Induction of motor neurons by Sonic hedgehog is ... Flipbook PDF

Induction of motor neurons by Sonic hedgehog is independent of floor plate differentiation Yasuto Tanabe, Henk Roelink a

---

126 Views
72 Downloads
FLIP PDF 1.12MB

DOWNLOAD FLIP

REPORT DMCA

Induction of motor neurons by Sonic hedgehog is independent of floor plate differentiation Yasuto Tanabe, Henk Roelink and Thomas M. Jessell Howard Hughes Medical Institute, Deptartment of Biochemistry and Molecular Biophysics, Center for Neurobiology and Behavior, Columbia University, 701 West 168th Street, New York, New York 10032, USA.

Background: The differentiation of floor plate cells and motor neurons in the vertebrate neural tube appears to be induced by signals from the notochord. The secreted protein encoded by the Sonic hedgehog (Shh) gene is expressed by axial midline cells and can induce floor plate cells in vivo and in vitro. Motor neurons can also be induced in vitro by cells that synthesize Sonic hedgehog protein (Shh). It remains unclear, however, if the motor-neuroninducing activity of Shh depends on the synthesis of a distinct signaling molecule by floor plate cells. To resolve this issue, we have developed an in vitro assay which uncouples the notochord-mediated induction of motor neurons from floor plate differentiation, and have used this assay to examine whether Shh induces motor neurons in the absence of floor plate differentiation. Results: Floor plate cells and motor neurons were induced in neural plate explants grown in contact with the notochord, but only motor neurons were induced when explants were separated from the notochord. COS cells

transfected with Shh induced both floor plate cells and motor neurons when grown in contact with neural plate explants, whereas only motor neurons were induced when the explants were grown at a distance from Shh-transfected COS cells. Direct transfection of neural plate cells with an Shh-expression construct induced both floor plate cells and motor neurons, with motor neuron differentiation occurring prior to, or coincidentally with, floor plate differentiation. The induction of motor neurons appears, therefore, not to depend on floor plate differentiation. Conclusions: The induction of motor neurons by Shh does not depend on distinct floor-plate-derived signaling molecules. Shh can, therefore, initiate the differentiation of two cell types that are generated in the ventral region of the neural tube. These results show that the early development of motor neurons involves the inductive action of Shh, whereas the survival of motor neurons at later stages of embryonic development requires neurotrophic factors.

Current Biology 1995, 5:651-658

Background

[15]. These findings suggest that Shh normally participates in floor plate induction.

A central problem in vertebrate neural development is to define the cellular and molecular mechanisms that control the differentiation of distinct classes of cells at specific positions within the neural tube. The generation of cell types in the ventral half of the neural tube depends on signals provided by axial mesodermal cells of the notochord [1-6]. Floor plate differentiation at the ventral midline of the neural tube depends on a contact-mediated signal from the notochord, whereas the differentiation of motor neurons more laterally can be induced by a diffusible factor that does not elicit floor plate differentiation [2,7-9]. Floor plate cells subsequently acquire both these inducing activities [5,7,9]. Sonic hedgehog (Shh), also known as Vhh-1 or Hhg-1, is a vertebrate homologue of the secreted protein encoded by the Drosophila gene hedgehog [10,11], and is expressed by the notochord and floor plate at the time that these midline cell groups exhibit their inductive activities [12-16]. Misexpression of Shh in vertebrate embryos can induce the differentiation of floor plate cells at ectopic locations in the neural tube [13-15]. In addition, COS cells expressing Shh can induce floor plate differentiation in vitro when grown in contact with neural plate explants

Whether Shh is also the diffusible factor derived from the notochord and floor plate that induces motor neurons remains unclear. Motor neurons are induced in neural plate explants grown in contact with cells that express Shh [15], but this could reflect the activity of a distinct factor secreted by the floor plate cells that are also induced in these explants. To resolve this issue, we have established in vitro assay conditions (Fig. 1) in which the induction of motor neurons in neural plate explants in response to notochord-derived signals can be uncoupled from floor plate differentiation. We have then examined whether Shh expressed in COS cells or neural plate cells can mimic the ability of the notochord to induce motor neurons in the absence of floor plate cells. Our results show that COS cells transfected with Shh acquire a diffusible activity that is sufficient to induce motor neurons in neural plate explants in the absence of floor plate differentiation. In addition, direct transfection of cells in neural plate explants with Sh induces, independently, motor neurons and floor plate cells. These findings suggest that Shh derived from the notochord normally initiates the differentiation of motor neurons in

© Current Biology 1995, Vol 5 No 6

651

652

Current Biology 1995, Vol 5 No 6

neural tube is sufficient to trigger ectopic floor plate cells [19,20] which, in turn, can induce ventral neurons in adjacent dorsal regions of the neural tube [19]. HNF33 expression provides, therefore, an early and reliable indicator of floor plate differentiation. As an independent marker of floor plate differentiation, we monitored the expression of the mRNA encoding the chemoattractant Netrin-1 (Table 1). Motor neuron differentiation was assessed by expression of the LIM homeodomain proteins Isl-1 and Isl-2 [21], by coexpression of SC1 with Isl-1 and Isl-2, and by expression of of Isl- 1, Isl-2 and choline acetyltransferase (ChAT) mRNAs (Table 1).

