HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Inazawa, T. Articles by Takahashi, K. Search for Related Content PubMed PubMed Citation Articles by Inazawa, T. Articles by Takahashi, K.

Basic fibroblast growth factor induction of neuronal ion channel expression in ascidian ectodermal blastomeres

Takeshi Inazawa *, Yasushi Okamura ¹ and Kunitaro Takahashi ²

MS 7348 Received 2 September 1997; accepted after revision 12 May 1998.

	ABSTRACT

Top Abstract Introduction Methods Results Discussion References

1. Cleavage-arrested anterior animal (a4-2) blastomeres isolated from eight-cell embryos of Halocynthia aurantium differentiated into neuronal type cells expressing neuron-specific ion channels when they were treated with basic fibroblast growth factor (bFGF). This induction process was very similar to that when a4-2 blastomeres were cultured in contact with anterior vegetal (A4-1) blastomeres from the same embryos or when treated with subtilisin, a serine protease.

2. Other growth factors, transforming growth factor (TGF) 1, activin A, epidermal growth factor (EGF) and nerve growth factor (NGF), had no effect on the default epidermal differentiation of cleavage-arrested a4-2 blastomeres.

3. Messenger RNA of the ascidian neuronal Na⁺ channel, TuNa I, was detected using RT-PCR in a4-2-derived partial embryos of Halocynthia aurantium as well as in the cleavage-arrested a4-2 blastomeres treated with bFGF, confirming the neural inducer activity of bFGF during ascidian embryogenesis.

4. bFGF was effective at concentrations as low as 1 ng ml^-1 in inducing neuronal ion channels in cleavage-arrested a4-2 blastomeres. EC₅₀ for neuronal differentiation was estimated to be around 8 ng ml^-1, and the maximum effect of 90 % neuronalization was obtained with above 100 ng ml^-1.

5. For induction of neuronal differentiation, bFGF was required to be continuously present 8 to 14 h after fertilization. A similar time window was required for cell-contact induction, but it was considerably shorter for subtilisin induction.

6. We discuss whether activation of receptor tyrosine kinase is a common pathway for neural induction by bFGF, subtilisin, and cell-contact with A4-1 blastomeres.

	INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

In vertebrate embryos, neural induction is thought to be the first essential step for development of the nervous system. Since the famous study by Spemann & Mangold (1924) on the organizer of amphibian embryos, many scientists have investigated signal transduction mechanisms triggered by the organizer. However, the molecular nature of this signal transduction is still to be clarified (Slack, 1994).

Ascidians have a notochord from the early stages of development, and their tadpole larvae have been thought to be an ancestral form of vertebrates (Okamura et al. 1993). It has been reported that neuron-specific ion channels are induced in the membrane of cleavage-arrested ectodermal blastomeres prepared from ascidian eight-cell embryos by contact with blastomeres of notochordal lineage or by application of a proteolytic enzyme, subtilisin, at critical stages (Okado & Takahashi, 1990a, b, 1993). This experimental system is suitable for investigations of the molecular mechanism of neural induction because it has been shown that induction of neuron-specific ion channels represents the differentiation of neuronal cells in the ascidian embryo, and is equivalent to neural induction in vertebrate embryos (Reverberi & Minganti, 1946; Okado & Takahashi, 1990a, b; Okamura et al. 1994).

In cultured Xenopus ectodermal cells Kengaku & Okamoto (1993, 1995) showed that the basic fibroblast growth factor (bFGF) acts as a neural inducer during the same time window in which the ectoderm in control embryos is competent for neural induction. In ascidian embryos the neural inducing effects of subtilisin, the most effective enzyme in this respect, suggest possible roles for proteolytic enzymes in the early neuronal development of vertebrate embryos. However, subtilisin is obtained from Bacillus subtilis, and it is unclear whether this enzyme exists physiologically in vertebrate embryos. On the other hand, the existence of bFGF in early vertebrate embryos has been convincingly demonstrated (Slack & Tannahill, 1992). Judging from these findings, bFGF, or other proteins from the same family, should be examined for their neural inducing activities in ascidian ectodermal blastomeres. We aim to elucidate the difference between the effects of proteolytic enzyme and those of growth factors during the early neurogenesis in ascidian embryos.

In the present study, the effects of bFGF on the development of neuron-specific ion channels were examined in the ascidian ectodermal blastomeres. Some of the results have been communicated in preliminary forms (Inazawa et al. 1992, 1994).

