Delta (Dl) function is required for proper specification of epidermal and neural lineages within the neurogenic ectoderm of Drosophila melanogaster. We have determined the spatial accumulation of five Dl transcripts that arise as the result of alternative RNA processing during embryogenesis. We find that these transcripts accumulate in all cells of the neurogenic ectoderm immediately preceding neuroblast segregation, indicating that transcription of Dl does not differ between presumptive neuroblasts and presumptive dermoblasts. Dl transcripts also accumulate transiently in mesodermal and endodermal cells, suggesting that Dl may function in developmental processes in addition to differentiation of the neurogenic ectoderm. We find that three of the Dl transcripts are localized to the base of the nucleus during cellularization. The apparent association of these three transcripts with polysomes suggests that they accumulate within the cytoplasm at the nuclear periphery and is consistent with the hypothesis that Dl encodes multiple translational products.

The central nervous system of Drosophila is derived from a subset of ectodermal cells located within the neurogenic regions of the embryo (Campos-Ortega and Hartenstein, 1985). Neurogenesis begins within these regions shortly after gastrulation when presumptive neuroblasts, progenitors of the nervous system, delaminate from the ectodermal epithelium and segregate into the interior of the embryo. The cells that remain within the epithelium after neuroblast segregation are dermoblasts, progenitors of the ventral and cephalic epidermis of the larva. Thus, the neurogenic ectoderm of Drosophila gives rise to both neural and epidermal lineages.

The zygotic function of six genes -Notch (A), Delta (Dl), Enhancer of split [E(spl)], neuralized (neu), mastermind (mam) and big brain (bib) -is known to be required to establish epidermal identity within neurogenic regions of the embryo (Poulson, 1937; Lehmann et al. 1983). Homozygosity for loss-of-function mutations in any one of these genes leads to hypertrophy of the central nervous system and reduction of the ventral and cephalic epidermis. Histological studies indicate that this phenotype results directly from the misrecruitment into the neural lineage of ectodermal cells that otherwise would enter the epidermal lineage (Lehmann et al. 1983).

Molecular and biochemical analyses of N have revealed that the Notch product is a transmembrane protein with an extracellular domain that contains 36 cysteine-rich, epidermal growth factor-like repeats (Wharton et al. 1985; Kidd et al. 1986). The structure of this polypeptide led to the hypothesis that N and other neurogenic genes may function in a system of cell-cell communication required for the proper differentiation of neurogenic ectoderm into neural and epidermal lineages (Wharton et al. 1985; Kidd et al. 1986). This hypothesis has received direct experimental support from embryonic cell transplantation experiments (Technau and Campos-Ortega, 1987). These experiments revealed that the neurogenic mutant phenotype of a cell lacking the function of N, Dl, mam, neu or bib can be rescued when the cell is transplanted among wild type ectodermal cells. Somatic mosaic analyses have further revealed that while single N mutant cells can be rescued by surrounding wild type cells, ‘patches’ of two or more cells are rarely rescued (P. Hoppe and R. Greenspan, in preparation), indicating that the nonautonomous function of N is locally restricted.

The subsequent molecular characterization of Dl and E(spl) has provided further support for the hypothesis that the neurogenic genes function in intercellular communication (Vassin et al. 1987; Kopczynski et al. 1988; Hartley et al. 1988). The predominant embryonic and maternal Dl transcripts encode a putative transmembrane protein (D1ZM) with an extracellular domain that contains nine cysteine-rich repeats homologous to the 36 EGF-like repeats present in the Notch polypeptide (Vassin et al. 1987; Kopczynski et al. 1988). Additional polypeptides distinct from D1ZM may be encoded by less abundant (minor) embryonic and maternal Dl transcripts (Kopczynski et al. 1988). A neurogenic gene within the E(spl) region has been identified recently (Hartley et al. 1988) that encodes a putative protein with homology to βtransducin, a protein that is involved in vertebrate phototransduction (reviewed in Gilman, 1987). These results reveal a potential for functional interactions among these products and the Notch product that is particularly interesting given the array of genetic interactions that has been described among N, Dl and E(spl) (Campos-Ortega et al. 1984; Dietrich and Campos-Ortega, 1984; Vassin et al. 1985; Alton et al. 1989; Shepard et al. 1989).

The accumulation of N and E(spl) transcripts in cell types other than epidermal and neural precursors has led to the suggestion that these genes may affect developmental processes in addition to differentiation of the neurogenic ectoderm (Hartley et al. 1987; Artavanis-Tsakonas, 1988; Hartley et al. 1988). We have determined the spatial accumulation patterns of the predominant and minor Dl transcripts during embryogenesis in order to address the question of whether Dl function is required solely for delineating epidermal and neural identities within the embryonic ectoderm. In contrast to a previous report (Vassin et al. 1987), we find that Dl transcripts are present in derivatives of all three germ layers of the embryo and are not restricted to neurogenic regions of the ectoderm. Thus, the spatial and temporal accumulation patterns that we observe suggest that Dl may act pleiotropically during embryogenesis. We also present evidence suggesting that the minor Dl transcripts are subcellularly localized within the cytoplasm to the nuclear periphery.

In situ hybridization

Embryo collection, fixation and paraffin embedding were performed as described in Ingham et al. (1985). Paraffin blocks containing embryos were cut into 5 μm sections that were collected on glass microscope slides coated with poly-D-lysine (50μgml−1 in 10mM-Tris-HCl pH8.0). Sections were dewaxed by two 10 min incubations in xylene, then rehydrated by incubation for 2 min each in 100 % EtOH, 95 % EtOH (v/v in H2O), 70% EtOH, 50% EtOH, 30% EtOH and PBS (130mM-NaCl, 7mM-Na2HPO4, 3mM-NaH2PO4). The slides were then treated with 0.25 % (v/v) acetic anhydride in 0.1 M-triethanolamine pH 8.0 for 10 min (Hayashi et al. 1978), rinsed in PBS, dehydrated through the above EtOH series and air dried. The subsequent hybridization, washing and autoradiography of the slides were performed as described in Ingham et al. (1985). After autoradiography, embryo sections were stained with Mayer’s hematoxylin (Clayden, 1971) for 1-4min and with Giemsa stain (Fisher) for 2 min, dehydrated through the above EtOH series, then incubated twice for 2 min in xylene before mounting in Permount (Fisher). Morphological features present in individual sections were identified, when necessary, on the basis of comparison with alternate Giemsa-stained sections.

Probes

Antisense RNA probes were generated using T3 or T7 RNA polymerase and 35S-UTP as described in Hartley et al. (1988). Each probe was labelled to a specific activity of 2×108ctsmin−1g−1 and used at a concentration of 0.3 ng μl−1 kb−1. Templates employed for the synthesis of probes (Fig. 1) included: a 1.5 kb Pstl/EcoRI DNA fragment corresponding to the 3′ end of cDNA insert D12 (Kopczynski et al. 1988) for 5.4Z and 4.5M, a 2.2 kb Dl genomic fragment (fragment F, ibid.) for 3.5z, a 0.4kb Dl genomic fragment (fragment H, ibid.) for 2.8z and a 2.5 kb Dl genomic fragment (fragment B, ibid.) for 5.4z. A 3 kb Ngenomic BglII fragment (ibid.) was used to generate a probe for N transcripts and a 118 bp Taql fragment from the 3′-untranslated region of the bet<z3-tubulin gene was used to generate a beta3-tubulin-specific probe (Kimble et al. 1989). A sense-strand RNA probe, synthesized using a 2.2 kb Dl genomic fragment (fragment L, Kopczynski et al. 1988) as a template, was used to determine levels of non-specific (‘background’) hybridization to sections (data not shown).

Fig. 1.

Transcript-specific probes used for in situ hybridization. TTie filled arrows above the restriction map represent the templates used to generate antisense-strand RNA probes that were employed to detect 5.4z, 3.5z, 2.8z and 5.4Z/4.5M transcripts, respectively (see MATERIALS AND METHODS). The hatched arrow represents the template for a sense-strand RNA probe used to assess background levels of hybridization. The open arrows below the restriction map represent a composite of the restriction fragments that hybridize to the specified Dl transcripts (Kopczynski et al. 1988). This diagram indicates that the restriction fragment used to generate the 2.8z-specific probe hybridizes to 5.4Z and 4.5M as well. However, RNA blot analysis with this fragment reveals that the 2.8z signal intensity obtained is approximately 20 times those observed for 5.4Z and 4.5M (data not shown). In addition, the in situ hybridization patterns obtained with this probe are identical to those obtained with probes specific for 5.4z or 3.5z (data not shown). We therefore conclude that this fragment can be employed to specifically detect the 2.8z transcript in in situ hybridization analyses. Numbers in parentheses are coordinates from a chromosomal walk that encompasses DI (ibid.) given in kilobasepairs of DNA. Only selected restriction sites are presented. (B), BamHI; (P), PstI; (R), EcoRI; (S), SalI; (Sc), SocI.

Fig. 1.