Fig. 1. Schematic outline of neural plate induction assays. Neural plate explants were embedded in a collagen gel (grey shading) and grown in contact with, or separated from, the notochord or cellular sources of Shh. the ventral half of the neural tube by a mechanism that is independent of floor plate differentiation.

Results and discussion Identification of floor plate cells and motor neurons inchick neural plate explants. Floor plate and motor neuron differentiation was monitored in explants derived from the intermediate region of the neural plate of stage-10 chick embryos [8], using immunocytochemical and reverse transcription-polymerase chain reaction (RT-PCR) assays. Floor plate differentiation was assessed, primarily, by expression of the winged-helix transcription factor HNF3[3 (Table 1). HNF33 is an early marker of floor plate differentiation in vivo [17,18], and its transcription in neural plate cells in vitro appears to be a direct response to notochordderived signals, as it can occur in the absence of protein synthesis [18]. Moreover, misexpression of HNF33 in the

Uncoupling motor neuron from floor plate differentiation in response to notochord signals When neural plate explants were grown alone in vitro for 36 hours, floor plate and motor neuron markers were not expressed (Fig. 2a,e,f; Table 2a). In contrast, neural plate explants grown in contact with notochord (Fig. la) for 36 hours expressed HNF3 mRNA and protein (Fig. 2b, d,e) and Netrin-1 mRNA (Fig. 2e), indicating the differentiation of floor plate cells. The same explants contained cells that expressed Isl-1 and/or Isl-2 (termed Isl + cells) in combination with SC1 (Fig. 2b-d,f), and in parallel assays explants expressed Isl-1, Isl-2 and ChAT mRNAs (Fig. 2f), indicating the differentiation of motor neurons.

Motor neuron induction by Shh Tanabe et al. To separate experimentally the motor neuron- and floor plate-inducing activities of the notochord, we prevented contact between the notochord and neural plate explants by interposing a membrane filter (Fig. lb). In the absence of contact, the notochord induced motor neuron differentiation (Fig. 2g,h), albeit less effectively as assessed by the number of Isl+ neurons (Table 2a). In contrast, the notochord did not induce floor plate differentiation across a filter, as assessed by the absence of HNF33 expression at 24 hours (data not shown) or 36 hours (Fig. 2g; Table

RESEARCH PAPER

2a). These results extend previous observations [7,81 in that they show that a notochord-derived diffusible factor can induce motor neurons in the absence of floor plate differentiation within the same neural plate explant. Uncoupling motor neuron from floor plate differentiation in response to Shh-dependent signals To examine whether Shh can mimic the contact-dependent and diffusible activities of the notochord, we grew neural plate explants for 36 hours, either in contact with

Fig. 2. Induction of floor plate and motor neuron differentiation by the notochord is distinguished by dependence on cell contact. (a) Neural plate explant grown for 36 h in the absence of the notochord and labelled with antibodies that detect HNF3P and Isl-1 and/or Isl-2 (Isl+ cells). No HNF3p + or Isl+ cells are detected. (b) Neural plate explant grown for 36 h in contact with notochord (n). HNF313 + (red) and Isl+ (green) cells are induced. HNF3[3+ cells are located closer to the notochord/neural plate junction (white dotted line) than are Isl+ cells. (c) Isl+ cells (green) induced in neural plate explants by contact with the notochord coexpress the surface immunoglobulin-like protein SC1 (red). Patches of SC1 + cells that do not express Isl proteins (arrowhead) correspond to floor plate cells [5]. (d) Contact with the notochord induces lsl-2+ cells (green) in neural plate explants. HNF3P3 cells (red) are also induced. (e) RT-PCR analysis of HNF313 and Netrin-1 mRNA induction by contact with the notochord. Lower bands, marked by the arrow, indicate competitive templates introduced to control for the efficiency of the RT-PCR reactions. Intermediate neural plate explants ([i]) and notochord (n)do not express either gene when cultured alone for 36 h. Contact with the notochord (n + [i]) induces HNF33 and Netrin-1 expression (upper bands). (f) RT-PCR analysis of Isl-1, Isl-2 and ChAT mRNA induction by contact with the notochord. Intermediate neural plate explants ([i]) and notochord (n)do not express Isl-1, Isl-2 or ChAT [81 when cultured alone for 36 h. Contact with the notochord (n + [i]) induces the expression of all three genes (upper bands). Lower bands, marked by the arrow, indicate internal standards introduced to control for the efficiency of the RT-PCR reactions. Results in (e) and (f) were obtained from RNA from the same set of explants. Similar results were obtained in 6 experiments. (g) Neural plate explants separated from the notochord by a Nucleopore filter and grown in vitro for 36 h contain Isl+ (green) but not HNF3, + (red) cells. (h) Isl+ cells (green) present in neural plate explants grown transfilter to the notochord express SC1 (red), indicating that they are motor neurons. Patches of SC1 + /I sl- cells were not detected, indicating the absence of floor plate differentiation. Similar results were obtained in 4 separate experiments using either Nucleopore or dialysis membrane filters. Scale bar: (a,c,h) = 20 Im; (b) = 100 I.m; (d,g) = 33 prm.