	METHODS

Top Abstract Introduction Methods Results Discussion References

Materials

Artificial sea water (ASW) (Jamarin U; Jamarin Laboratory, Osaka, Japan) was used for both culturing ascidian embryos and experiments on isolated blastomeres. The major ionic composition of Jamarin U is (mM): Na⁺, 409; K⁺, 9; Ca²⁺, 10; Mg²⁺, 47; Cl^-, 482·3; SO₄^-, 23·7; HCO₃^-, 2·3. Cytochalasin B (CB), used to arrest cleavage, was purchased from Aldrich and was prepared as a 2 mg ml^-1 solution in dimethyl sulphoxide and added as a one thousandth part to ASW. Recombinant bovine basic fibroblast growth factor (bFGF) and acidic fibroblast growth factor (aFGF) were purchased from Progen Biotechnik (Heidelberg). Recombinant human transforming growth factor (TGF) 1 and 2 were purchased from King Brewing Co., Ltd (Kakogawa, Japan), mouse epidermal growth factor (EGF) from Collaborative Research Inc. (Bedford, MA, USA), mouse nerve growth factor (NGF) from Wako Pure Chemicals (Osaka, Japan) and activin A was a gift from Dr Y. Etoh (Central Research Laboratories, Ajinomoto Co., Yokohama, Japan). Subtilisin was from Boehringer-Mannheim Biochemicals. A blocking antibody against bFGF, 3H3, was a gift from Takeda Chemical Industries, Ltd (Japan).

Animals

Figure 1A depicts a schematic illustration of the experimental design. Halocynthia aurantium embryos were used in this study. Adults were acquired in spring at Wakkanai in northern Japan and were maintained in aquaria with circulating sea water at 3°C. The animals were anaesthetized by chilling in sea water at 4°C and killed by gently crushing with forceps the intersiphonic region between in- and outlet siphons containing the oesophageal ganglia, the CNS of the adult ascidian. The gonads were dissected out and opened up to allow the oocytes to mature in sea water at 10°C for between 3 and 10 h. Eggs were fertilized with sperm obtained from another animal and reared at 9 ± 0·5°C in filtered sea water as described in a previous report (Saitoe et al. 1996). Under these conditions, embryos reached the eight-cell stage at 5 h after fertilization. Subsequent development was as follows; gastrula, 10-20 h; neurula, 20-30 h; tailbud, 30 h and hatching, 55-60 h (Fig. 1).

		View larger version [in this window] [in a new window]
Figure 1. Normal time course of development of the ascidian embryo with a schematic illustration of experimental design and typical current traces from epidermally and neuronally differentiated cell types A, time course of the normal development of an ascidian embryo indicating the positioning of a4-2 (stippled area) and A4-1 (filled area) blastomeres (top). Immediately below the time course of development the experimental procedure is represented: isolated a4-2 cells were cultured alone (uninduced), induced by contact with A4-1 blastomeres or with subtilisin or bFGF applied between 8 and 12 h after fertilization. Cytochalasin B was applied in all cultures from 5 h after fertilization to arrest cleavage. Positioning of the recording electrodes in current- and voltage-clamp modes is shown for each preparation (centre) and an example of typical potential traces with responsible components indicated is shown to the right (dK, delayed K+ current; irK, inward rectifier K+ current; L-Ca, L-type Ca2+ current). B, typical current recordings under voltage clamp conditions from an uninduced a4-2 blastomere (at 58·8 h) showing epidermal differentiation (left) and a neuronally induced a4-2 blastomere (at 54·3 h) (right). Command step potentials are indicated for each trace. External solution, Jamarin U.

For clarification the four types of blastomere, as named by Conklin (1905) (see also Nishida & Satoh, 1983), are as follows: a4-2, anterior animal blastomere, mainly epidermal and anterior CNS lineages; b4-2, posterior animal blastomere, mainly epidermal, brainstem and spinal cord lineages; A4-1, anterior vegetal blastomere, mainly notochordal, brainstem and endodermal lineages; B4-1, posterior vegetal blastomere, mainly muscular, mesenchymal and endodermal lineages.

The eight-cell embryos (at 5 h) were transferred into ASW containing 2 µg ml^-1 CB to arrest cleavage, and dechorionated manually with fine tungsten needles under a binocular microscope. From dechorionated eight-cell embryos, a4-2 blastomeres were isolated with fine glass needles. A4-1 blastomeres were also isolated as inducer cells for cell contact-triggered induction. Eight hours after fertilization (equivalent to the 32-cell stage of normally developing embryos) the blastomeres were transferred into ASW containing 0·2 µg ml^-1 CB. This transfer made the blastomeres in the cell contact-induced preparation adhere more tightly to each other (Okado & Takahashi, 1990a). At various times after this stage most of the a4-2 blastomeres were subjected to induction procedures using cell contact, subtilisin or bFGF, and further cultured at 9°C. No blastomeres cultured in ASW containing 0·2 µg ml^-1 CB had cleaved by the time of electrical recording.

