a stretch of retina which had a relatively unlabeled region (corresponding to the central retina) flanked by regions with robust expression of RA4 antigen (Fig. 8B). This spatial pattern of distribution of RA4 immunoreactivity is analogous to that observed in E6 retina in vivo, where RA4 immunoreactivity is found predominantly localized in the periphery (Fig. 4C). In the untreated group, we observed a region of the retina labeled for RA4 immunoreactivity flanked by unlabeled regions (Fig. 8D). The pattern of RA4 immunoreactivity was analogous to that observed in E4 retina in vivo (Fig. 4B). Between E4 and E6 the gradient of RA4 immunoreactivity changes such that RA4 immunoreactivity is observed largely in the peripheral retina (Fig. 4). Concomitant with this change in the spatial expression of RA4, the relative number of RA4-expressing RGCs doubles in E6 retina in comparison to that in E4 (Fig. 5). This increase in RA4positive RGCs is likely due to the fact that more RGCs are born in the periphery than in the central retina as development progresses (Prada et al., 1991). The perturbation of Delta–Notch signaling seems to accelerate the developmental process such that the changes in the temporal and spatial distribution of RA4 positive RGCs observed in vivo occurs earlier in vitro in response to antisense treatment. Thus E4 retina, following treatment with C-Delta-1 antisense oligonucleotide for 24 hr, began to resemble E6 retina from the viewpoint of RA4 immunoreactivity. Second, we tested the hypothesis that if a decrease in

tively. However, the decrease in levels of C-Notch-1 transcript was not as dramatic as the decrease observed in the levels of C-Delta-1 transcript in response to antisense treatment. The decrease observed in levels of C-Delta-1 and CNotch-1 transcript is consistent with previous findings that antisense treatment may lead to degradation of specific transcripts which is in turn responsible for the biological effects of antisense oligonucleotides (Paules et al., 1989; O’Keefe et al., 1989; Rosner et al., 1991; Brysch and Schlingensiepen, 1994; Austin et al., 1995).

Change in Delta-1 Expression Alters the Spatial Distribution of RA4-Positive Cells in the Peripheral Retina The spatial distribution of RA4-positive RGCs in response to changes in C-Delta-1 expression was studied in two ways. First, we carried out immunocytochemical analysis of retina treated with C-Delta-1 antisense oligonucleotides. Hemisected eye cups were cultured to ascertain the

FIG. 7. Effect of antisense oligonucleotides on C-Delta-1 and CNotch-1 transcripts. E4 retinal explants were cultured for 24 hr in 50 mM oligonucleotides followed by RNA extraction, cDNA synthesis, and PCR amplification using Delta-1- and Notch-1-specific primers. The PCR products were analyzed by Southern analyses using probes corresponding to Delta-1 and Notch-1 cDNA. b-Actin transcript was amplified as a measure of constitutive expression. Explant cultures were treated with Notch antisense (A), Notch-1 sense (B), Delta-1 antisense (C), Delta sense (D), or b-globin antisense (E) oligonucleotides.

Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.

AID

DB 8546

/

6x22$$$$$1

04-09-97 20:11:01

dba

99

Delta–Notch Signaling in the Retina

FIG. 8. Effect of antisense oligonucleotide on the spatial distribution of RA4 immunoreactivity. Hemisected E4 eye cups were cultured with and without Delta-1 antisense oligonucleotide followed by cryosectioning and immunocytochemistry using RA4 antibody. Indirect immunofluorescence on sections obtained from those treated with Delta-1 antisense oligonucleotide shows RA4 immunoreactivity largely distributed in the peripheral region (B), whereas untreated explant culture shows RA4 immunoreactivity largely distributed in the central region (D). A and C are Nomarski images. Arrow, central retina; arrowhead, peripheral retina. Scale bars, 100 mm.