653

654

Current Biology 1995, Vol 5 No 6

(Fig. c), or separated from (Fig. ld), COS cells transfected with sense or antisense rat Shh cDNA expression constructs [15]. Neural plate explants grown in contact with COS cells transfected with the sense Shh cDNA contained floor plate cells, assessed by expression of HNF3P (Fig. 3a and Fig. 3g, HNF3P3, lane 1; Table 2b) and Netrin- (Fig. 3g, Net-1, lane 1), and motor neurons, assessed by the presence of Isl+/SC1 + and Isl-2+ neurons (Fig. 3a, b,c), and Isl- and ChAT mRNA expression (Fig. 3h, lanes 1). Isl-2 was expressed in 74 2 % (mean + s.e.m., n = 6 explants) of neurons that expressed Isl-1 (Fig. 3c, and data not shown), which probably reflects the earlier expression of Isl-1 by embryonic motor neurons in vivo [21].

not express HNF3I3 mRNA or protein (Fig. 3d,g lane 3) or Netrin- 1 mRNA (Fig. 3g, lane 3), indicating that floor plate differentiation did not occur under these conditions. In contrast, motor neuron differentiation was induced in the absence of contact, as assessed by the presence of Isl+/SC1+ and Isl-2 + neurons (Fig. 3d-f; Table 2b), and Isl- I and ChAT mRNA expression (Fig. 3h, lanes 3). Isl-2 was expressed in 74 + 5 % (mean + s.e.m., n = 8 explants) of neurons that expressed Isl-1 (Fig. 3f, and data not shown). Taken together, these results indicate that the diffusible activity derived from Shh-transfected COS cells induces the entire complement of motor neuron markers detected in response to induction by the notochord.

Neural plate explants grown in the absence of contact with COS cells transfected with the sense Shh cDNA did

In previous studies, we have found that medium conditioned by Shh-transfected COS cells does not induce

Fig. 3. COS cells that express Shh exhibit contact-dependent floor plateand diffusible motor-neuron-inducing activities. (a) A neural plate explant grown in contact with Shh-transfected COS cells for 36 h contains HNF3[3+ (red) and Isl+ (green) cells. The two cell groups are intermingled. Apparently yellow cells probably represent the superimposition of two distinct nuclei in the confocal section. (b) Isl+ neurons (green) in neural plate explants grown in contact with Shh-transfected COS cells express SC1 (red). Isl+ neurons that do not coexpress SC1 probably represent newly-differentiated motor neurons [21]. (c) sl-2 + neurons in intermediate neural plate explants grown in contact with Shh-transfected COS cells. (d) A neural plate explant separated from Shh-transfected COS cells in a collagen gel and grown for 36 h contains Isl+ (green) but not HNF33 + (red) cells. (e) Isl+ neurons (green) induced at a distance from Shh-transfected COS cells coexpress SC1 (red) and are motor neurons. (f) sl-2+ neurons induced at a distance from Shh-transfected COS cells. Intermediate neural plate explants grown in contact with or at a distance from COS cells transfected with antisense Shh cDNA did not contain HNF3I3 +, Isl+ or sl-2 + cells (Table 2, and data not shown). (g) RT-PCR analysis of floor plate induction by Shh-transfected COS cells (s, sense; as, antisense). HNF33 and Netrin-1 expression is induced in neural plate explants grown in contact with Shh-transfected COS cells (lanes 1). but not with antisense Shh-transfected COS cells (lanes 2). HNF33 and Netrin-1 expression is not induced in neural plate explants grown at a distance from Shhtransfected (lanes 3) or antisense Shhtransfected (lanes 4) COS cells. In the same experiment, notochord (n) grown in contact with neural plate explants induces both HNF3[3 and Netrin-1 expression (lanes 5). (h) RT-PCR analysis of motor neuron induction by Shh-transfected COS cells (s, sense; as; antisense). Isl-1 and ChATexpression is induced in neural plate explants grown in contact with Shh-transfected COS cells (lanes 2). Isl1 and ChATexpression are also induced in neural plate explants grown at a distance from Shh-transfected COS cells (lanes 3). Isl-1 and ChAT expression is not induced in neural plate explants exposed to COS cells transfected with antisense Shh (lanes 2 and 4). Notochord (n) grown in contact with neural plate explants induces both Isl-I and ChAT(lanes 5). Results shown in Panels (a-h) have been replicated in 6 different experiments. Scale bar: (a,d) = 16 p.m; (b,e) = 20 pim; (c,f) = 33 pm.