Transcript-specific probes used for in situ hybridization. TTie filled arrows above the restriction map represent the templates used to generate antisense-strand RNA probes that were employed to detect 5.4z, 3.5z, 2.8z and 5.4Z/4.5M transcripts, respectively (see MATERIALS AND METHODS). The hatched arrow represents the template for a sense-strand RNA probe used to assess background levels of hybridization. The open arrows below the restriction map represent a composite of the restriction fragments that hybridize to the specified Dl transcripts (Kopczynski et al. 1988). This diagram indicates that the restriction fragment used to generate the 2.8z-specific probe hybridizes to 5.4Z and 4.5M as well. However, RNA blot analysis with this fragment reveals that the 2.8z signal intensity obtained is approximately 20 times those observed for 5.4Z and 4.5M (data not shown). In addition, the in situ hybridization patterns obtained with this probe are identical to those obtained with probes specific for 5.4z or 3.5z (data not shown). We therefore conclude that this fragment can be employed to specifically detect the 2.8z transcript in in situ hybridization analyses. Numbers in parentheses are coordinates from a chromosomal walk that encompasses DI (ibid.) given in kilobasepairs of DNA. Only selected restriction sites are presented. (B), BamHI; (P), PstI; (R), EcoRI; (S), SalI; (Sc), SocI.

Polysomal RNA preparation

Fertilized eggs were collected for a 3h interval at 25°C, aged an additional 2h at 25°C, then harvested and dechorionated as described in Alton et al. (1988). The dechorionated embryos were blotted dry, then 2 g of embryos were horn ogenized in 15 ml ice-cold polysome buffer [40 mM-Tris-HCl pH8.5, 5mM-KCl, 10mM-MgCl2, 0.5% (v/v) diethylpyrocarbonate] using a Dounce homogenizer. The homogenate was centrifuged at 5000 revs min−1 for 2 min at 4 °C in a Beckman JS13 rotor, and 2 ml of the supernatant were layered over a 20-50% (w/v) sucrose gradient in Mg2+-sucrose buffer (20 mM-Tris-HCl pH 8.5, 50mM-KCl, 10mM-MeCl2, 1mgml−1 heparin) and centrifuged at 25 000 revs min−1 for 2h at 4°C in a Beckman SW27 rotor. The gradient was collected as 1.5 ml fractions through an ISCO UA-5 spectrophotometer. These fractions were subsequently pooled to give the five larger fractions indicated in Fig. 9. Total RNA was prepared from these fractions as described in Anderson and Lengyel (1981). The purified RNA pellets were each resuspended in 160μ1 of deionized, distilled H2O and stored at −70°C.

Fig. 9.

Dl transcripts cosediment with polysomes in sucrose gradients. (A) Fractionation of a 2–5 h embryo homogenate in a 20–50% sucrose gradient (in the presence of Mg2+). Fractions A-E correspond to large polysomes (A), median-sized polysomes (B), small polysomes (C), monosomes (D) and free ribosomal subunits (E) as inferred from the A26tl profile. Total RNA was purified from these fractions and equal volumes of each RNA sample were fractionated on denaturing agarose gels, transferred to nylon membranes and hybridized with transcript-specific probes. The RNA blots are presented below the A26O profile. The majority of each transcript is present in fractions A-C. (B) Fractionation of polysomes isolated from a 2–5 h embryo homogenate after incubation in 40mM-Na2EDTA. The A26o profile shows that the majority of the polysomal RNA sediments at a position in the 20-50% gradient expected for free ribosomal subunits (fraction E). Total RNA preparations from fractions A-E were analyzed by RNA blot analysis as in (A). The RNA blots below the A2WJ profile reveal that the majority of each transcript sediments in fractions D and E.

Fig. 9.

Dl transcripts cosediment with polysomes in sucrose gradients. (A) Fractionation of a 2–5 h embryo homogenate in a 20–50% sucrose gradient (in the presence of Mg2+). Fractions A-E correspond to large polysomes (A), median-sized polysomes (B), small polysomes (C), monosomes (D) and free ribosomal subunits (E) as inferred from the A26tl profile. Total RNA was purified from these fractions and equal volumes of each RNA sample were fractionated on denaturing agarose gels, transferred to nylon membranes and hybridized with transcript-specific probes. The RNA blots are presented below the A26O profile. The majority of each transcript is present in fractions A-C. (B) Fractionation of polysomes isolated from a 2–5 h embryo homogenate after incubation in 40mM-Na2EDTA. The A26o profile shows that the majority of the polysomal RNA sediments at a position in the 20-50% gradient expected for free ribosomal subunits (fraction E). Total RNA preparations from fractions A-E were analyzed by RNA blot analysis as in (A). The RNA blots below the A2WJ profile reveal that the majority of each transcript sediments in fractions D and E.

To achieve efficient EDTA-disruption of polysomes, 8 ml of homogenate were layered onto a 3 ml sucrose pad (1.5 M-sucrose in Mg2+-sucrose buffer) and centrifuged at 35 000 revs minfor 90 min at 4°C in a Beckman SW41 rotor. The polysome pellet was then resuspended in 8 ml of EDTA gradient buffer (20 mM-Tris-HCl pH8.5, 50mM-KCl, 40 mM-Na2EDTA, 1mgml−1 heparin) and 2ml of resuspended polysomes were layered onto a 20–50% (v/v) sucrose gradient in EDTA gradient buffer. This gradient was run in parallel with the Mg23−containing sucrose gradient and processed as described above.

RNA blot analysis

RNA samples were denatured with glyoxal, fractionated on agarose gels and transferred to nylon membranes as described in Kopczynski et al. (1988). 32P-labeling of the DNA probes and hybridization of the blots were performed as described in Feinberg and Vogelstein (1983). The transcript-specific probes used were Dl genomic fragments B (5.4z), F (3.5z), I (2.8z) and L (5.4Z/4.5M) (Fig. 1, Kopczynski et al. 1988).

The Dl transcripts

Dl encodes at least six distinct transcripts: 5.4Z, the predominant embryonic transcript; 5.4z, 3.5z and 2.8z, the minor embryonic transcripts; 4.5M, the predominant maternally-loaded transcript; and 3.6m, a minor maternally-loaded transcript (Kopczynski et al. 1988). The 4.5M transcript appears to encode the same polypeptide as 5.4Z and is probably expressed zygotically as well as maternally (ibid.).

We have assessed the spatial accumulation patterns of Dl transcripts using the probes that are depicted in Fig. 1. The hybridization patterns we present for the predominant transcripts, which were obtained with the probe that recognizes 5.4Z and 4.5M, are identical to those obtained with a probe specific for 5.4Z after the zygotic accumulation of Dl transcripts begins (data not shown). We present the spatial accumulation pattern of only one minor transcript, 3.5z, because we observe no difference between the patterns obtained with a 3.5z-specific probe and probes specific for 5.4z or 2.8z (Fig. 1, data not shown). The accumulation of 3.6m transcripts has not been assessed because we have not yet defined a probe specific for this transcript. The stages of embryonic development referred to below have been defined by Campos-Ortega and Hartenstein (1985).

The pattern through gastrulation

Maternally-loaded Dl transcripts are uniformly distributed throughout the embryo during the nine nuclear division cycles that precede syncytial blastoderm (Fig. 2A). The minor zygotic transcripts are not detectable above background [as assessed by hybridization with a sense strand probe (data not shown)] during this time (Fig. 2B). Zygotic accumulation of the predominant and minor transcripts begins at the start of cellularization (early stage 5) and is localized to the presumptive neurogenic regions of the embryo (Figs 2C, D, 3A, B). The transcripts are initially distributed in a dorsoventral gradient within the ventral neurogenic region (Fig. 3A, B), but become evenly distributed within this region as cellularization proceeds (Fig. 3C, D).

Fig. 2.

Distribution of Dl transcripts from presyncytial blastoderm through gastrulation. Parasagittal and sagittal sections of embryos hybridized with the 5.4Z/4.5M (left column) or 3.5z (right column) probes are shown. All sections are oriented dorsal side up and anterior to the left. (A, B) Alternate parasagittal sections through a stage 3 embryo (darkfield illumination). Maternally-loaded 4.5M transcripts are evenly distributed throughout the embryo; 3.5z transcripts are not detectable above background. (C, D) Parasagittal sections through mid-(C) and early (D, darkfield) stage 5 embryos. 5.4Z/4.5M and 3.5z transcripts are most prevalent within the procephalic (pNR) and ventral (vNR) neurogenic regions of the embryo. (E, F) Alternate sagittal sections through a stage 6 embryo. 5.4Z/4.5M and 3.5z transcripts are present within most cells of the embryo, including germ band mesodermal cells (ms) and cells in a region of the embryo that will give rise to the cephalic mesoderm and anterior midgut (arrowheads). Transcripts are not detectable in cells of the presumptive foregut (fg). (G, H) Alternate parasagittal sections through a stage 7 embryo. 5.4Z/4.5M and 3.5z transcripts are present in ectodermal, mesodermal and endodermal [anterior and posterior midgut (pm)] cells. Cells of the presumptive foregut remain unlabeled at this stage. The arrowheads point to labeled cells of the cephalic mesoderm and anterior midgut primordium. Exposure times: (A, B, D, G) 4 days; (C, E, F, H) 10 days.