Delta–Notch signaling increased the relative number of RA4-positive RGCs, then activation of this signaling pathway early in development should prevent the specification of RGCs and therefore compromise the spatial expression of RA4 antigen. We incubated E3.5 eye cups for 24 hr in the presence of Drosophila S2 cells expressing Drosophila Delta (40). The control consisted of culturing eye cups in excess of S2 cells that do not express Delta. At the end of the culture the retina should be comparable to that in E4.5 embryo in vivo. The conservation of Delta function across species has been shown by the ability Drosophila Delta to bind specifically with Xenopus Notch (Rebay et al., 1991) and influence RGC differentiation (Austin et al., 1995). The analysis of S2-Delta-treated retina showed that RA4 immunoreactivity remained confined to the central retina and few, if any, RA4-positive cells could be observed in the periphery (Fig. 9B). Even in the central retina few cells were found to be positive for RA4 immunoreactivity. This is similar to the pattern of RA4 immunoreactivity that is observed in E3.5 retina in vivo (Fig. 4A). In contrast, retinal sections obtained from eye cups incubated with control S2 cells showed RA4 immunoreactivity analogous to that observed in the retina of E4 embryo (Fig. 9D).

DISCUSSION Neurogenesis in the chick retina follows a temporal pattern which occurs over an extended period (Fig. 1C). The

continuing expression of C-Delta-1 and C-Notch-1 in the developing retina is similar to that observed by Myat et al. (1996) in the developing chick brain vesicles where these genes are expressed for the duration of neurogenesis. In tissues like retina where fates of different cell types are specified over an extended period, Delta–Notch signaling offers a regulatory mechanism to maintain a population of uncommitted precursors, possibly by lateral inhibition (Chitnis, 1995; Lewis, 1996; Kopan and Turner, 1996; Dorsky et al., 1997). Therefore the utilization of Delta–Notch signaling may depend on the successive rate of neurogenesis. Thymidine birthdating analysis shows that the overall rate of neurogenesis is highest during E6 (Prada et al., 1991). This is the stage during which the majority of RGCs, horizontal cells, amacrine cells, and photoreceptor cells become postmitotic. Most of these neuronal cell types are terminally differentiated by E8. Therefore the extent of neurogenesis decreases between E8 and E12. The increase in the levels of expression of Delta-1 and Notch-1 at E4 and E7.5 coincides with the period of neurogenesis when the majority of neurons are born and this may reflect the tissue’s increased need for Delta–Notch signaling to sort out competent cells from equivalent precursors. The decrease observed in the levels of Delta-1 and Notch-1 expression at E10 corresponds to a time when neurogenesis is declining and when Delta– Notch signaling is involved in regulating the competence of precursors that give rise to bipolar and Mu¨ller cells only (Fig. 1C). As the number of possible cell fates decreases for

Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.

AID

DB 8546

/

6x22$$$$$1

04-09-97 20:11:01

dba

100

Ahmad, Dooley, and Polk

FIG. 9. Effect of Drosophila Delta on the spatial distribution of RA4 immunoreactivity. Optic cup obtained from E3.5 embryos were cultured for 24 hr in excess of S2 cells or S2 cells expressing Drosophila Delta-1, followed by cryosectioning and immunocytochemistry using RA4 antibody. Indirect immunofluorescence shows a decrease in RA4 immunoreactivity when retinae were exposed to S2 cells expressing Delta (B) compared with retinae that were exposed to S2 cells without Delta expression construct (D). A and C are Nomarski images. Arrow, central retina; arrowhead, peripheral retina. Scale bars, 50 mm.