Motor neuron induction by Shh Tanabe et al. Motor neuron induction by Shh Tanabe et at.

RESEARCH PAPER RESEARCH PAPER

Fig. 4. Induction of floor plate and motor neuron differentiation by transfection of neural plate explants with Shh. (a) RT-PCR analysis of floor plate and motor neuron marker expression in neural plate explants analyzed 48 h after transfection with a CMV Shh cDNA expression construct. HNF3P, Netrin-1, Isl-1 and Isl-2 are induced in Shh-transfected explants but not in mock-transfected (-) explants. Isl-1 was also detected albeit at lower levels in Shh-transfected neural plate explants grown in the absence of neurotrophin-3 (data not shown). Cells that expressed HNF3i and Isl immunoreactivity could also be detected (data not shown), although there was an extremely high background, possibly because of cell damage as a consequence of the transfection protocol. (b) Time-course of HNF3 and Is/lexpression in neural plate explants transfected with a CMV Shh cDNA expression construct. (i) In this experiment, neither Isl-1 nor HNF33 are expressed 10 h or 20 h after transfection (lanes 1 and 2), but are detected at 30 h and 40 h (lanes 3 and 4). Netrin-1 and Isl-2 are also expressed after 30 h (data not shown). (ii) In this experiment, Isl-i expression is not apparent at 10 h (lane 1) and can first be detected at 22 h (lane 2). In contrast, HNF3 expression is not detected at either 22 h or 24 h (lanes 2 and 3, although the gene is expressed at 40 h (lane 4). Results showing that Isl-1 expression occurs before or coincident with HNF3P expression were obtained in 4 separate experiments. In a further 3 experiments, Isl-i expression was detected, although HNF3P3 expression could not be detected.

floor plate or motor neuron differentiation in neural plate explants [15]. In the present experiments, the induction of motor neuron differentiation in neural plate explants grown at a distance from Shh-transfected COS cells might result from the provision of a higher concentration, or of a constant source, of Shh. COS cells transfected with antisense Shh did not induce loor plate or motor neuron differentiation under any condition (Fig. 3g,h, lanes 2 and 4, and data not shown). Expression of Shh, therefore, confers COS cells with a contact-dependent floor-plate-inducing activity and a diffusible motor-neuron-inducing activity that does not elicit floor plate differentiation. The most likely explanation of these results is that Shh itself mediates both these activities. COS cells transfected with Shh also exhibit a diffusible activity that can induce ventral forebrain neurons [22] and ventralize segmental plate mesoderm [23]. Independent induction of motor neuron and floor plate differentiation after transfection of neural plate explants with Shh To eliminate the possibility that COS cells transfected with Shh are induced to secrete a distinct factor that mediates the floor-plate- or motor-neuron-inducing activities described above, we tested whether Shh can itself induce motor neuron and floor plate differentiation. Cells within neural plate explants were transfected directly with an Shh expression construct (Fig. le). When assayed 48 hours after transfection with Shh, neural plate explants

expressed HNF3I3, Netrin-1, Isl-l and Isi-2 mRNAs (Fig. 4a). Shh is, therefore, sufficient to induce floor plate and motor neuron differentiation in neural plate explants. To determine whether the induction of motor neurons in neural plate explants transfected with Shh occurs independently of floor plate differentiation, we analyzed the time-course of expression of HNF33 and Isl-l. Expression of Isl-l in neural plate explants transfected with Shh was first detected after about 22 hours, and either preceded (Fig. 4b(ii)) or occurred coincidentally (Fig. 4d(i)) with that of HNF3, depending on the particular experiment. Thus, motor neuron differentiation in neural plate explants transfected with Shh occurs prior to, or synchronously with, floor plate differentiation. These results indicate that Shh acts on neural plate cells to induce the differentiation of motor neurons in a manner that does not require the prior differentiation of floor plate cells. Previous studies in chick embryos have shown that cells in lateral regions of the neural plate are exposed to a motor-neuron-inducing signal from the notochord prior to the differentiation of floor plate cells [8]. Our results support the idea that Shh mediates this notochordderived signal. Shh as an inducer of motor neurons The present findings provide evidence that the ability of the notochord to induce motor neuron differentiation via a diffusible factor can be attributed to the activity of