Fig. 2.

Distribution of Dl transcripts from presyncytial blastoderm through gastrulation. Parasagittal and sagittal sections of embryos hybridized with the 5.4Z/4.5M (left column) or 3.5z (right column) probes are shown. All sections are oriented dorsal side up and anterior to the left. (A, B) Alternate parasagittal sections through a stage 3 embryo (darkfield illumination). Maternally-loaded 4.5M transcripts are evenly distributed throughout the embryo; 3.5z transcripts are not detectable above background. (C, D) Parasagittal sections through mid-(C) and early (D, darkfield) stage 5 embryos. 5.4Z/4.5M and 3.5z transcripts are most prevalent within the procephalic (pNR) and ventral (vNR) neurogenic regions of the embryo. (E, F) Alternate sagittal sections through a stage 6 embryo. 5.4Z/4.5M and 3.5z transcripts are present within most cells of the embryo, including germ band mesodermal cells (ms) and cells in a region of the embryo that will give rise to the cephalic mesoderm and anterior midgut (arrowheads). Transcripts are not detectable in cells of the presumptive foregut (fg). (G, H) Alternate parasagittal sections through a stage 7 embryo. 5.4Z/4.5M and 3.5z transcripts are present in ectodermal, mesodermal and endodermal [anterior and posterior midgut (pm)] cells. Cells of the presumptive foregut remain unlabeled at this stage. The arrowheads point to labeled cells of the cephalic mesoderm and anterior midgut primordium. Exposure times: (A, B, D, G) 4 days; (C, E, F, H) 10 days.

Fig. 3.

Distribution of Dl transcripts during cellularization. Transverse and parasagittal sections hybridized with the 5.4Z/4.5M (left column) or 3.5z (right column) probes are shown. (A, B) Alternate transverse sections through an early stage 5 embryo at approximately 50% egg length (100% egg length at anterior end). Arrowheads mark the approximate boundaries of the presumptive ventral neurogenic region (vNR). Labeled nuclei are most apparent in the ventral neurogenic region, indicating that the 5.4Z/ 4.5M and 3.5z embryonic transcripts are most prevalent within this region of the embryo. However, maternally-loaded 4.5M transcripts are present throughout the cortical cytoplasm at this stage. Note that the highest levels of 5.4Z/4.5M and 3.5z transcripts are present over the ventral-most nuclei of the ventral neurogenic region. (C, D) Alternate transverse sections through a late stage 5 embryo at approximately 40% egg length. The zygotic accumulation of 5.4Z/4.5M transcripts remains most apparent within the presumptive ventral neurogenic regions of the ectoderm, although some hybridization is evident over nuclei in regions representing the presumptive mesoderm (ms) and dorsal ectoderm (dec). The accumulation of 3.5z transcripts in the presumptive mesoderm and dorsal ectoderm is much more evident. 5.4Z/4.5M and 3.5z transcripts appear evenly distributed within the ventral neurogenic region at this stage. (E, F) Parasagittal sections through the anterior regions of two late stage 5 embryos. 5.4Z/4.5M transcripts are distributed over nuclei (nuc) and throughout the cortical cytoplasm at this stage; 3.5z transcripts are localized to the base of the blastoderm nuclei. Exposure times: (A, B) 4 days; (C-F) 10 days.

Fig. 3.

Distribution of Dl transcripts during cellularization. Transverse and parasagittal sections hybridized with the 5.4Z/4.5M (left column) or 3.5z (right column) probes are shown. (A, B) Alternate transverse sections through an early stage 5 embryo at approximately 50% egg length (100% egg length at anterior end). Arrowheads mark the approximate boundaries of the presumptive ventral neurogenic region (vNR). Labeled nuclei are most apparent in the ventral neurogenic region, indicating that the 5.4Z/ 4.5M and 3.5z embryonic transcripts are most prevalent within this region of the embryo. However, maternally-loaded 4.5M transcripts are present throughout the cortical cytoplasm at this stage. Note that the highest levels of 5.4Z/4.5M and 3.5z transcripts are present over the ventral-most nuclei of the ventral neurogenic region. (C, D) Alternate transverse sections through a late stage 5 embryo at approximately 40% egg length. The zygotic accumulation of 5.4Z/4.5M transcripts remains most apparent within the presumptive ventral neurogenic regions of the ectoderm, although some hybridization is evident over nuclei in regions representing the presumptive mesoderm (ms) and dorsal ectoderm (dec). The accumulation of 3.5z transcripts in the presumptive mesoderm and dorsal ectoderm is much more evident. 5.4Z/4.5M and 3.5z transcripts appear evenly distributed within the ventral neurogenic region at this stage. (E, F) Parasagittal sections through the anterior regions of two late stage 5 embryos. 5.4Z/4.5M transcripts are distributed over nuclei (nuc) and throughout the cortical cytoplasm at this stage; 3.5z transcripts are localized to the base of the blastoderm nuclei. Exposure times: (A, B) 4 days; (C-F) 10 days.

The predominant and minor transcripts accumulate in all regions of the ectoderm during gastrulation with the exception of the presumptive foregut (stages 6 and 7, Fig. 2E-H). Dl transcripts are also present in the anterior and posterior midgut invaginations and at low levels in the mesoderm (Fig. 2E-H). Thus, Dl transcripts are expressed in all three germ layers of the embryo at this time.

Two significant differences between the predominant and minor transcript accumulation patterns become apparent during cellularization. First, the minor transcripts are localized to the base of each nucleus, whereas the predominant transcripts are distributed throughout the cortical cytoplasm (Fig. 3E, F). This localization of minor transcripts could represent accumulation at a basal position within the nucleus or accumulation in the cytoplasm that surrounds the basal half of the nucleus. These possibilities cannot be distinguished by in situ hybridization (however, see below). Second, the accumulation of the minor transcripts appears to extend into the dorsal ectoderm and presumptive mesoderm significantly earlier than the zygotic accumulation of the predominant transcripts (Fig. 3C, D). However, it is possible that the zygotic accumulation of predominant transcripts begins at the same time within these regions, but is more difficult to detect due to the more diffuse cytoplasmic distribution of the predominant transcripts. The presence of the predominant maternally-loaded transcript throughout the cortical cytoplasm of the precellular embryo (Fig. 3A) may also obscure the initial accumulation of predominant transcripts outside of the neurogenic regions.

The pattern during germ band elongation and extended germ band

The spatial accumulation of the predominant and minor Dl transcripts from approximately three-and-one-half through seven hours of embryogenesis (stage 8 through stage 11) is presented in Fig. 4. The accumulation of Dl transcripts in endodermal derivatives (anterior and posterior midgut) remains relatively constant throughout this interval. However, the accumulation of Dl transcripts in the ectoderm and mesoderm changes continuously as these germ layers give rise to their respective differentiated tissues. We therefore present the temporal changes in Dl transcript accumulation patterns within the context of ectodermal and mesodermal differentiation during this interval.

Fig. 4.

Distribution of Dl transcripts from germ band elongation through extended germ band. Parasagittal sections of embryos hybridized with the 5.4Z/4.5M (left column) or 3.5z (right column) probes are shown. All sections are oriented as in Fig. 1. (A, B) Alternate sections through a stage 8 embryo. 5.4Z/4.5M and 3.5z transcripts are present throughout the ectoderm with the exception of the foregut (fg) and proctodeum primordia. 5.4Z/4.5M and 3.5z transcripts are also present in the dorsal wall of the posterior midgut primordium (pm) and at low levels in the mesoderm (ms). (C, D) Alternate sections through a stage 10 embryo. 5.4Z/4.5M and 3.5z transcripts are present at relatively high levels in both the mesodermal and peripheral ectodermal cell layers. Lower levels of transcripts are present between these cell layers where neuroblasts (nb) that will give rise to the ventral nerve cord have segregated. 5.4Z/4.5M and 3.5z transcripts are also present in the primordia of the anterior (am) and posterior midgut. The arrowheads point to unlabeled cells of the proctodeum. (E, F) Alternate sections of a mid-stage 11 embryo. This embryo was sectioned such that the bottom of the section represents a ventral region of the embryo and the top of the section a more lateral region of the embryo. 5.4Z/4.5M and 3.5z transcripts are less abundant in the neuroblast layer than in the peripheral ectoderm, anterior midgut and posterior midgut. 5.4Z/4.5M transcripts are apparent ventrally in metameric clusters of mesodermal cells and laterally in mesodermal cells located between the tracheal pits; 3.5z transcripts are observed in the same region of the mesoderm laterally, but do not appear to be restricted ventrally to metameric clusters of mesodermal cells in this section. (G, H) Alternate sections of a late stage 11 embryo (H, darkfield illumination). 5.4Z/4.5M and 3.5z transcript levels are reduced throughout the mesoderm and ventral ectoderm. Three ventrolateral clusters of ectodermal cells consisting of neuroblasts and underlying dermoblasts remain heavily labeled with the 5.4Z/4.5M probe. 5.4Z/4.5M and 3.5z transcripts are apparent in the anterior midgut, posterior midgut and dorsal wall of the proctodeum at this stage. Abbreviations: (pNR) procephalic neurogenic region; (st) stomodeum; (vNR) ventral neurogenic region. Exposure times: (A, B, E, F) 10 days; (C, D, G, H) 4 days.