In the RT-PCR experiment, partial embryos were obtained as follows. a4-2 blastomeres were isolated in sea water without CB, from eight-cell embryos. The isolated a4-2 blastomeres were cultured in sea water containing 0-20 ng ml^-1 bFGF or 5 ng ml^-1 activin A without CB. These a4-2 blastomeres showed cleavage and developed into partial embryos.

Electrical recordings

Fifty-five to sixty hours after fertilization, when normally developed embryos had hatched, various voltage-dependent ionic currents in the cell membrane of cultured a4-2 blastomeres were electrically assayed to identify their differentiated cell types. A conventional two-electrode voltage-clamp method was employed. Typical current traces in Jamarin U under voltage clamp of epidermal and neuronal type cells, obtained from a4-2 blastomeres after 55-58 h, are illustrated in Fig. 1B. In neuronal type cells the current was composed of typical Na⁺ channel and delayed K⁺ rectifier channel components, while in epidermal type cells the current included a slowly inactivating Ca²⁺ channel component and/or some unidentified small outward component, as reported previously (Okado & Takahashi, 1990a, b).

To isolate each component of the ionic channel currents, three types of extracellular solutions were used (Table 1), and differentiated cell types were classified according to combinations of ion channel current types expressed and expression of the epidermal cell-specific tunica membrane, as summarized in Table 2.

Table 1. Ionic composition of extracellular solutions (mM)

Table 2. Identification of differentiated cell types

Detection of the voltage-gated Na⁺ channel TuNa I mRNA by RT-PCR

Six to ten a4-2 blastomeres or partial embryos were collected and mRNAs were extracted using the acid guanidinium thiocyanate- phenol-chloroform method (Chomczynski & Sacchi, 1987). cDNAs were synthesized with superscript II (Gibco) using random hexamers. PCR primers radiolabelled with ³²P--ATP by polynucleotide kinase were used for amplification. TuNa I-specific primers were: NaI-2, TGTGGATTCATGGCATATGG; NaI-4c, CGTCTTTCAGTGCTTTGACAG. Cytoskeletal actin primers were: Cact6, ACAACGAACTTCGTGTAGCC; Cact7c, CCATCACCGGAGTCCATAAC. These actin primers discriminate between DNAs of cytoskeletal actin and those from DNAs of muscle type actin of Halocynthia. Primers for myosin were: MY1, AGAAAGGAGGATCATTC; MY2c, GGTATCTCTGCTTAAAG. These myosin primers were based on the sequence of the Halocynthia aurantium (Y. Okamura, unpublished results) obtained from PCR-based cloning using primers specific to the Halocynthia roretzi myosin gene (Makabe et al. 1990). Amplification of TuNa I was performed according to the following protocol: twenty-eight cycles of 94°C for 55 s, 54°C for 70 s, 72°C for 60 s. Amplification of cytoskeletal actin cDNA was performed over twenty-two cycles of the same protocol as for TuNa I. With these cycles, TuNa I, actin and myosin were linearly amplified, as shown by pilot experiments. PCR products were separated on a 6 % polyacrylamide gel and quantified with a phosphoimager (BAS-2000, Fuji).

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Neuronal differentiation of an isolated ectodermal blastomere induced by bFGF

Figure 2A depicts representative ionic currents under voltage-clamp conditions, recorded after 60 h development at 9°C in cleavage-arrested ectodermal blastomeres (a4-2) isolated from eight-cell embryos, which were treated with basic fibroblast growth factor (bFGF) during the competent stage. As shown in the bFGF traces, the inward Na⁺ channel current (-10 mV) and outward delayed K⁺ rectifier current (+25 mV), which are characteristic of neuronal differentiation, were clearly observed. In this experiment 100 ng ml^-1 (5 nM) bFGF was continuously applied from 8 to 12 h after fertilization, equivalent to the period of 32- to 128-cell stages of normally developing embryos. Figure 2A shows the cases of induction by subtilisin and cell contact for comparison. There were no essential differences in the distributions of amplitudes of Na⁺ channel current (-10 mV) and delayed K⁺ channel current (+25 mV) among the three types of neural induction (Fig. 2B). Figure 2C shows that neuron-specific Na⁺ and K⁺ channel expression were induced in 84 % of the bFGF-treated a4-2 blastomeres. Confirming a previous report (Okado & Takahashi, 1993), cells were also neuronally induced by treatment with a serine protease, subtilisin. Subtilisin was applied at 1 mg ml^-1 in ASW from 9 to 11 h after fertilization. Approximately 89 % of subtilisin-treated cells differentiated into neuronal cells. Also as in previous reports (Okado & Takahashi, 1990a, b), 93 % of cleavage-arrested a4-2 blastomeres cultured in contact with A4-1 blastomeres, expressed neuron-specific ion channels.