655

655

656

Current Biology 1995, Vol 5 No 6 Shh. Importantly, the induction of motor neuron differentiation in response to Shh does not appear to require the prior differentiation of floor plate cells. This finding provides direct evidence on the identity of an inductive factor responsible for the differentiation of motor neurons in the embryonic central nervous system. Our results are consistent with the idea that, in vivo, Shh diffuses from the notochord and acts on lateral neural plate cells to induce the expression of transcription factors, including Isl-1 and Isl-2, that commit neural progenitor cells to a motor neuron fate. We cannot, however, exclude the possibility that the initiation of motor neuron differentiation by Shh is mediated solely by the contact-dependent induction of a distinct secreted factor in neural plate cells. If this factor exists, our results would indicate that its induction in neural plate cells by Shh is independent of floor plate differentiation. The mechanism by which Shh induces the differentiation of two distinct ventral cell types, floor plate cells and motor neurons, is not resolved by the present studies. Drosophila and vertebrate hedgehog proteins undergo autoproteolysis to generate an amino-terminal cleavage product, which is associated with the cell surface, and a carboxy-terminal cleavage product, which is diffusible [24-26]. The induction of motor neuron and floor plate differentiation, however, does not appear to result from distinct biological activities that reside in the two cleavage products of Shh. The amino-terminal autoproteolytic cleavage product of Drosophila hedgehog has been shown to mediate the biological activities of the hedgehog precursor [25,27], and in recent studies we have found that the amino-terminal cleavage product of Shh has both floor-plate- and motor-neuron-inducing activity [28]. Taken together with the present finding, these results raise the possibility that, in vivo, motor neuron and floor plate fates are specified by different concentrations of the amino terminal cleavage product of Shh, in a manner similar to that proposed for transforming growth factor 3 (TGF[)-related proteins in the patterning of mesodermal tissues [29-31]. The ability to induce motor neuron differentiation distinguishes Shh from other factors that are known to regulate the embryonic development of motor neurons (Fig. 5). Members of the neurotrophin, TGF3 and fibroblast growth factor (FGF) families, and neurally active cytokines such as ciliary neurotrophic factor (CNTF) and leukemia inhibitory factor (LIF), support the survival of embryonic motor neurons [32,33], but do not elicit the differentiation of motor neurons from progenitor cells in the neural plate (Fig. 4a, and data not shown). Neurotrophin-3 (NT3) has been shown to increase the number of motor neurons in dissociated neural tube cultures, possibly by enhancing the survival and/or proliferation of progenitor cells [34-36]. The initial induction of motor neuron differentiation, therefore, appears to be mediated by Shh, whereas the later progression of motor neuron development may involve the actions of distinct families of secreted signaling proteins.

Fig. 5. Sequential actions of secreted signaling factors during embryonic motor neuron development. The diagram illustrates the distinction between the activities of Shh and those of other factors that regulate motor neuron development. Shh appears to initiate the differentiation of motor neurons in neural plate cells, whereas neurotrophins and other neurotrophic factors regulate later steps in the embryonic development of motor neurons. (GDNF isglial-derived neurotrophic factor.)

Conclusions The data in this paper show that Shh can induce motor neurons in a manner that is independent of floor plate differentiation. The independent induction by Shh of two distinct cell types generated in the ventral region of the neural tube suggests that the differentiation of other classes of ventral neurons may also be dependent on Shhmediated signaling. Previous studies have shown that dorsalin-1, a member of the bone morphogenic protein (BMP) family, can suppress motor neuron differentiation and promote neural crest cell differentiation. The opposing actions of Shh and BMPs may, therefore, control the induction and patterning of distinct cell types along the dorsoventral axis of the neural tube.

Materials and methods Neural plate induction assays Intermediate neural plate explants were dissected from the caudal region of the neural plate of Hamburger-Hamilton [37] (HH) stage-10 chick embryos as described [8]. Notochord explants were dissected after dispase treatment from the caudal region of HH stage-10 chick embryos. Conjugates between notochord and neural plate explants were prepared in collagen gels. When required, notochord and neural plate explants were separated by Nucleopore polycarbonate (pore size 0.1 ,am, COSTAR) or dialysis membrane (Spectrum, Spectra/Por membrane, molecular weight cut off: 50 kD) filters. Explants were grown in defined medium as described [8].