Fig. 4.

Distribution of Dl transcripts from germ band elongation through extended germ band. Parasagittal sections of embryos hybridized with the 5.4Z/4.5M (left column) or 3.5z (right column) probes are shown. All sections are oriented as in Fig. 1. (A, B) Alternate sections through a stage 8 embryo. 5.4Z/4.5M and 3.5z transcripts are present throughout the ectoderm with the exception of the foregut (fg) and proctodeum primordia. 5.4Z/4.5M and 3.5z transcripts are also present in the dorsal wall of the posterior midgut primordium (pm) and at low levels in the mesoderm (ms). (C, D) Alternate sections through a stage 10 embryo. 5.4Z/4.5M and 3.5z transcripts are present at relatively high levels in both the mesodermal and peripheral ectodermal cell layers. Lower levels of transcripts are present between these cell layers where neuroblasts (nb) that will give rise to the ventral nerve cord have segregated. 5.4Z/4.5M and 3.5z transcripts are also present in the primordia of the anterior (am) and posterior midgut. The arrowheads point to unlabeled cells of the proctodeum. (E, F) Alternate sections of a mid-stage 11 embryo. This embryo was sectioned such that the bottom of the section represents a ventral region of the embryo and the top of the section a more lateral region of the embryo. 5.4Z/4.5M and 3.5z transcripts are less abundant in the neuroblast layer than in the peripheral ectoderm, anterior midgut and posterior midgut. 5.4Z/4.5M transcripts are apparent ventrally in metameric clusters of mesodermal cells and laterally in mesodermal cells located between the tracheal pits; 3.5z transcripts are observed in the same region of the mesoderm laterally, but do not appear to be restricted ventrally to metameric clusters of mesodermal cells in this section. (G, H) Alternate sections of a late stage 11 embryo (H, darkfield illumination). 5.4Z/4.5M and 3.5z transcript levels are reduced throughout the mesoderm and ventral ectoderm. Three ventrolateral clusters of ectodermal cells consisting of neuroblasts and underlying dermoblasts remain heavily labeled with the 5.4Z/4.5M probe. 5.4Z/4.5M and 3.5z transcripts are apparent in the anterior midgut, posterior midgut and dorsal wall of the proctodeum at this stage. Abbreviations: (pNR) procephalic neurogenic region; (st) stomodeum; (vNR) ventral neurogenic region. Exposure times: (A, B, E, F) 10 days; (C, D, G, H) 4 days.

Dl function is required during the period of neuroblast segregation to establish epidermal identity in the neurogenic ectoderm (Lehmann et al. 1983). The pattern of Dl transcript accumulation within the neurogenic ectoderm during this period is presented in detail in Fig. 5. The predominant and minor transcripts accumulate in all cells of the ventral and procephalic neurogenic ectoderm during stage 8 immediately prior to neuroblast segregation (Figs4A, B, 5A, B). Dl transcripts are still present in the peripheral ectodermal cells during stage 10 after the majority of the germ band neuroblasts have segregated (Figs 4C, D, 5C, D). Dl transcripts are also present in neuroblasts during this stage, but the neuroblast layer does not appear as heavily labeled in longitudinal sections as the mesodermal and peripheral ectodermal cell layers (Fig. 4C, D). Transverse sections of stage 10 embryos hybridized with the 5.4Z/4.5M probe reveal areas of lower grain density within the neuroblast layer, suggesting that these transcripts are not present at comparable levels in all neuroblasts (Fig. 5C). This is clearly the case for 3.5z since both labeled and unlabeled neuroblasts are distinguishable (Figs 5D, 6D). The difference between the 5.4Z/4.5M and 3.5z hybridization patterns within the neuroblast layer is apparently due to persistence of the nuclear localization of the minor transcripts during this interval. The levels of the predominant and minor Dl transcripts in the neuroblast layer decrease during stage 11 as neuroblast segregation is completed (Figs 4E-H, 5E, F).

Fig. 5.

Distribution of Dl transcripts during neuroblast segregation. Transverse and parasagittal sections of embryos hybridized with the 5.4Z/4.5M (left column) or 3.5z (right column) probes are shown. (A, B) Alternate transverse sections through a stage 8 embryo at approximately 70% egg length. Arrowheads mark the boundaries of the procephalic (pNR) and ventral (vNR) neurogenic regions. All ectodermal cells within these regions accumulate 5.4Z/4.5M and 3.5z transcripts. Hybridization to these transcripts is only slightly above background within the mesoderm (ms) at this stage. (C, D) Alternate transverse sections through a stage 10 embryo at approximately 30% egg length. 5.4Z/4.5M and 3.5z transcripts are present at similar levels in the mesodermal and peripheral ectodermal cell layers. Regions of reduced hybridization within the neuroblast population that separates these cell layers are observed with the SATf 4.5M probe. Individual labeled and unlabeled neuroblasts are distinguishable with the 3.5z probe (D, small arrowheads). Most cells within the peripheral ectoderm are labeled at this stage, although unlabeled cells are occasionally observed with the 3.5z probe. (E, F) Alternate parasagittal sections through a mid-stage 11 embryo. SATLj 4.5M and 3.5z transcripts are present at low levels in the mesodermal and neuroblast (nb) cell layers relative to the peripheral ectoderm. Neuroblast segregation is complete at this stage. Abbreviations: (cf) cephalic furrow; (dec) dorsal ectoderm; (pm) posterior midgut; (vec) ventral ectoderm. Exposure times: (A, B) 10 days; (C, D) 20 days; (E, F) 4 days.

Fig. 5.

Distribution of Dl transcripts during neuroblast segregation. Transverse and parasagittal sections of embryos hybridized with the 5.4Z/4.5M (left column) or 3.5z (right column) probes are shown. (A, B) Alternate transverse sections through a stage 8 embryo at approximately 70% egg length. Arrowheads mark the boundaries of the procephalic (pNR) and ventral (vNR) neurogenic regions. All ectodermal cells within these regions accumulate 5.4Z/4.5M and 3.5z transcripts. Hybridization to these transcripts is only slightly above background within the mesoderm (ms) at this stage. (C, D) Alternate transverse sections through a stage 10 embryo at approximately 30% egg length. 5.4Z/4.5M and 3.5z transcripts are present at similar levels in the mesodermal and peripheral ectodermal cell layers. Regions of reduced hybridization within the neuroblast population that separates these cell layers are observed with the SATf 4.5M probe. Individual labeled and unlabeled neuroblasts are distinguishable with the 3.5z probe (D, small arrowheads). Most cells within the peripheral ectoderm are labeled at this stage, although unlabeled cells are occasionally observed with the 3.5z probe. (E, F) Alternate parasagittal sections through a mid-stage 11 embryo. SATLj 4.5M and 3.5z transcripts are present at low levels in the mesodermal and neuroblast (nb) cell layers relative to the peripheral ectoderm. Neuroblast segregation is complete at this stage. Abbreviations: (cf) cephalic furrow; (dec) dorsal ectoderm; (pm) posterior midgut; (vec) ventral ectoderm. Exposure times: (A, B) 10 days; (C, D) 20 days; (E, F) 4 days.

The accumulation of Dl transcripts within other regions of the ectoderm is the same for the predominant and minor transcripts and can be followed in Figs 4, 5, and 6. Dl transcripts accumulate in all cells of the ectoderm during stage 8 with the exception of some cells in specific regions of the proctodeum and presumptive foregut (Figs 4A, B, 5A, B). Transcript accumulation subsequently falls off in segmentally repeated arrays of dorsolateral ectodermal cells during stage 10 (Fig. 6A). These cells are apparently incorporated into the tracheal pits during stage 11 (Fig. 6B). Dl transcript accumulation is similarly diminished within ectodermal cells of the labial segment just prior to the incorporation of these cells into the salivary gland invaginations at the end of stage 11 (Fig. 6C, D). The accumulation of Dl transcripts in the ventral ectoderm is restricted to ventrolateral clusters of cells at late stage 11 (Fig. 4G, H), and by stage 12 Dl transcript levels are reduced throughout the ventral epidermis (Figs 6E, 8C).

Fig. 6.

Dl transcript levels are reduced in differentiating ectoderm. Parasagittal and transverse sections hybridized with the 5.4Z/4.5M (A, B, C, E) or 3.5z (D) probes are shown. (A) Parasagittal section through the dorsolateral ectoderm (dec) of a stage 10 embryo showing alternate regions of high and low transcript levels. (B) Parasagittal section through the dorsolateral ectoderm of a mid-stage 11 embryo showing relatively low levels of transcripts in ectodermal cells that have invaginated to form the tracheal pits (arrowheads). (C) Parasagittal section through the anterior region of a mid-stage 11 embryo. Transcripts are not detectable in ectoderm of the labial segment (lb). Labeled cells of the developing stomato gastric nervous system are apparent in this section (arrowheads). (D) Transverse section through a late stage 11 embryo at approximately 70% egg length. Transcripts are absent from the labial epidermis which has begun to invaginate to form the salivary gland placodes (arrowheads). Transcripts are present in the epidermis of the procephalic lobe (pl) just above the labial segment. Only one of the two neuroblasts (nb) visible above the labial epidermis is labeled. (E) Transverse section through a late stage 12 embryo at approximately 60% egg length. Transcripts are not detectable within the ventral epidermis (vep) below the ventral nerve cord (vnc), but are present in lateral and dorsal regions of the epidermis. Abbreviations: (am) anterior midgut; (ms) mesoderm; (pr) proctodeum. Exposure times: (A-C) 4 days; (D, E) 20 days.