		View larger version [in this window] [in a new window]
Figure 2. Neuronal differentiation induced by bFGF, subtilisin or cell contact with A4-1 blastomeres Cleavage-arrested a4-2 blastomeres separated from eight-cell embryos of Halocynthia aurantium were cultured until 60 h after fertilization in isolation without treatment (Control), in contact with A4-1 blastomeres (Contact), in isolation and treated with 100 ng ml-1 bFGF in ASW from 8 to 12 h after fertilization (bFGF), or in isolation and treated with 1 mg ml-1 subtilisin in ASW from 9 to 11 h after fertilization (Subtilisin). A, observed current traces under two-electrode voltage-clamp conditions. With depolarization to -10 mV, fast-inactivating inward Na+ currents were observed and with further depolarization to +25 mV delayed outward K+ currents appeared in most bFGF-treated, subtilisin-treated or A4-1-contacted cells, these currents are characteristic of neuronal cells (see Fig. 1). B, distributions of current amplitudes of peak Na+ current (INa,peak, -10 mV) and delayed K+ current (IdK, +25 mV). C, percentages of identified differentiation types with each treatment. Differentiated cell types were defined according to Table 2. Here and in subsequent figures, values in parentheses indicate the number of blastomeres examined with each treatment.

In Fig. 3, Na⁺, delayed K⁺ rectifier and slow Ca²⁺ channel currents isolated using the solutions listed in Table 1 are illustrated for bFGF- and contact-induced a4-2 cells. Exactly the same combination of neuron-specific ion channels was observed in bFGF- as in contact-induced cells. Although not illustrated, the I-V curves for Na⁺, K⁺ and Ca²⁺ channel currents of bFGF-induced cells were identical to those of contact- and subtilisin-induced cells in terms of the characteristics of these currents, such as the threshold, peak and reversal potentials and macroscopic kinetics of currents.

		View larger version [in this window] [in a new window]
Figure 3. Separated ion channel currents expressed in bFGF- and cell contact-induced cells Currents were recorded from a bFGF- and a cell contact-induced a4-2 blastomere under the voltage-clamp conditions. Replacing ASW with the solutions listed in Table 1, the currents through each ion channel were isolated. Those currents associated with neuronal cells, fast-inactivating inward Na+ currents in Na-Mn-TEA solution and delayed outward K+ currents in TMA-Mn-K solution, were recorded from most bFGF-treated cells. Slow Ca2+ channel currents (Sr2+ currents) in TMA-Sr-Mn solution with depolarization were observed. Characteristics of these currents, such as the threshold, peak and reversal potentials, were the same among bFGF-, cell contact- and, although not shown, subtilisin-induced cells.

Effects of other growth factors

To explore whether other growth factors show neural induction properties similar to those of bFGF, the cleavage-arrested a4-2 blastomeres were treated with TGF1, TGF2, activin A, EGF, NGF, and acidic FGF (Fig. 4). All of these growth factors were applied at a concentration of 100 ng ml^-1 from 8 to 12 h after fertilization. Although activin A, TGF1 and TGF2, like bFGF, are all known to be mesoderm inducing in amphibian embryos (Knöchel et al. 1987; Slack et al. 1987; Rosa et al. 1988; Asashima et al. 1990), none of them induced neuronal or mesodermal differentiation in ectodermal a4-2 blastomeres. The treated blastomeres became mostly epidermal cells expressing slow Ca²⁺ channel currents and/or the tunica membrane at the surface as in the non-treated cleavage-arrested cells, although TGF2 somewhat suppressed epidermal differentiation.

The bFGF receptors contain a tyrosine kinase domain at the cytoplasmic side (Lee et al. 1989) and kinase activities appear to be important for various bFGF effects (Schlessinger & Ullrich, 1992). Among the growth factors tested, EGF and NGF receptors possess tyrosine kinase activities similar to bFGF receptors (Hunter & Cooper, 1981; Klein et al. 1991), but neither of them induced neuronal differentiation of a4-2 blastomeres. These results are consistent with those in Xenopus embryonic cell cultures reported by Kengaku & Okamoto (1993, 1995). Acidic FGF may have slight neural induction properties because the percentage of epidermally differentiated cells decreased in comparison with the control, and because the delayed rectifier K⁺ channel, which is one of the characteristics of neuronally differentiated cells, was expressed in some of the aFGF-treated cell population (stippled area in the aFGF column in Fig. 4).