Motor neuron induction by Shh Tanabe et al. For the induction of floor plate cells and motor neurons by Shh-transfected COS cells, intermediate neural plate explants were placed on a monolayer of transfected COS cells, embedded in the collagen gel and cultured for 36 h in F12/N3 medium. To prepare transfilter assays, intermediate neural plate explants were separated from COS cells by a polymerized collagen gel, by Nucleopore Polycarbonate filter or by dialysis membrane (see Fig. 2 legend). Detection of neural markers HNF3P was detected with rabbit antibodies [17,18], Isl-proteins were detected by antibodies that recognize both Islet-1 and Islet-2 (Isl+ cells), and by Isl-2-specific monoclonal antibodies [21,38] (S. Morton, unpublished). The SC1 glycoprotein was detected with SC1-specific monoclonal antibody [39]. Neural plate explants were fixed with 4 % paraformaldehyde at 4 C for -2 h and washed with phosphate-buffered saline (pH 7.4) at 4 C for 1-2 h. Explants were incubated with primary antibodies overnight at 4 C, then with FITC-conjugated goat anti-mouse gG (Boehringer Mannheim) or Texas red-conjugated goat anti-rabbit gG (Molecular Probes) for 1-2 h at 22 C. The explants were then washed and mounted on slides in 50% glycerol:50 % 0.2 M carbonate buffer, pH 9.0 containing paraphenylene diamine (0.4 mg ml-'). Explants were examined on a Zeiss Axiophot microscope equipped with epifluorescence optics. Optical sectioning of explants was performed on a Bio-Rad MRC-500 confocal microscope. Competitive PCR analysis RT-PCR analysis was performed essentially as described [8]. Total RNA was extracted from 10-20 explants cultured in collagen gel with 5 g of glycogen as carrier [40]. An internal standard for competitive PCR analysis was prepared by deleting (in HNF33, Isl-1) or inserting (in Isl-2, Netrin-1, CAT) a 200-300 base-pair fragment within the sequence to be amplified. Plasmid DNAs were linearized and transcribed in vitro to prepare sense-oriented RNA. 100 fg of competitive template RNA was added to the total RNA of each sample and was reverse transcribed using MoMLV-RT (Gibco BRL). One tenth of each reaction product was subjected to PCR using. specific primers flanking the deleted or inserted site of each clone. The primers used were: HNF33: 5'-TCACCATGGCCATCCAGCAGTCG and 5'-CAGCAGGTGCTGCGCTGGAGAGG; Netrin- 1: 5'-TGGGCAGCACCGAGGAC and 5'-CCTTCCATCCCTCAATA; Isl- 1: 5'-TCAAACCTACTTTGGGGTCTTA and 5'-ATCGCCGGGGATGAGCTGGCGGCT; Isl-2: 5'-TGCTGAACGAGAAGCAG and 5'-TGGTAGGTCTGCACCTCCA; CHiA T: 5'-TCCATACGCCGATTTGATGAGGGC and 5'-CTATTGCTTGTCAAATAGGTCTCA. Each PCR cycle was at 94 C for 1 min, 54 C for 1 min and 72 C for 1 min. Twenty two cycles were used for amplifying Isl-2, Isl- 1, HNF33 and Netrin- 1, and twenty cycles for CAT. The PCR products were detected by Southern-blot hybridization with 32P-labeled DNA probes. Blots are aligned such that the tissue-derived band is above the internal standard. Sizes of tissue-derived PCR bands are: HNF33 510bp, Netrin-I 232 bp, Isl- 1 427 bp, Isl-2 304 bp and CHA T 283 bp.

COS cell transfections Transfections with sense or antisense Shh expression plasmids were performed as described [15]. Briefly, 1 g of DNA and 12 pIg ml-' of Lipofectamine (Gibco BRL) in Dulbecco's

RESEARCH PAPER modified Eagles medium (DMEM) supplemented with 1 % glutamine was added to the 80-90 % confluent COS cells in 35 mm dishes. After 5 h of incubation, the transfection reaction was stopped by replacing the medium with DMEM supplemented with 10 % calf serum. Induction assays were carried out after 36 h of incubation. Neural plate transfections CMV- or RSV-LTR-based Shh expression plasmids were transfected directly into intermediate neural plate explants using Lipofectamine (Gibco BRL). 400 ng of DNA and 2 xg of Lipofectamine were mixed in 100 Ixl of F12/N3 and added to neural plate explants. The explants were incubated for 5 h, rinsed and cultured in collagen gels as described [8]. In experiments on Shh-transfected explants, 28 cycles of amplification were used on 1/100th of the tissue-derived cDNA product. The viability of neural plate explants subjected to the transfection protocol was impaired (data not shown). We therefore supplemented the culture medium with neurotrophin-3 (10 ng ml - l, Genentech, Inc.), which has no floor-plate- or motor-neuron-inducing activity (Fig. 4a and data not shown), but which enhances the number of motor neurons that differentiate in dissociated neural tube cultures [34]. Acknowledgements: We thank T. Yamada for help in establishing in vitro assays, S. Morton, S. Pfaff and T. Tsuchida for sl-2-specific antibodies, A. Ruiz i Altaba for the HNF33 antibody and chick HNF3P primer sequences, A. Rosenthal (Genentech) for neurotrophin-3 and M. Tessier-Lavigne for Netrin- primer sequences. We also thank R. Axel, J. Dodd, T. Edlund, S. Pfaff, G. Struhl and A. Tomlinson for comments on the manuscript, and I. Schieren and C. Bireline for help in preparing the text and figures. This work was supported by N.I.M.H. grant MH5073301. Y.T. is an Associate and T.MJ. an Investigator of the Howard Hughes Medical Institute.