a precursor during development, utilization of the Delta– Notch pathway is restricted, and therefore the amount of transcript required may decrease as well. The temporal correlation of Delta-1 and Notch-1 expression with birthdating suggests that Delta–Notch signaling is utilized to sort out competent cells during successive stages of neurogenesis in the chick retina. This notion is supported by the observation that a decrease in Delta–Notch signaling prematurely increases the relative number of RGCs and therefore influences the temporal specification of RGCs. In the retina neurogenesis follows a spatial pattern. Autoradiographic studies have shown that the central to peripheral gradient in cell differentiation holds true for most types of retinal neurons (Prada et al., 1991). The first postmitotic RGC is detected near the posterior pole of the retina and the wave of ganglion cell differentiation spreads gradually from the center and reaches the peripheral boundary by E8/ 9 (5). This central to peripheral gradient in ganglion cell differentiation is correlated with the expression pattern of the RGC-specific marker RA4. Based on the spatial distribution of C-Delta-1 and C-Notch-1 expression, we feel that the spatial aspect of RA4 is regulated by an inverse gradient of Delta/Notch activity. This notion is supported by the observation that a decrease in C-Delta-1/C-Notch-1 expression in E4 retina not only leads to an increase in the relative number of RA4-positive cells but also these cells are recruited predominantly in the periphery, which is characteristic of the number and distribution observed in E6 retina in vivo. Since RA4 antigen is expressed by RGCs soon after they become postmitotic and RA4-positive cells have morphological and biochemical properties of retinal ganglion cells (McLoon and Barnes, 1989), it is likely that attenuation of Delta–Notch signaling promotes specification of ganglion cells from retinal precursors. We therefore propose that inhibition of the Delta–Notch signaling pathway causes precursors in the peripheral retina to prematurely

acquire retinal ganglion cell phenotype and express RA4 antigen (Fig. 10). A corollary to this observation is that accentuating Delta–Notch signaling is likely to keep the precursors uncommitted. The confinement of RA4 immunoreactivity to a few cells in the explant culture exposed to Drosophila Delta suggests that the precursors that would have expressed RA4 antigen failed to do so most likely due to the enhancement of Delta–Notch signaling. Another important aspect of our study is the observation of spatial distribution of C-Delta-1 and C-Notch-1 transcripts toward the vitreal and ventricular surfaces, respectively. Like other regions in the developing CNS the ven-

FIG. 10. Schematic representation of Delta–Notch signaling and differentiation of ganglion cells in retina. At E3.5 few RA4-positive ganglion cells are found in the center of the retina whereas cells expressing C-Notch-1 and C-Delta-1 are expressed in a gradient that increases from center to periphery. More RA4-positive cells are generated in stages E4, E4.5, and E6 as Delta–Notch signaling decreases (shown by lighter circles) from the center to the periphery presumably due to a gradient decrease in C-Notch-1 and C-Delta-1 expression. Treatment of E4 retina with Notch-1/Delta-1 antisense oligonucleotides causes a premature attenuation of Delta–Notch signaling resulting in an increase in the relative number of RA4 positive cells and their distribution similar to that observed in E6 retina. m, retinal ganglion cells with RA4 immunoreactivity in soma; l, Delta–Notch signaling.

Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.