Thus only bFGF exhibited clear neural induction activity in ascidian ectodermal blastomeres. This is in accordance with the result reported by Kengaku & Okamoto (1993) who found that in cultured Xenopus ectodermal cells no other growth factors, besides acidic FGF, induced neuronal differentiation. Therefore, in the following experiments only bFGF was used to examine the neural induction mechanism in ascidian embryos.

		View larger version [in this window] [in a new window]
Figure 4. The inducing activity of various growth factors Cleavage-arrested a4-2 blastomeres were treated with 100 ng ml-1 bFGF, aFGF, activin A, TGF1, TGF2, EGF or NGF in ASW from 8 to 12 h after fertilization. Percentages of differentiated cell types are illustrated. Subtilisin- and contact-induced, and non-treated cleavage-arrested (Control) a4-2 cells were also examined for comparison. The subtilisin-treatment was carried out at a concentration of 1 mg ml-1 in ASW from 9 to 11 h after fertilization. Differentiated cell types were identified according to criteria in Table 2.

RT-PCR detection of TuNa I mRNA

A voltage-gated sodium channel gene TuNa I, which has been cloned from cDNA libraries of Halocynthia roretzi tadpole larvae and which was proved by in situ hybridization to be transcribed exclusively in neurons (Okamura et al. 1994; Okada et al. 1997), is a reliable molecular neuronal marker. A TuNa I homologue has been shown to be present in Halocynthia aurantium. In the following experiments, neuronal differentiation of bFGF-treated a4-2 blastomeres of Halocynthia aurantium was confirmed by RT-PCR detection of TuNa I mRNA.

After electrical recordings, six to ten neuronally differentiated a4-2 blastomeres were collected, and mRNAs were examined using the RT-PCR method. Non-treated cleavage-arrested epidermally differentiated a4-2 blastomeres were also examined (Fig. 5A). TuNa I mRNA was detected in cleavage-arrested a4-2 blastomeres which were neuronally induced by bFGF. TuNa I mRNA was also detected in subtilisin- or contact-induced neuronal cells, but not in non-treated cleavage-arrested epidermal cells. Thus, neuron-specific transcription of TuNa I mRNA was again confirmed, showing concurrence between the results from electrophysiological and molecular biological methods.

Although the cleavage-arrested system has many advantages especially because of the simple morphology and ease of electrical recording either under constant current or under voltage-clamp conditions it is suspected that cytochalasin B (CB) may have aberrant effects on the induction process. Using TuNa I as a neuronal marker, it was possible to detect the expression of neuron-specific Na⁺ channels without the need for electrical recording. In the following experiments, the induction of neuronal cells by bFGF was confirmed in the partial embryos derived from isolated a4-2 blastomeres (not mesodermal lineage) in the absence of CB.

		View larger version [in this window] [in a new window]
Figure 5. RT-PCR detection of TuNa I mRNA A, six to ten cleavage-arrested a4-2 blastomeres were collected and mRNAs examined using RT-PCR. Neuronal differentiation of the blastomeres with inducing treatments had been confirmed in electrical recording prior to RT-PCR. TuNa I mRNA was detected in the cleavage-arrested a4-2 blastomeres which were neuronally induced by bFGF. TuNa I mRNA was also detected in subtilisin- and contact-induced neuronal cells, but not in non-treated (Control) cleavage-arrested epidermal cells. Cytoskeletal actin mRNA was also examined to qualify the integrity of extracted mRNA. B, a4-2 blastomeres were isolated from eight-cell embryos which were not cleavage arrested. The a4-2 blastomeres developed into partial embryos in the absence of cytochalasin B. Representative pictures are shown for each set of partial embryos cultured in the presence of 0, 5 or 20 ng ml-1 bFGF. Arrowheads indicate tunica borders. C, a4-2 blastomeres, isolated from eight-cell embryos which were not cleavage arrested, were cultured in sea water containing 0-20 ng ml-1 bFGF or 5 ng ml-1 activin A. From the developed partial embryos for each treatment, ten were collected and subjected to the examination with the RT-PCR method. TuNa I, myosin and cytoskeletal actin mRNAs were examined. Intact whole embryos (Whole) were also examined.