References 1. van Straaten HWM, Hekking JMW, Wiertz-Hoessels EL, Thors F, Drukker J: Effect of the notochord on the differentiation of a floor plate area in the neural tube of the chick embryo. Anat Embryol 1988, 177:317-324.

2. Placzek M, Tessier-Lavigne M, Yamada T, Jessell TM, Dodd J: Mesodermal control of the neural cell identity: floor plate induction by the notochord. Science 1990, 250:985-988.

3. Bovolenta P, Dodd J: Perturbation of neuronal differentiation and axon guidance in the spinal cord of mouse embryos lacking a floor plate: analysis of Danforth's short-tail mutation. Development 1991, 113:625-639.

4. Hirano S, Fuse S, Sohal CS: The effect of the floor plate on pattern and polarity in the developing central nervous system. Science 1991, 251:310-313.

5. Yamada T,Placzek M, Tanaka H, Dodd , essell TM: Control of cell pattern in the developing nervous system: polarizing activity of the floor plate and notochord. Cell 1991, 64:635-647.

6. Goulding M, Lumsden A, Gruss P: Signals from the notochord and floor plate regulate the region-specific expression of two pax genes in the developing spinal cord. Development 1993, 117:1001-1016.

7. Placzek M, Jessell TM, Dodd : Induction of floor plate differentiation by contact-dependent, homeogenetic signals. Development 1993, 117:205-218

8. Yamada T, Pfaff SL, Edlund T, Jessell TM: Control of cell pattern in the neural tube: motor neuron induction by diffusible factors from notochord and floor plate. Cell 1993, 73:673-686.

9. Hatta K, Kimmel CB, Ho RK, Walker C: The cyclops mutation blocks specification of the floor plate of the zebrafish central ner-

vous system. Nature 1991, 350:339-341. NUsslein-Volhard C, Wieschaus E: Mutations affecting segment number and polarity in Drosophila. Nature 1980, 287:795-801. 11. Lee JJ, von Kessler DP, Parks S, Beachy PA: Secretion and localized 10.

transcription suggest a role in positional signaling for products of the segmentation gene hedgehog. Cell 1992, 71:33-50.

657

658

Current Biology 1995, Vol 5 No 6 12. 13.

14.

15.

16. 17.

18.

19. 20. 21.

22.

23. 24. 25. 26. 27.

Riddle R, Johnson RL, Laufer E, Tabin C: Sonic hedgehog mediates the polarizing activity of the ZPA. Cell 1993, 75:1401-1416. Echelard Y, Epstein DJ, St-Jacques B, Shen L, Mohler J, McMahon JA, et al.: Sonic hedgehog, a member of a family of putative signaling molecules, is implicated in the regulation of CNS polarity. Cell 1993, 75:1417-1430. Krauss S, Concordet JP, Ingham PW: A functionally conserved homologue of the Drosophila segment polarity gene hedgehog is expressed in tissues with polarizing activity in zebrafish embryos. Cell 1993, 75:1431-1444. Roelink H, Augsberger A, Heemskerk J, Korzh V, Norlin S, Ruiz i Altaba A, et al.: Floor plate and motor neuron induction by Shh, a vertebrate homolog of hedgehog expressed by the notochord. Cell 1994, 76:761-775. Chang DT, Lopez A, von Kessler DP, Chang C, Simandl BK, Zhao R, et al.: Products, genetic linkage and limb patterning activity of a murine hedgehog gene. Development 1994, 120:3339-3353. Ruiz i Altaba A, Prezioso VR, Darnell JE, Jessell TM: Sequential expression of HNF-3a and HNF-33 by embryonic organizing centres: the dorsal lip/node, notochord, and floor plate. Mech Dev 1993, 44:91-108. Ruiz i Altaba A, Placzek M, Baldassare M, Dodd J, Jessell TM: Early stages of notochord and floor plate development in the chick embryo defined by normal and induced expression of HNF-3B. Dev Biol 1995, in press. Sasaki H, Hogan BLM: HNF-3p as a regulator of floor plate development. Cell 1994, 76:103-115. Ruiz i Altaba A, Jessell TM, Roelink H: Restrictions to floor plate induction by hedgehog and winged-helix genes in the neural tube of frog embryos. Mol Cell Neurosci 1995, 6:in press. Tsuchida T, Ensini M, Morton SB, Baldassare M, Edlund T, Jessell TM, et al.: Topographic organization of embryonic motor neurons defined by expression of LIM homeobox genes. Cell 1994, 79: 957-970. Ericson J, Muhr J, Placzek M, Lints T, Jessell TM, Edlund T: Sonic hedgehog induces the differentiation of ventral forebrain neurons: a common signal for ventral patterning along the rostrocaudal axis of the neural tube. Cell 1995, in press. Fan C-M, Tessier-Lavigne M: Patterning of mammalian somites by the surface ectoderm and the notochord: evidence for sclerotome induction by a hedgehog homologue. Cell 1994, 79:1175-1186. Lee JJ, Ekker SC, von Kessler DP, Porter JA, Sun BI, Beachy PA: Autoproteolysis in hedgehog protein biogenesis. Science 1994, 266: 1528-1537. Porter JA, Ekker SC, Young KE, von Kessler DP, Lee JJ, Moses K, et al.: The product of hedgehog autoproteolytic cleavage active in local and long-range signalling. Nature 1995, 374:363-366. Bumcrot DA, Takada R, McMahon AP: Proteolytic processing yields two secreted forms of Sonic hedgehog. Mol Cell Biol 1995, 15:2294-2303. MJ Fietz, A Jacinto, AM Taylor, C Alexandre, PW Ingham: Secretion of the amino-terminal fragment of the Hedgehog protein isnecessary