AID

DB 8546

/

6x22$$$$$1

04-09-97 20:11:01

dba

101

Delta–Notch Signaling in the Retina

tricular surface in the embryonic retina represents the principle site of cell division (Robinson, 1991). Committed cells withdraw from the cell-cycle and begin to migrate away from the ventricular surface toward the vitreal surface where they settle and undergo terminal differentiation. Consistent with the fact that the activation of the Notch pathway suppresses differentiation (Coffman et al., 1993; Dorsky, 1995; Austin et al., 1995), C-Notch-1 is expressed in ventricular zone which mostly harbors dividing progenitors. In contrast, C-Delta-1 transcripts are localized in cells situated largely toward the vitreal surface. This differential distribution of C-Notch-1 and C-Delta-1 transcripts is similar to that observed in the developing chick hind brain and spinal cord (Myat et al., 1996) and suggests a very early distinction of two population of cells in the developing retina; those that are expressing C-Notch-1 are proliferating progenitors and those expressing C-Delta-1 represent a population of cells which are either proliferating but poised to exit mitosis (precursors) or those which have recently withdrawn from the cell cycle (nascent neurons). Our study cannot distinguish between the two possibilities for the CDelta-1-expressing cells but evidence for both does exist. First, Henrique et al. (1995), using the double-staining technique to identify BrdU incorporation and C-Delta-1 transcripts, have shown that both C-Delta-1-expressing precursors (C-Delta-1/BrdU/) and nascent neurons (C-Delta1/BrdU0) coexist in the developing neural tube, albeit in different proportions. Second, Dorsky et al. (1997) have shown that Delta-1-expressing cells in Xenopus retina are predisposed toward adopting differentiated phenotypes as they are released from lateral inhibition due to the lack of Notch-activating signal (see below). The spatial distribution of C-Notch-1- and C-Delta-1-expressing cells which are in different stages of differentiation may facilitate lateral inhibition to maintain a population of uncommitted retinal progenitors whose competence can be regulated by subtle fluctuations in Delta–Notch signaling. This is similar to mechanism proposed by Myat et al. (1996) in the developing brain. Cells that express C-Delta-1 are precursors poised to exit mitosis and/or nascent neurons. As they migrate out of the ventricular zone they interact with proliferating cells which express C-Notch-1. This interaction delivers the lateral inhibition and keeps the proliferating cell from differentiating prematurely. However, according to the feedback loop mechanism proposed in Drosophila (Heitzler et al., 1996) and Xenopus (Dorsky et al., 1997) the activation of Notch signaling is likely to reduce the expression of CDelta-1 in proliferating cells, thereby reducing the Notchactivating signal on the neighboring cells which are expressing C-Delta-1. The consequence would be that C-Delta-1expressing cells will be released from lateral inhibition and become competent to differentiate. Therefore a perturbation in the Delta–Notch signaling, either by misexpression of Delta-1 (Dorsky et al., 1997) or by decreasing C-Delta-1/ C-Notch-1 expression as carried out in our study will result in premature differentiation of precursors. Based on the results presented it can be suggested that

successive C-Delta-1 activation of the Notch receptor is required in order for retinal precursors to follow a specific fate both temporally and spatially. This is especially critical when numerous cell types are being generated over an extended period. The fact that C-Delta-1 and C-Notch-1 are expressed in spatiotemporal concert not only suggests the involvement of Notch signaling in sorting out fates of retinal neurons in general but also implies that C-Delta-1 may be the dominant ligand in this signaling pathway. The latter suggestion is supported by the finding that Serrate, another ligand of Notch identified in chick, has not been detected in the developing retina and therefore may not contribute to retinal neurogenesis in this animal model (Myat et al., 1996; Ahmad et al., 1996). While we limited this report to RGC differentiation, it will be necessary to examine the effects of perturbing Delta–Notch signaling on the remaining retinal cell types.

ACKNOWLEDGMENTS We thank Steve McLoon for RA4 antibody, David Ish-Horowicz for cDelta-1 and cNotch-1 cDNA, Spyros Artavanis-Tsakonas for S2 cells and Notch-1 antibody, Wally Thoreson and Harsha Acharya for critical reading of the manuscript, Huai Yun Han for technical assistance, and Walter Williams and Bill Wassom for photography and graphics. This work was supported by NEI (R2910313).

REFERENCES Ahmad, I. (1995). Mash-1 is expressed during rod photoreceptor differentiation and binds an E-box element, Eopsin-1 in the rat opsin gene. Dev. Brain Res. 90, 184–189. Ahmad, I., Leinders-Zufall, T., Kocsis, J. D., Shepherd, G. M., Zufall, F., and Barnstable, C. J. (1994). Retinal ganglion cells express a cGMP-gated cation conductance activatable by nitric oxide donors. Neuron 12, 155–165. Ahmad, I., Polk, D., and Dooley, C. (1996). Expression of the putative Notch-1 ligands, Delta-1 and Jagged in the vertebrate retina. Invest. Ophthalmol. Visual Sci. Res. 375S, 200. Ahmad, I., Zagouras, P., and Artavanis-Tsakonas, S. (1995). Involvement of Notch-1 in mammalian retinal neurogenesis: Association of Notch-1 activity with both immature and terminally differentiated cells. Mech. Dev. 53, 73–85. Artavanis-Tsakonas, S., Delidakis, C., and Fehon, R. G. (1991). The Notch locus and cell biology of neuroblast segregation. Annu. Rev. Cell Biol. 7, 427–452. Artavanis-Tsakonas, S., Matsuno, K., and Fortini, M. (1995). Notch signalling. Science 268, 225–232. Austin, C. P., Feldman, D. E., Ida, J. A., and Cepko, C. L. (1995). Vertebrate retinal ganglion cells are selected from competent progenitors by the action of Notch. Development 121, 3637–3650. Brysch, W., and Schlingensiepen, K. (1994). Design and application of antisense oligonucleotides in cell culture, in vivo and as therapeutic agents. Cell. Mol. Biol. 14, 557–568. Cagan, R. L., and Ready, D. F. (1989). Notch is required for successive cell decisions in the developing Drosophila retina. Genes Dev. 3, 1099–1112. Chen, A., and McConnel, S. K. (1995). Cleavage orientation and the

Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.

AID

DB 8546

/

6x22$$$$$1

04-09-97 20:11:01

dba

102

Ahmad, Dooley, and Polk

asymmetric inheritance of Notch-1 immunoreactivity in mammalian neurogenesis. Cell 82, 631–641. Chitnis, A., Henrique, D., Lewis, J., Ish-Horowicz, D., and Kitner, C. (1995). primary neurogenesis in Xenopus embryos regulated by a homologue of the Drosophila neurogenic gene Delta. Nature 375, 761–766. Chitnis, A. J. (1995). The role of Notch in lateral inhibition and cell fate specification. Mol. Cell. Biol. 6, 311–321. Coffman, C., Harris, W., and Kitner, C. R. (1990). Xotch, the xenopus homolog of Drosophila Notch. Science 249, 1438–1441. Coffman, C., Skogland, P., Harris, W. A., and Kitner, C. R. (1993). Expression of an extracellular deletion of Xotch diverts cell fate in the Xenopus embryos. Cell 73, 659–671. Couso, J. P., Knust, E., and Martinez Arias, S. (1995). Serrate and Wingless cooperate to induce vestigial gene expression and wing formation in Drosophila. Curr. Biol. 5, 1437–1448. Diaz Benjumea, F. J., and Cohen, S. M. (1995). Serrate signal through Notch to establish a wingless-dependent organizer at the dorsal/ventral compartment boundary of the Drosophila wing. Development 121, 4215–4225. Dorsky, R. I., Chang, S. W., Rapaport, D. H., and Harris, W. A. (1997). Regulation of neuronal diversity in the Xenopus retina by Delta signalling. Nature 385, 67–70. Dorsky, R. I., Rapaport, D. H., and Harris, W. A. (1995). Xotch inhibits cell differentiation in the Xenopus retina. Neuron 14, 487– 496. Fehon, R. G., Kooh, P. J., Rebay, I., Regan, C. L., Xu, T., Muskavitch, M. A. T., and Artavanis-Tsakonas, S. (1990). Molecular interactions between the protein products of the neurogenic loci Notch and Delta, two EGF-homologous genes in Drosophila. Cell 61, 523–534. Fitzgerald, K., and Greenwald, I. (1995). Interchangeability of Caenorhabditis elegans DSL proteins and intrinsic signalling activity of their extracellular domain in vivo. Development 121, 4275–4282. Fleming, R. J., Scottgale, T. M., Diederick, R. J., and Artavanis-Tsakonas, S. (1990). The gene Serrate encodes a putative EGF-like transmembrane protein essential for proper ectodermal development. Genes Dev. 4, 2188–2201. Fortini, M. E., Rebay, I., and Artavanis-Tsakonas, S. (1993). An activated Notch receptor blocks cell-fate commitment in the developing Drosophila eye. Nature 365, 555–567. Gao, D., and Kimble, J. (1995). APX-1 can substitute for its homolog lag2 to direct cell interactions throughout Caenorhabditis elegans development. Proc. Natl. Acad. Sci. USA 92, 9839–9842. Greenwald, I., and Rubin, G. M. (1992). Making a difference: The role of cell–cell interactions in establishing separate identities for equivalent cells. Cell 68, 271–281. Grillo, M., and Margolis, F. L. (1991). Use of reverse transcriptase polymerase chain reaction to monitor expression of intronless genes. Biotechniques 9, 262–268. Gu, Y., Hukriede, N. A., and Fleming, R. J. (1995). Serrate expression can functionally replace Delta activity during neurobalst segregation in the Drosophila embryo. Development 121, 535– 544. Hamburger, V., and Hamilton, H. L. (1951). A series of normal stages in the development of the chick embryo. J. Morphol. 88, 49–92. Harman, A. M., and Beazley, L. D. (1989). Generation of retinal cells in the Wallaby Setonix brachyurus (quokka). Neuroscience 28, 219–232. Heitzler, P., Bourouis, M., Ruel, L., Carteret, C., and Simpson, P.