Fig. 6.

Dl transcript levels are reduced in differentiating ectoderm. Parasagittal and transverse sections hybridized with the 5.4Z/4.5M (A, B, C, E) or 3.5z (D) probes are shown. (A) Parasagittal section through the dorsolateral ectoderm (dec) of a stage 10 embryo showing alternate regions of high and low transcript levels. (B) Parasagittal section through the dorsolateral ectoderm of a mid-stage 11 embryo showing relatively low levels of transcripts in ectodermal cells that have invaginated to form the tracheal pits (arrowheads). (C) Parasagittal section through the anterior region of a mid-stage 11 embryo. Transcripts are not detectable in ectoderm of the labial segment (lb). Labeled cells of the developing stomato gastric nervous system are apparent in this section (arrowheads). (D) Transverse section through a late stage 11 embryo at approximately 70% egg length. Transcripts are absent from the labial epidermis which has begun to invaginate to form the salivary gland placodes (arrowheads). Transcripts are present in the epidermis of the procephalic lobe (pl) just above the labial segment. Only one of the two neuroblasts (nb) visible above the labial epidermis is labeled. (E) Transverse section through a late stage 12 embryo at approximately 60% egg length. Transcripts are not detectable within the ventral epidermis (vep) below the ventral nerve cord (vnc), but are present in lateral and dorsal regions of the epidermis. Abbreviations: (am) anterior midgut; (ms) mesoderm; (pr) proctodeum. Exposure times: (A-C) 4 days; (D, E) 20 days.

The accumulation of the predominant and minor Dl transcripts in the differentiating mesoderm can be followed in Figs 4, 5 and 7. Low levels of Dl transcripts are present in the mesoderm of stage 8 embryos during early germ band elongation (Figs 4A, B, 5A, B). The mesoderm becomes organized into a monolayer of cells during stage 9 within which Dl transcripts accumulate to relatively high levels by early stage 10 (Figs 4C, D, 5C, D). A second, more internal layer of mesodermal cells is established by mid-stage 11 that will subsequently give rise to the visceral musculature (Campos-Ortega and Hartenstein, 1985). Dl transcripts do not accumulate in these cells (Fig. 7C), though Dl transcripts are present in other mesodermal cells located over the developing ventral nerve cord and between the tracheal pits (Figs4E, F, 6B, 7C). These regions of the mesoderm no longer contain detectable levels of Dl transcripts by the end of stage 11 (Figs 4G, H, 7E).

Fig. 7.

Dl transcript levels are reduced in differentiating mesoderm. Parasagittal sections of embryos hybridized with the 5.4Z/4.5M probe (left column) or a beta3-tubulin probe (right column). (A, B) Alternate parasagittal sections through the anterior region of a stage 10 embryo. Dl transcripts are not detectable in cells of the cephalic mesoderm that accumulate beta3-tubulin transcripts (arrowheads). (C, D) Parasagittal sections through dorsolateral regions of the germ band of two stage 11 embryos. Dl transcripts are present in ectodermal and mesodermal cells between the tracheal pits, but are absent from the inner-most layer of mesodermal cells that represents the presumptive visceral mesoderm (vms). beta3-tubulin transcripts are present in the presumptive visceral mesodermal cells, but not in the mesodermal cells between the tracheal pits. (E, F) Alternate parasagittal sections through an early stage 12 embryo. Dl transcripts are no longer present between the tracheal pits in those cells of the somatic mesoderm (sms) that accumulate betoJ-tubulin transcripts. Abbreviations: (am) anterior midgut; (vec) ventral ectoderm. Exposure times: (A-F) 4 days.

Fig. 7.

Dl transcript levels are reduced in differentiating mesoderm. Parasagittal sections of embryos hybridized with the 5.4Z/4.5M probe (left column) or a beta3-tubulin probe (right column). (A, B) Alternate parasagittal sections through the anterior region of a stage 10 embryo. Dl transcripts are not detectable in cells of the cephalic mesoderm that accumulate beta3-tubulin transcripts (arrowheads). (C, D) Parasagittal sections through dorsolateral regions of the germ band of two stage 11 embryos. Dl transcripts are present in ectodermal and mesodermal cells between the tracheal pits, but are absent from the inner-most layer of mesodermal cells that represents the presumptive visceral mesoderm (vms). beta3-tubulin transcripts are present in the presumptive visceral mesodermal cells, but not in the mesodermal cells between the tracheal pits. (E, F) Alternate parasagittal sections through an early stage 12 embryo. Dl transcripts are no longer present between the tracheal pits in those cells of the somatic mesoderm (sms) that accumulate betoJ-tubulin transcripts. Abbreviations: (am) anterior midgut; (vec) ventral ectoderm. Exposure times: (A-F) 4 days.

The pattern of Dl transcript accumulation in the mesoderm was analyzed further by comparing the accumulation of Dl transcripts to the accumulation of transcripts that encode beta3-tubulin, a marker for mesodermal differentiation (Gasch et al. 1988; Leiss et al. 1988; Kimble et al. 1989). 6eio3-tubulin transcripts first accumulate during stage 10 in cells of the cephalic mesoderm (Gasch et al. 1988). Adjacent sections of a stage 10 embryo reveal that the predominant Dl transcripts are present at much lower levels in the cephalic mesodermal cells that accumulate the beta3-tubulin transcript than in the surrounding mesodermal and anterior midgut cells (Fig. 7A, B). We also find that as beta3-tubulin transcripts accumulate during stage 11, first in the innermost mesodermal cell layer and then in the remaining mesodermal cells, the accumulation of predominant Dl transcripts decreases within these regions of the mesoderm in the same temporal order (Fig. 7C-F). Similar results were obtained with the 3.5z probe (data not shown). Thus, the reduction in Dl transcript levels observed in specific cells of the mesoderm correlates with the onset of 6eta3-tubulin transcript accumulation in those cells.

The pattern from germ band shortening to hatching

The number of embryonic cell types in which Dl transcripts accumulate continues to decline until the end of embryogenesis (Fig. 8). No differences between the accumulation patterns of the predominant and minor transcripts have been observed during this interval (data not shown).

Fig. 8.

Distribution of Dl and N transcripts from germ band shortening to hatching. Horizontal and parasagittal sections hybridized with the 5.4Z/4.5M probe (A, C, E, G, I-L) or a N probe (B, D, F, H) are shown. (A, B) Alternate horizontal sections through a mid-stage 12 embryo. Dl transcripts are present in the optic lobes (ol) and posterior midgut (pm) and in subsets of cells within the proctodeum (pr) and somatic mesoderm (arrowheads). N transcripts are present in many of the same cell types, but are relatively more abundant in the epidermis and less abundant in the posterior midgut than Dl transcripts. (C, D) Alternate parasagittal sections through a late stage 12 embryo. Dl transcripts are abundant in the brain (br), ventral nerve cord (vnc), posterior midgut and anterior and posterior regions of the foregut. Dl transcript levels are much reduced in the ventral epidermis and anterior midgut (am). N transcripts are abundant in the ventral epidermis, foregut and developing nervous system and are present at lower levels in the anterior and posterior midgut. (E, F) Alternate horizohtal sections through a stage 14 embryo. Dl transcripts are present in the optic lobes, brain, posterior midgut and cells at the anterior tip of the embryo that may represent sensory organs of the head. Dl transcripts are also present in the germ band in clusters of cells just below the epidermis and in regions that correspond topologically to the developing gonads (arrowheads). N transcripts are still abundant in the epidermis and brain, but are present only at very low levels in the posterior midgut. (G, H) Parasagittal sections through two stage 17 embryos. Dl transcripts are present in the hindgut (hg), proventriculus (pv), peripheral cells of the central nervous system and cells that may represent anterior sense organs (arrowhead). N transcripts are present in the same regions of the embryo as Dl transcripts, but are relatively more abundant in the anterior regions of the embryo than Dl transcripts. Dl and N transcripts are absent from the ventral epidermis (vep) at this stage. (I) Horizontal section through a stage 14 embryo showing the accumulation of Dl transcripts in clusters of cells just below the epidermis. (J) Parasagittal section through the anterior half of a stage 14 embryo. Dl transcripts are present in peripheral cells at the base of the nerve cord, in the brain and at the anterior tip of the embryo (arrowhead). High levels of Dl transcripts are also present in the pharynx and in the posterior region of the esophagus (es). (K) Horizontal section through a stage 17 embryo. The labelled cells of the esophagus present at stage 14 have become incorporated into the proventriculus. (L) Glancing section through the ventral nerve cord of a stage 17 embryo. The relatively heavy labeling of the peripheral cells of the central nervous system is apparent in this section. Exposure times: (A, C, E, G, I-L) 20 days; (B, D, F, H) 4 days.