28.

29. 30.

31. 32. 33. 34.

35. 36.

37. 38. 39. 40. 41. 42.

and sufficient for hedgehog signalling in Drosophila. Curr Biol 1995, 5:643-650. Roelink H, Porter J, Chiang C, Tanabe Y, Chang DT, Beachy PA, et al.: Control of cell pattern in the neural tube: floor plate and motor neuron induction by different concentrations of the amino terminal cleavage product of Sonic Hedgehog autoproteolysis. Cell 1995, in press. Ruiz i Altaba A, Melton D: Interaction between peptide growth factors and homeobox genes in the establishment of anterior-posterior polarity in frog embryos. Nature 1989, 341:33-38. Green J, New HV, Smith J:Responses of embryonic Xenopus cells to activin and FGF are separated by multiple dose response thresholds and correspond to distinct axes of the mesoderm. Cell 1992, 71: 731-739. Gurdon JB, Harger P, Mitchell A, Lemaire P: Activin signalling and response to a morphogen gradient. Nature 1994, 371:487-492. Lindsay RM, Wiegand Si, Altar CA, Di Stefano PS: Neurotrophic factors: from molecule to man. Trends Neurosci 1994, 17:182-190. Snider WD: Functions of the neurotrophins during nervous system development: what the knockouts are teaching us. Cell 1994, 77: 627-638. Averbuch-Heller L, Pruginin M, Kahane N, Tsoulfas P, Parada L, Rosenthal A, et al.: Neurotrophin 3 stimulates the differentiation of motoneurons from avian tube progenitor cells. Proc Natl Acad Sci USA 1994, 91:3247-3251. Kalcheim C, Carmeli C, Rosenthal A: Neurotrophin 3 is a mitogen for cultured neural crest cells. Proc Natl Acad Sci USA 1992, 89: 1661-1665. Pinco O, Carmeli C, Rosenthal A, Kalcheim C: Neurotrophin 3 affects proliferation and differentiation of distinct neural crest cells and is present in the early neural tube of avian embryos. Neurobiol 1993, 24:1626-1641. Hamburger H, Hamilton H: A series of normal stages in the development of the chick embryo. J Morphol 1951, 88:49-92. Ericson J, Thor 5, Edlund T, Jessell TM, Yamada T: Early stages of motor neuron differentiation revealed by expression of homeobox gene Islet-1. Science 1992, 256:1555-1560. Tanaka H, Obata K: Developmental changes in unique cell surface antigens of chick embryo spinal motor neurons and ganglion cells. Dev Biol 1984, 106:26-37. Chomczymski P, Sacchi N: Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 1987, 162:156-159. Serafini TE, Kennedy TE, Galko MJ, Mirzayan C, Jessell TM, TessierLavigne M: The netrins define a family of axon outgrowth-promoting proteins homologous to C. elegans UNC-6. Cell 1994, 78:409-424. Kennedy TE, Serafini TE, de la Torre JR, Tessier-Lavigne M: Netrins are diffusible chemotropic factors for commissural axons in the embryonic spinal cord. Cell 1994, 78:425-435. Received: 10 March 1995; revised: 13 April 1995. Accepted: 13 April 1995.