a4-2 blastomeres were cultured in sea water containing 0-20 ng ml^-1 bFGF or 5 ng ml^-1 activin A. The a4-2 blastomeres developed into partial embryos in the absence of CB (Fig. 5B). Ten partial embryos from each treatment were collected, and tested for the presence of TuNa I mRNA using the RT-PCR method (Fig. 5C). Myosin and cytoskeletal actin mRNAs were also tested for. The partial embryos with no treatment developed into epidermal balls surrounded by tunica, an epidermis-specific extracellular substance. TuNa I gene expression was not detected in these embryos, being compatible with the previous finding that the A4-1 blastomere is required for TuNa I gene induction (Okamura et al. 1994). By treatment with 5 ng ml^-1 bFGF, some cells were not associated with tunica, and thus tunica extended only to one side, forming a bulge. However, TuNa I mRNA was not detected in these embryos, indicating that suppression of epidermal differentiation occurred without TuNa I induction by a low dose of bFGF. When a4-2 blastomeres were treated with 20 ng ml^-1 bFGF, expression of the tunica further decreased. This suggests that the suppression of epidermal differentiation occurs in a dose-dependent manner. In these embryos TuNa I gene transcription was detected, but mesodermal features such as myosin gene transcription or notochord-specific vacuole structure were not detected, indicating that neuronal differentiation was induced directly by bFGF and not as a secondary effect through contact with mesodermal tissue. Activin A induced neither TuNa I transcription nor suppressed epidermal differentiation. This finding is compatible with the result of experiments on cleavage-arrested blastomeres (Fig. 4). This series of experiments was repeated in three separate batches and similar results were obtained (in one case, TuNa I gene was detected with 5 ng ml^-1 bFGF). The dose of 5-20 ng ml^-1 bFGF is also compatible with that for induction of Na⁺ currents in cleavage-arrested a4-2 blastomeres (Fig. 6).

Dose dependency

Figure 6A shows the dose dependence of bFGF effects on a4-2 blastomeres for induction of neuron-specific ion channels. bFGF was continuously applied from 8 to 12 h after fertilization for all doses. Neuron-specific ion channels were induced at a concentration as low as 1 ng ml^-1. The percentage of cells differentiated to the neuronal type increased as the concentration of bFGF was increased; the neuronal differentiation became more than 70 % at 30 ng ml^-1, and the effect seemed to be almost saturated at 100 ng ml^-1. Assuming the Michaelis-Menten type relation, EC₅₀ for neuronal differentiation was estimated to be around 8 ng ml^-1, and the maximum effect was estimated to be around 90 % (Fig. 6B). At 100 ng ml^-1 bFGF the percentage of neuronally differentiated cells was above 80 % which was almost at the same level as that induced by subtilisin or by cell contact. From these results the concentration of bFGF employed in the rest of this study was fixed at 100 ng ml^-1.

		View larger version [in this window] [in a new window]
Figure 6. Dose dependency of the inducing activity of bFGF Cleavage-arrested a4-2 blastomeres were treated with 0-100 ng ml-1 bFGF in ASW from 8 to 12 h after fertilization. Subtilisin-treated, contact-induced and non-treated cleavage-arrested cells were also examined for comparison. Subtilisin-treatment was 1 mg ml-1 in ASW from 9 to 11 h after fertilization. A, percentages of differentiated cell types are illustrated. The cell types were identified according to Table 2. B, Michaelis-Menten fitting for the dose-response curve of the neural induction by bFGF. The percentage differentiation is calculated as follows: % = 90 [bFGF]/([bFGF] + 8), this equation predicts that EC50 is around 8 ng ml-1 and that the effect saturates at the level at which 90 % of cells were induced to become neuronal type.

Time windows for neural induction by bFGF, A4-1 cell contact and subtilisin

To determine the latest starting point at which bFGF is effective for inducing the neuron-specific ion channel expression, application of bFGF was started at various times after fertilization, and bFGF was not washed out until 60 h when electrical recordings were performed (Fig. 7A). Neuronal differentiation was induced in 88 % of a4-2 blastomeres when application of bFGF was started at 8 h after fertilization. However, the percentage rapidly decreased thereafter. Neuron-specific ion channels were not detected when bFGF application was started at 14 h after fertilization.