Fig. 8.

Distribution of Dl and N transcripts from germ band shortening to hatching. Horizontal and parasagittal sections hybridized with the 5.4Z/4.5M probe (A, C, E, G, I-L) or a N probe (B, D, F, H) are shown. (A, B) Alternate horizontal sections through a mid-stage 12 embryo. Dl transcripts are present in the optic lobes (ol) and posterior midgut (pm) and in subsets of cells within the proctodeum (pr) and somatic mesoderm (arrowheads). N transcripts are present in many of the same cell types, but are relatively more abundant in the epidermis and less abundant in the posterior midgut than Dl transcripts. (C, D) Alternate parasagittal sections through a late stage 12 embryo. Dl transcripts are abundant in the brain (br), ventral nerve cord (vnc), posterior midgut and anterior and posterior regions of the foregut. Dl transcript levels are much reduced in the ventral epidermis and anterior midgut (am). N transcripts are abundant in the ventral epidermis, foregut and developing nervous system and are present at lower levels in the anterior and posterior midgut. (E, F) Alternate horizohtal sections through a stage 14 embryo. Dl transcripts are present in the optic lobes, brain, posterior midgut and cells at the anterior tip of the embryo that may represent sensory organs of the head. Dl transcripts are also present in the germ band in clusters of cells just below the epidermis and in regions that correspond topologically to the developing gonads (arrowheads). N transcripts are still abundant in the epidermis and brain, but are present only at very low levels in the posterior midgut. (G, H) Parasagittal sections through two stage 17 embryos. Dl transcripts are present in the hindgut (hg), proventriculus (pv), peripheral cells of the central nervous system and cells that may represent anterior sense organs (arrowhead). N transcripts are present in the same regions of the embryo as Dl transcripts, but are relatively more abundant in the anterior regions of the embryo than Dl transcripts. Dl and N transcripts are absent from the ventral epidermis (vep) at this stage. (I) Horizontal section through a stage 14 embryo showing the accumulation of Dl transcripts in clusters of cells just below the epidermis. (J) Parasagittal section through the anterior half of a stage 14 embryo. Dl transcripts are present in peripheral cells at the base of the nerve cord, in the brain and at the anterior tip of the embryo (arrowhead). High levels of Dl transcripts are also present in the pharynx and in the posterior region of the esophagus (es). (K) Horizontal section through a stage 17 embryo. The labelled cells of the esophagus present at stage 14 have become incorporated into the proventriculus. (L) Glancing section through the ventral nerve cord of a stage 17 embryo. The relatively heavy labeling of the peripheral cells of the central nervous system is apparent in this section. Exposure times: (A, C, E, G, I-L) 20 days; (B, D, F, H) 4 days.

Dl transcripts are present through the end of embryogenesis in the nerve cord, optic lobe primordia and anterior regions of the embryo that could represent sensory organs of the head (Fig. 8A, C, E, G, J, L). The accumulation of Dl transcripts in the nerve cord becomes restricted during this interval to the peripheral cells of the central nervous system, which include neuroblasts, ganglion mother cells and newly born neurons (Fig. 8C, G, J, L). We are unable to determine within which of these cell types Dl transcripts accumulate since these cells are not morphologically distinguishable after stage 12. Dl transcripts may also accumulate in the developing peripheral nervous system during this time (Fig. 8E, I).

Mesodermal accumulation of Dl transcripts is observed during stage 12 in regularly spaced clusters of subepidermal cells (Fig. 8A). We do not observe hybridization of the beta3-tubulin probe to transcripts within these cells (data not shown), but we infer that they are mesodermal based on their location and the absence of a segregated peripheral nervous system during this stage (Campos-Ortega and Hartenstein, 1985). Similarly spaced subepidermal cell clusters are labeled during stage 14 (Fig. 8E, I), but these clusters apparently contain fewer cells than those observed during stage 12. These clusters could represent a subset of the previously labeled mesodermal cells or cells of the peripheral nervous system. Transcripts are also present during stage 14 in regions that correspond to the developing gonads (Fig. 8E). By stage 17, Dl transcripts are no longer detectable in mesodermal cells (Fig. 8G).

Other changes in Dl transcript accumulation during this interval include the disappearance of Dl transcripts from the anterior midgut during stage 12 (Fig. 8C) and from the posterior midgut and entire epidermis during stage 17 (Fig. 8G). The only nonneural accumulation of Dl transcripts that remains during stage 17 is in the proventriculus and in portions of the pharynx and hindgut (Fig. 8G, K).

The Dl pattern compared to N

The dynamic pattern of Dl transcript accumulation prior to germ band shortening contrasts with the ubiquitous accumulation of N transcripts during this interval (Hartley et al. 1987). By the end of embryogenesis, however, the N and Dl transcript accumulation patterns are almost identical (Fig. 8G, H; ibid., Vassin et al. 1987). A comparison of the progressive loss of N and Dl transcripts from specific regions of the embryo is presented in Fig. 8. It is evident that the temporal order of events that leads to their respective final accumulation patterns is not the same. For instance, N transcript levels remain high in the epidermis long after Dl transcript levels drop (Fig. 8E, F). Conversely, Dl transcript levels remain high in the posterior midgut long after N transcript levels drop (Fig. 8E, F). These data reveal that Dl transcript accumulation is regulated independently of N transcript accumulation within many regions of the embryo. In the developing central nervous system, however, the accumulation of Dl and N transcripts is indistinguishable throughout this period (Fig. 8A-D, G, H, J). Thus, the expression of Dl and N in the central nervous system during the later stages of neurogenesis may be coordinately regulated.

Sedimentation of the minor Dl transcripts in polysome gradients

The resolution of our in situ hybridization analysis does not allow us to determine whether the minor Dl transcripts are localized within nuclei or within the cytoplasm at the nuclear periphery. We therefore fractionated 2 to 5h embryo homogenates on sucrose gradients to compare the sedimentation rates of the minor embryonic Dl transcripts to the sedimentation rates of polysomes. Fig. 9A reveals that the major fraction of each minor transcript cosediments with polysomes in the gradient. Fig. 9B further demonstrates that the sedimentation of these transcripts in the gradient is EDTA-sensitive, a characteristic of polysome-bound transcripts (Penman et al. 1968). These results are consistent with the hypothesis that the minor embryonic Dl transcripts are associated with polysomes and therefore localized within the cytoplasm. It remains possible, however, that the minor transcripts accumulate within the nucleus as EDTA-sensitive ribonucleo-protein particles that cosediment with polysomes in sucrose gradients.

Our previous molecular analysis of Dl revealed that at least six distinct transcripts are encoded by this gene (Kopczynski et al. 1988). The two predominant transcripts, 5.4Z and 4.5M, appear to encode the same polypeptide which we have designated D1ZM (ibid.). The structure of the minor transcripts indicates that these transcripts must encode polypeptides significantly different from D1ZM if they are translated. It was therefore important to characterize the accumulation of the predominant and the minor transcripts in our analysis of the spatial expression of Dl. Our results reveal that the predominant and minor Dl transcripts accumulate in the same tissues at approximately the same times throughout embryogenesis. The only apparent difference in their timing is the accumulation of the minor transcripts within the dorsal ectoderm and presumptive mesoderm approximately 20 min prior to the detectable accumulation of the predominant transcripts in these regions. This difference and the differences observed between the respective hybridization patterns in the neuroblast layer during stage 10 are probably due to the different subcellular distributions of these transcripts.

The association of the minor embryonic transcripts with the basal half of the nucleus represents a newly described form of transcript localization in Drosophila. The cosedimentation of these transcripts with polysomes suggests that they are localized within the cytoplasm. Interestingly, the transcripts of a number of genes involved in early pattern formation are localized within the cytoplasm above the nucleus during cellularization (Hafen et al. 1984; Ingham et al. 1985; Harding et al. 1986; Kilcherr et al. 1986; MacDonald et al. 1986; Gergen and Butler, 1988). It has recently been suggested that this apical localization may function to prevent diffusion of these transcripts out of their specific domains of accumulation (Edgar et al. 1987). This same hypothesis could be invoked to explain the subcellular localization of the minor Dl transcripts, although the functional significance of restricting Dl transcripts to neurogenic regions at the start of cellularization is not obvious. Alternatively, the subcellular localization of the minor transcripts could reflect their translation in association with a specific subcompartment of the endoplasmic reticulum. Further characterization of these transcripts and their putative translation products will be required to address these issues.

The apparent association of all four embryonic Dl transcripts (5.4Z, 5.4z, 3.5z and 2.8z) with polysomes supports the hypothesis that Dl encodes multiple polypeptides. The accumulation of these transcripts within the same cells during embryogenesis further suggests that the embryonic Dl product may function as a multimeric complex. If this were the case, then genetic interactions among Dl alleles would exhibit some degree of complexity. Indeed, interallelic complementation has been observed among various Dl mutations (Vâssin and Campos-Ortega, 1987; Alton et al. 1988; M.A.T. Muskavitch, unpublished data).