(1996). Genes of the Enhancer of Split and Achaete–Scute complexes are required for a regulatory loop between Notch and Delta during lateral signalling in Drosophila. Development 122, 161– 171. Heitzler, P., and Simpson, P. (1991). The choice of cell fate in the epidermis of Drosophila. Cell 64, 1083–1092. Henderson, S. T., Gao, D., Lambie, E. J., and Kimble, J. (1994). Lag2 may encode a signaling ligand for GLP-1 and LIN-12 receptors of C. elegans. Development 120, 2913–2924. Henrique, D., Adam, J., Myat, A., Chitnis, A., Lewis, J., and IshHorowicz, D. (1995). Expression of Delta homologue in prospective neurons in chick. Nature 375, 787–790. Holt, C. E., Bretsch, T. W., Ellis, H. M., and Harris, W. (1988). Cellular determination in the Xenopus retina is independent of lineage and birthdate. Neuron 1, 15–26. Kahn, A. J. (1974). An audioradiographic analysis of the time of appearance of neurons in the developing chick neural retina. Dev. Biol. 38, 30–40. Kopan, R., and Nye, J. S. (1995). Vertebrate ligands for Notch. Curr. Biol. 5, 966–969. Kopan, R., and Turner, D. L. (1996). The Notch pathway: Democracy and aristocracy in the selection of cell fate. Curr. Opin. Neurobiol. 6, 594–601. Kopczynski, C. C., Alton, A. K., Fechtel, K., Kooh, P. J., and Muskavitvh, M. A. T. (1988). Delta, a Drosophila neurogenic gene, is transcriptionally complex and encodes a protein related to blood coagulation factors and epidermal growth factor of vertebrates. Genes Dev. 2, 1723–1735. La Vail, M. M., Rapaport, D. H., and Rakic, P. (1991). Cytogenesis in the monkey retina. J. Comp. Neurol. 309, 86–114. Lewis, J. (1996). Neurogenic genes and vertebrate neurogenesis. Curr. Opin. Biol. 6, 3–10. Lindsell, C. E., Shawber, C. J., Boulter, J., and Weinmaster, G. (1995). Jagged: A mammalian ligand that activates Notch-1. Cell 80, 909–917. McLoon, S. C., and Barnes, R. B. (1989). Early differentiation of retinal ganglion cells: An axonal protein expressed by premigratory and migrating retinal ganglion cells. J. Neurosci. 9, 1424–1432. Muskavitch, M. A. T. (1994). Delta-Notch signaling and Drosophila cell fate choice. Dev. Biol. 166, 415–430. Myat, A., Henrique, D., Ish-Horowicz, D., and Lewis, J. (1996). A chick homologue of Serrate and its relationship with Notch and Delta homologues during central neurogenesis. Dev. Biol. 174, 233–247. O’Keefe, S. J., Wolfe, H., Kiessling, A. A., and Cooper, G. M. (1989). Microinjection of antisense c-mos oligonucleotide prevents meiosis II in the maturing mouse egg. Proc. Natl. Acad Sci. USA 86, 7038–7042. Paules, R. S., Buccione, R., Moschel, R. C., Vande Woude, G. F., and Kelly, T. J. (1989). Mouse mos protooncogene product is present and functions during oogenesis. Proc. Natl. Acad Sci. USA 86, 5395–5399. Prada, C. , Puga, J., Perez-Mendez, L., Lopez, R., and Ramirez, G. (1991). Spatial and temporal patterns of neurogenesis in the chick retina. Eur. J. Neurosci. 3, 559–569. Price, J., Turner, D., and Cepko, C. (1987). Lineage analysis in the vertebrate nervous system by retrovirus mediated gene transfer. Proc. Natl. Acad Sci. USA 84, 156–160. Reaume, A. G., Canlon, R. A., Zirgibl, R., Yamaguchi, T. P., and Rossant, J. (1992). Expression analysis of Notch homologin the mouse embryo. Dev. Biol. 154, 377–387. Rebay, I., Fleming, R. J., Fehon, R. G., Cherbas, L., Cherbas, P.,

Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.

AID

DB 8546

/

6x22$$$$$1

04-09-97 20:11:01

dba

103

Delta–Notch Signaling in the Retina

and Artavanis-Tsakonas (1991). Specific EGF-repeats of Notch mediate interactions with Delta and Serrate: Implications of Notch as a multifunctional receptor. Cell 67, 687–699. Reh, T. A., and Tully, T. (1986). Regulation of tyrosine-hydroxylase-containing amacrine cell number in larval frog retina. Dev. Biol. 114, 463–469. Represa, J., Leon, Y., and Miner, C. (1991). The int-2 proto-oncogene is responsible for the induction of the inner ear. Nature 353, 561– 563. Robinson, S. R. (1991). Development of the mammalian retina. In ‘‘Neuroanatomy of Visual Pathways and Their Development’’ (B. Dreher and S. R. Robinson, Eds.), Vol. 3, pp. 69–128. Macmillan, UK. Rosner, M. H., Desanto, R. J., Arnheiter, H., and Staredt, L. M. (1991). Oct-3 is a maternal factor required for the first mouse embryonic division. Cell 64, 1103–1110. Sidman, R. L. (1961). Histogenesis of mouse retina studied with thymidine 3H. In ‘‘Structure of the Eye’’ (G. K. Smelser, Ed.), pp. 487–506. Academic Press, New York. Speicher, S. A., Thomas, U., Hinz, U., and Knust, E. (1994). The Serrate locus of Drosophila and its role in morphogenesis of the wing imaginal discs: Control of cell proliferation. Development 120, 535–544. Tax, F. E., Yeargers, J. J., and Thomas, J. H. (1994). Sequence of C. elegans lag2 reveals a cell-signalling domain shared with Delta and Serrate of Drosophila. Nature 368, 150–154. Thomas, U., Speicher, S. A., and Knust, E. (1991). The Drosophila

gene Serrate encodes an EGF-Like transmembrane protein with a complex expression pattern in embryos and wing disc. Development 111, 749–761. Turner, D. L., and Cepko, C. L. (1987). A common progenitor for neurons and glia persists in rat retina late in development. Nature 238, 131–136. Turner, D. L., Snyder, E. Y., and Cepko, C. L. (1990). Lineage-independent determination of cell type in the embryonic mouse retina. Neuron 4, 833–845. Vaessin, H., Bremer, K. A., Knust, E., and Campos-Ortega, J. A. (1987). The neurogenesis gene Delta of Drosophila melanogaster is expressed in neurogenic territories and encodes a putative transmembrane protein with EGF-like repeats. EMBO J. 6, 3430– 3440. Waid, D. K., and McLoon, S. C. (1995). Immediate differentiation of ganglion cells following mitosis in the developing retina. Neuron 14, 117–124. Weinmaster, G., Roberts, V. J., and Lemke, G. (1991). A homolog of Drosophila Notch expressed during mammalian development. Development 113, 199–205. Wetts, R., and Fraser, S. E. (1988). Multipotent precursors can give rise to all major cell types of the frog retina. Science 239, 1142– 1145. Young, R. W. (1985). Cell differentiation in the retina of the mouse. Anat. Rec. 212, 499–205. Received for publication December 19, 1996 Accepted February 20, 1997

Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.

AID

DB 8546

/

6x22$$$$$1

04-09-97 20:11:01

dba