Next, to determine the minimum required period for bFGF application, the treatment was started at 8 h and bFGF was washed out at various times after fertilization (Fig. 7B). Since it was necessary to completely eliminate the activity of residual bFGF, a blocking antibody, 3H3 (1 µg ml^-1), was applied after each wash-out of bFGF, and the cells were kept in ASW containing the antibody thereafter. 3H3 at a concentration of 1 µg ml^-1 is known to completely block the bFGF-dependent growth of a transplanted mouse myeloma cell line (Hori et al. 1991). When bFGF was not washed out until 60 h, 88 % of a4-2 blastomeres expressed neuron-specific ion channels in their membranes. When the effect of bFGF was blocked by 3H3 at 14 h after fertilization, the percentage of neuronally differentiated cells (78 %) was not significantly different from that of the cells in which bFGF effects were not blocked. However, the percentage decreased to 24 % when 3H3 was applied at 12 h after fertilization, and neuron-specific ion channels were not detected when 3H3 was applied at 8 h after fertilization, i.e. bFGF and its antibody were given at the same time.

Figure 7C depicts the superimposed curves illustrated in Fig. 7A and B. The densely hatched area indicates the competent time window, from 8 to 14 h, during which a continuous presence of bFGF was required to induce the full neuronal differentiation in the treated cell population.

		View larger version [in this window] [in a new window]
Figure 7. Determination of the competent time window for neuronal differentiation induced by bFGF Experiments to study the time requirement for bFGF treatment. A, effect of starting time () of bFGF treatment on the percentage of neuronally differentiated cells. Cleavage-arrested a4-2 blastomeres were treated with 100 ng ml-1 bFGF in ASW from the indicated points to the time of recording. The Y axis indicates percentage of neuronally differentiated cells, and the X axis indicates starting time of the bFGF treatment. Although almost the full inducing effect was obtained by bFGF treatment from 8 h after fertilization, the effect decreased rapidly thereafter. This means that to ensure maximum induction bFGF treatment needs to be started by 8 h after fertilization (hatched area). B, effect of finish time () of bFGF treatment. Cleavage-arrested a4-2 blastomeres were treated with 100 ng ml-1 bFGF in ASW from 8 h after fertilization. At each indicated point, bFGF was washed out and the activity of residual bFGF was blocked by immunoneutralizing monoclonal antibody 3H3. The Y axis indicates the percentage of neuronally differentiated cells, and X axis indicates washing and blocking points of the bFGF treatment. By blocking bFGF activity at the same time as its application at 8 h after fertilization, no cell differentiated into the neuronal type. When the blocking point was later than 14 h after fertilization, the inducing effect was almost maximal in the population of treated cells. This means that bFGF treatment needs to be continued until 14 h after fertilization (hatched area). C, time window for bFGF treatment. The curves illustrated in A and B are superimposed here. The overlap of the hatched areas in A and B, i. e. from 8 to 14 h after fertilization, is shown. It means that for fully effective induction bFGF must be applied continuously from 8 to 14 h after fertilization.

Similar experiments were performed to determine the competent time windows for A4-1 cell contact and subtilisin neural induction (Fig. 8B and C). For cell contact induction, the pairing of cells, which had been separately prepared, was carried out at various times after fertilization to determine the latest effective time ( in Fig. 8B), and the a4-2 and A4-1 cells were separated at indicated times to determine the minimum contact period required ( in Fig. 8B). The time requirement for A4-1 cell contact, which is thought to be the natural induction in the ascidian embryos, was identical to that for bFGF application. However, when the competent time window was similarly determined for subtilisin induction, applying Streptomyces subtilisin inhibitor (SSI) after washing out the enzyme as previously reported by Okado & Takahashi (1993), the window was found to be much narrower than those for bFGF and cell contact, lasting just 1 h.

		View larger version [in this window] [in a new window]
Figure 8. Competent time windows for neuronal differentiation induced by various treatments Similar graphs to Fig. 7C were plotted for neural induction by A4-1 contact and subtilisin. A, time window for bFGF. Identical graph as shown in Fig. 7C, repeated here for ease of comparison. B, time window for A4-1 cell contact. To ensure full neural induction a4-2 cells needed to be in continuous contact with A4-1 cells from 8 to 14 h after fertilization. This time window coincided with that for bFGF. * Data for the starting times of contact were taken from a previous report (Okado & Takahashi, 1990). C, time window for subtilisin. To ensure full neural induction subtilisin was required to be present only from 10·5 to 11·5 h after fertilization. This time window was considerably shorter than those for bFGF and A4-1. ** Data for the blocking times were transcribed from a previous report (Okado & Takahashi, 1993), in which the activity of residual subtilisin after washing out was blocked by Streptomyces subtilisin inhibitor (SSI).