Cell transplantation experiments have revealed that Dl can act nonautonomously to promote the entry of a cell that lacks Dl function into the epidermal lineage (Technau and Campos-Ortega, 1987). We therefore wanted to determine whether the expression of Dl within the neurogenic region is restricted to a subset of ectodermal cells. We find that this is not the case. Dl transcripts accumulate in most if not all cells of the neurogenic ectoderm immediately prior to and during neuroblast segregation. This result is particularly interesting since the transcripts of two other neurogenic genes, N (Hartley et al. 1987) and E(spl) (Hartley et al. 1988), also appear to accumulate ubiquitously within the neurogenic ectoderm. Thus, there is no apparent distinction between presumptive neuroblasts and presumptive dermoblasts with respect to their accumulation of Dl, N, or E(spl) transcripts.

It has been proposed that the neurogenic genes may function in a system of intercellular communication by which segregating neuroblasts prevent neighboring ectodermal cells from entering the neural lineage (de la Concha et al. 1988). This model requires that a functional distinction exists between neuroblasts and adjacent cells within the ectoderm. The apparently uniform distributions of Dl, N and E(spl) transcripts within neurogenic regions do not provide an evident basis for such a distinction, although post-transcriptional regulation of one or more of these genes cannot be excluded. The asymmetric expression of another neurogenic gene or a gene not yet known to be involved in lineage specification within the ectoderm may provide the basis for this distinction. Alternatively, it is possible that the neurogenic genes function in an intercellular communication system that represses the entry of all neurogenic ectodermal cells into the neural lineage. Assumption of the neural fate would then occur as specific cells lose their susceptibility to this inhibitory influence. In this model, the neurogenic genes would be expressed in most or all cells of the neurogenic ectoderm, as we and others have observed. However, the available data do not exclude either of these models from further consideration.

Dl and N transcripts accumulate in the peripheral cells of the developing central nervous system long after neuroblast segregation is completed. The progressive loss of Dl transcripts from the neuroblast layer prior to germ band shortening suggests that the relatively dense peripheral labeling of the developing central nervous system may correspond to ganglion mother cells and/or newly born neurons. This possibility is particularly intriguing given that the differentiation of sibling neurons has been shown to involve cell-cell interactions in grasshopper embryos (Kuwada and Goodman, 1985). Accumulation in the peripheral nerve cord could also, or alternatively, represent the reinitiation of Dl expression during late neurogenesis in neuroblasts that will divide post-embryonically (Truman and Bate, 1988).

Vässin et al. (1987) inferred that the spatial accumulation pattern of the predominant Dl transcripts during embryogenesis reflects the specific expression of Dl in regions of the embryo within which precursors of the central and peripheral nervous systems develop. However, these authors (ibid.) did not document the accumulation of Dl transcripts in the mesoderm. Our results suggest that the dynamic pattern of Dl transcript accumulation during embryogenesis reflects the initial expression of Dl in many different cell types of the embryo followed by the progressive reduction of Dl expression in those cell types as they differentiate. This hypothesis is supported by our observation that Dl transcript levels decrease in mesodermal cells as they begin to accumulate 6eta3-tubulin transcripts, as well as in ectodermal cells immediately prior to their incorporation into tracheal pits and salivary glands, respectively. The peripheral labeling of the central nervous system is also consistent with this hypothesis since the less differentiated cells of the developing nervous system are located at the periphery of the nerve cord. The late expression of Dl transcripts in the foregut and hindgut could represent an exception to this general hypothesis. However, other genes essential for normal pattern formation are expressed equally late in these regions (Fjose et al. 1985; Ingham et al. 1985; Kornberg et al. 1985; Mlodzik et al. 1985; Krause et al. 1988), suggesting that the differentiation of these tissues may not be complete until very late in embryogenesis.

What role might Dl play in undifferentiated tissues? The predominant Dl product, D1ZM, is a putative transmembrane protein with an extracellular domain that contains a cysteine-rich motif that has been implicated in protein-protein interactions in other systems (Carpenter and Cohen, 1979; Gray et al. 1983; Scott et al. 1983; Sudhof et al. 1985; Wharton et al. 1985; Kidd et al. 1986; Furie and Furie, 1988; Jones et al. 1988; Montell and Goodman, 1988; Rees et al. 1988). The structure of D1ZM suggests that it is involved directly in the cell-cell interactions that are required for proper differentiation of the neurogenic ectoderm. It is therefore possible that Dl is involved throughout the embryo in a number of developmental processes that require intercellular communication for accurate differentiation. Indeed, the Dl loss-of-function phenotype includes defects in the development of muscles, gut, gonads and other tissues (Lehmann et al. 1983), but these defects have been assumed to result from indirect effects of gross neural hypertrophy and reduction of the epidermis. With the aid of stage- and tissue-specific molecular markers, it should be possible to determine whether Dl plays such a pleiotropic role during embryonic development.

The authors thank David Hartley and Spyros Artavanis-Tsakonas for introducing us to in situ hydridization techniques, Mary Kimble for providing the beta3-tubulin probe and Drs Thomas Kaufman, Thomas Blumenthal and the members of our group for careful reading of the manuscript. C.C.K. was supported by a training grant in molecular biology from the National Institutes of Health. M.A.T.M. was supported by a Junior Faculty Research Award from the American Cancer Society. This work was supported by a grant from the National Institutes of Health.

Alton
,
A. K.
,
Fechtel
,
K.
,
Kopczynski
,
C. C.
,
Shepard
,
S. B.
,
Kooh
,
P. J.
and
Muskavitch
,
M.
A. T
. (
1989
).
Molecular genetics of Delta, a locus required for ectodermal differentiation in Drosophila. Dev. Genet
.
10
,
261
272
.
Alton
,
A. K.
,
Fechtel
,
K.
,
Terry
,
A. L.
,
Meikle
,
S. B.
and
Muskavitch
,
M. A. T.
(
1988
).
Cytogenetic definition and morphogenetic analysis of Delta, a gene affecting neurogenesis in Drosophila melanogaster
.
Genetics
118
,
235
245
.
Anderson
,
K. V.
and
Lengyel
,
J. A.
(
1981
).
Changing rates of DNA and RNA synthesis in Drosophila embryos
.
Devi Biol
.
82
,
127
138
Artavanis-Tsakonas
,
S.
(
1988
).
The molecular biology of the Notch locus and the fine tuning of differentiation in Drosophila
.
Trends Genet
.
4
,
95
100
.
Campos-Ortega
,
J. A.
and
Hartenstein
,
V.
(
1985
).
The Embryonic Development of Drosophila melanogaster
.
Springer
:
Berlin
.
Campos-Ortega
,
J. A.
,
Lehmann
,
R.
,
Jimenez
,
F.
and
Dietrich
,
U.
(
1984
).
A genetic analysis of early neurogenesis in Drosophila
.
In Organizing Principles of Neural Development
. (ed.
S.C.
Sharma
), pp.
129
144
.
Plenum Press
:
New York
.
Carpenter
,
G.
and
Cohen
,
S. N.
(
1979
).
Epidermal growth factor
.
A. Rev. Biochem
.
48
,
193
216
.
Clayden
,
E. C.
(
1971
).
In Practical Section Cutting and Staining
, p.
249
.
Churchill Livingstone:London
.
De La Concha
,
A.
,
Dietrich
,
U.
,
Weigel
,
D.
and
Campos-Ortega
,
J. A.
(
1988
).
Functional interactions of neurogenic genes of Drosophila melanogaster
.
Genetics
118
,
499
508
.
Dietrich
,
U.
and
Campos-Ortega
,
J. A.
(
1984
).
The expression of neurogemc loci in imaginai epidermal cells of Drosophila melanogaster
.
J. Neurogenet
.
1
,
315
332
.
Edgar
,
B. A.
,
Odell
,
G. M.
and
Schubiger
,
G.
(
1987
).
Cytoarchitecture and the patterning of fushi tarazu expression in the Drosophila blastoderm
.
Genes and Dev
.
1
,
1226
1237
.
Feinberg
,
A. P.
and
Vogelstein
,
B.
(
1984
).
A techmque for radiolabelling DNA restriction endonuclease fragments to high specific activity
.
Analyt. Biochem
.
137
,
266
267
.
Fjose
,
A.
,
McGinnis
,
W. J.
and
Gehring
,
W. J.
(
1985
).
Isolation of a homeobox-containing gene from the engrailed region of Drosophila and the spatial distribution of its transcripts
.
Nature, Land
.
313
,
284
289
.
Furie
,
B.
and
Furie
,
B. C.
(
1988
).
The molecular basis of blood coagulation
.
Cell
53
,
505
518
.
Gasch
,
A.
,
Hinz
,
U.
,
Leiss
,
D.
and
Renkawitz-Pohl
,
R.
(
1988
).
The expression of beta3- and beta-tubulin genes of Drosophila melanogaster is spatially regulated during embryogenesis
.
Mol. gen. Genet
.
211
,
8
16
.
Gergen
,
J. P.
and
Butler
,
B. A.
(
1988
).
Isolation of the Drosophila segmentation gene runt and analysis of its expression during embryogenesis
.
Genes and Dev.
2
,
1179
1193
.
Gilman
,
A. G.
(
1987
).
G proteins: transducers of receptorgenerated signals
.
A. Rev. Biochem
.
56
,
615
649
.
Gray
,
A.
,
Dull
,
T. J.
and
Ullrich
,
A.
(
1983
).
Nucleotide sequence of epidermal growth factor cDNA predicts a 128,000-molecular weight protein precursor
.
Nature, Lond
.
303
,
722
725
.
Hafen
,
E.
,
Kuroiwa
,
A.
and
Gehring
,
W. J.
(
1984
).
Spatial distribution of transcripts from the segmentation gene fushi tarazu during Drosophila embryonic development
.
Cell
37
,
833
841
.
Harding
,
K.
,
Rushlow
,
C.
,
Doyle
,
H. J.
,
Hoey
,
T.
and
Levine
,
M.
(
1986
).
Cross-regulatory interactions among pair-rule genes in Drosophila
.
Science
233
,
953
959
.
Hartley
,
D. A.
,
Preiss
,
A.
and
Artavanis-Tsakonas
,
S.
(
1988
).
A deduced gene product from the Drosophila neurogenic locus, Enhancer of split, shows homology to mammalian G-protein beta subunit
.
Cell
55
,
785
795
.
Hartley
,
D. A.
,
Xu
,
T.
and
Artavanis-Tsakonas
,
S.
(
1987
).
The embryonic expression of the Notch locus of Drosophila melanogaster and the implications of point mutations in the extracellular EGF-like domain of the predicted protein
.
EM BO J.
6
,
3407
3417
.
Hayashi
,
S.
,
Gillam
,
I. C.
,
Delaney
,
A. D.
and
Tener
,
G. M.
(
1978
).
Acetylation of chromosome squashes of Drosophila melanogaster decreases the background in autoradiographs from hybridization with [125I]-labelled RNA
.
J. Histochem. Cytochem
.
26
,
671
-
619
.
Ingham
,
P. W.
,
Howard
,
K. R.
and
Ish-Horowicz
,
D.
(
1985
).
Transcription pattern of the Drosophila segmentation gene hairy
.
Nature, Lond
.
318
,
439
445
.
Jones
,
F. S.
,
Burgoon
,
M. P.
,
Hoffmann
,
S.
,
Crossin
,
K. L.
,
Cunningham
,
B. A.
and
Edelman
,
G. M.
(
1988
).
A cDNA clone for cytotactin contains sequences similar to epidermal growth factor-like repeats and segments of fibronectin and fibrinogen
.
Proc. natn. Acad. Sci. U.S.A
.
85
,
2186
2190
.
Kidd
,
S.
,
Kelley
,
M. W.
and
Young
,
M. W.
(
1986
).
Sequence of the Notch locus of Drosophila’, relationship of the encoded protein to mammalian clotting and growth factors
.
Molec. cell. Biol
.
6
,
3094
3108
.
Kilcherr
,
F.
,
Baumgartner
,
S.
,
Bopp
,
D.
,
Frei
,
E.
and
Noll
,
M.
(
1986
).
Isolation of the paired gene of Drosophila and its spatial expression during early embryogenesis
.
Nature, Lond
.
317
,
40
44
.
Kimble
,
M.
,
Incardona
,
J. P.
and
Raff
,
E. C.
(
1989
).
A variant β-tubulin isoform of Drosophila melanogaster (betaT) is expressed primarily in tissues of mesodermal origin in embryos and pupae, and is utilized in populations of transient microtubules
.
Devi Biol
.
131
,
415
429
.
Kopczynski
,
C. C.
,
Alton
,
A. K.
,
Fechtel
,
K.
,
Kooh
,
P. J.
and
Muskavitch
,
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 and Dev.
2
,
1723
1735
.
Kornberg
,
T.
,
Siden
,
I.
,
O’Farrell
,
P.
and
Simon
,
M.
(
1985
).
The engrailed locus of Drosophila: in situ localization of transcripts reveals compartment-specific expression
.
Cell
40
,
45
53
.
Krause
,
H. M.
,
Klemenz
,
R.
and
Gehring
,
W. J.
(
1988
).
Expression, modification and localization of the fushi tarazu protein in Drosophila embryos
.
Genes and Dev
.
2
,
1021
1036
.
Kuwada
,
J.
and
Goodman
,
C. S.
(
1985
).
Neuronal determination during embryonic development of the grasshopper nervous system
.
Devi Biol
.
110
,
114
126
.
Lehmann
,
R.
,
Jimenez
,
F.
,
Dietrich
,
U.
and
Campos-Ortega
,
J. A.
(
1983
).
On the phenotype and development of mutants of early neurogenesis in Drosophila melanogaster
.
Wilhelm Roux’s Arch, devl Biol.
192
,
62
74
.
Leiss
,
D.
,
Hinz
,
U.
,
Gasch
,
A.
,
Mertz
,
R.
and
Renkawitz-Pohl
,
R.
(
1988
).
fcetaJ-tubulin expression characterizes the differentiating mesodermal germ layer during Drosophila embryogenesis
.
Development
104
,
525
531
.
MacDonald
,
P. M.
,
Ingham
,
P. W.
and
Struhl
,
G.
(
1986
).
Isolation, structure and expression of even-skipped: a second pair-rule gene of Drosophila containing a homeobox
.
Cell
47
,
721
734
.
Mlodzik
,
M.
,
Fjose
,
A.
and
Gehring
,
W. J.
(
1985
).
Isolation of caudal, a Drosophila homeobox-containing gene with maternal expression, whose transcripts form a concentration gradient at the pre-blastoderm stage
.
EMBO J
.
4
,
2961
2969
.
Montell
,
D. J.
and
Goodman
,
C. S.
(
1988
).
Drosophila substrate adhesion molecule: sequence of laminin Bl chain reveals domains of homology with mouse
.
Cell
53
,
463
473
.
Penman
,
S.
,
Vesco
,
C.
and
Penman
,
M.
(
1968
).
Localization and kinetics of formation of nuclear heterodisperse RNA and polynbosome-associated messenger RNA in HeLa cells
.
J. molec. Biol
.
34
,
49
69
.
Poulson
,
D. F.
(
1937
).
Chromosomal deficiencies and embryonic development of Drosophila melanogaster
.
Proc. natn. Acad. Sci. U.S.A
.
23
,
133
137
.
Preiss
,
A.
,
Hartley
,
D. A.
and
Artavanis-Tsakonas
,
S.
(
1988
).
The molecular genetics of Enhancer of split, a gene required for embryonic neural development in Drosophila
.
EMBO J
.
7
,
3917
3927
.
Rees
,
D. J. G.
,
Jones
,
I. M.
,
Hanford
,
P. A.
,
Walter
,
S. J.
,
Snouf
,
M. P.
,
Smith
,
K. J.
and
Brownlee
,
G. G.
(
1988
).
The role of fl-hydroxyaspartate and adjacent carboxylate residues in the first EGF domain of human factor IX
.
EMBO J
.
7
,
2053
2061
.
Scott
,
J.
,
Urdea
,
M.
,
Quiroga
,
M.
,
Sanchez-Pescador
,
R.
,
Fong
,
N.
,
Selby
,
M.
,
Rutter
,
W. J.
and
Bell
,
G. I.
(
1983
).
Structure of a mouse submaxillary messenger RNA encoding epidermal growth factor and seven related proteins
.
Science
221
,
236
240
.
Sudhof
,
T. C.
,
Goldstein
,
J. L.
,
Brown
,
M. S.
and
Russell
,
D. W.
(
1985
).
The LDL receptor gene: a mosaic of exons shared with different proteins
.
Science
228
,
815
822
.
Technau
,
G. M.
and
Campos-Ortega
,
J. A.
(
1987
).
Cell autonomy of expression of neurogenic genes of Drosophila melanogaster
.
Proc. natn. Acad. Sci. U.S.A
.
84
,
4500
4504
.
Truman
,
J. W.
and
Bate
,
M.
(
1988
).
Spatial and temporal patterns of neurogenesis in the central nervous system of Drosophila melanogaster
.
Devi Biol
.
125
,
145
157
.
Vâssin
,
H.
,
Bremer
,
K. A.
,
Knust
,
E.
and
Campos-Ortega
,
J. A.
(
1987
).
The neurogenic gene Delta of Drosophila melanogaster is expressed in neurogenic territories and encodes a putative transmembrane protein with EGF-like repeats
.
EMBO J
.
6
,
3431
3440
.
Vâssin
,
H.
,
Vielmetter
,
J.
and
Campos-Ortega
,
J. A.
(
1985
).
Genetic interactions in early neurogenesis of Drosophila melanogaster
.
J. Neurogenet
.
2
,
291
308
.
Wharton
,
K. A.
,
Johansen
,
K. M.
,
Xu
,
T.
and
Artavanis-Tsakonas
,
S.
(
1985
).
Nucleotide sequence from the neurogenic locus Notch implies a gene product that shares homology with proteins containing EGF-like repeats
.
Cell
43
,
567
581
.