Inward rectifier K⁺ channel expression

As previously reported, differences in current amplitude through inward rectifier K⁺ (irK) channels in a4-2 blastomeres at the late gastrula stage are the earliest signs of cells branching down neuronal or epidermal pathways. In the epidermally committed cleavage-arrested a4-2 blastomeres, the irK current amplitudes increase after 15 h and reach a plateau at 25 h. However, in blastomeres neuronally induced, either by subtilisin or cell contact, the irK current amplitudes increase after 15 h and then decrease at around 20 h (Okamura & Takahashi, 1993). Hence it was inferred that the cause of this difference is that irK gene transcription, which seems to start before the competent period, is suppressed immediately after neural induction (Okamura & Takahashi, 1993). Thus it is important to examine whether the bFGF treatment suppresses the increase of irK current amplitudes.

In the present experiment, the time courses of changes in the irK current amplitudes were compared among four groups of cleavage-arrested a4-2 blastomeres: bFGF-, subtilisin- and contact-induced and non-treated groups (Fig. 9). In epidermally differentiated cells without induction, the irK current amplitude increased until 20 h after fertilization, and remained at the plateau level until 30 h after fertilization. In both subtilisin-treated and A4-1 cell contact-induced neuronally differentiated cells, the amplitude of irK current increased until 18 h after fertilization, and started to decrease at around 20 h after fertilization. These results are in accord with the previous report (Okamura & Takahashi, 1993). In bFGF-treated neuronally differentiated cells, changes in the irK current amplitude showed a similar time course to those of the subtilisin-treated or A4-1 cell contact-induced cells. Thus, in the early processes of neuronal determination, the three methods of induction seemed to result in similar signal transduction for changes in electrical membrane properties.

		View larger version [in this window] [in a new window]
Figure 9. Changes in inward rectifier K+ currents during early phases of epidermal and neuronal differentiation Cleavage-arrested a4-2 blastomeres were treated with 100 ng ml-1 bFGF () in ASW from 8 to 12 h after fertilization. From those cells irK currents were recorded with hyperpolarizing pulses up to -150 mV in ASW from 5 to 30 h after fertilization. After this irK recording, the blastomeres were cultured until 60 h and finally differentiation types were examined. Subtilisin-treated (1 mg ml-1 from 9 to 11 h after fertilization, ), A4-1 cell contact-induced (), and non-treated cleavage-arrested () cells were also examined for comparison. Amplitudes of irK currents with hyperpolarizing pulse to -135 mV were plotted.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

The present study revealed that bFGF induced neuron-specific ion channels in ectodermal blastomeres, similar to the cases where neuronal differentiation was induced by cell contact with presumptive notochordal blastomeres or by application of a serine protease, subtilisin. Other growth factors, such as activin A or EGF, did not show such an effect. The continuous presence of bFGF from 8 to 14 h after fertilization was required to fully express neuron-specific ion channels in the treated cells. This time window was identical to that for induction by cell contact. The time courses of irK channel expression during early stages of neuronal determination were also the same among bFGF application, subtilisin application and cell contact. RT-PCR experiments revealed that the mRNA encoding a neuron-specific Na⁺ channel gene TuNa I was transcribed in bFGF-treated blastomeres. All of the above results suggest that bFGF or its family proteins are neural inducer candidates for ascidian embryos as well as for Xenopus ectodermal cells (Kengaku & Okamoto, 1993, 1995).

In the present study, effects of bFGF, such as induction of neuron-specific ion channels, were tacitly assumed to affect the ectodermal blastomeres directly. However, there is the possibility that bFGF induces mesodermal cells, and subsequently that neuronal cells are induced by the mesodermal cells. bFGF has been known to exert the mesoderm-inducing activity (Slack et al. 1987) and the mesodermal cells apparently possess the neural induction activity (Spemann & Mangold, 1924; Okado & Takahashi, 1988, 1990a). Moreover, it has been reported that bFGF faciliates notochord differentiation in partial embryos from ascidian notochord precursor blastomeres (Nakatani et al. 1996). However, it is unlikely that induction of neuron-specific ion channels and TuNa I mRNA is mediated by bFGF-induced notochordal cells for the following reasons. First, notochord-specific features were not developed in bFGF-treated a4-2-derived partial embryos (Nakatani et al. 1996). Second, the present experimental system consisted of a single cell; multiple cell type characteristics are not usually expressed in single cleavage-arrested blastomeres from ascidian embryos (Hirano et al. 1984; Okado & Takahashi, 1990a), instead the exclusivity of differentiation is the predominant nature of those cells (Weiss, 1939).

In Xenopus embryos the minimal dose of bFGF required to induce neuronal differentiation is extremely low (5 pM) compared with that for mesoderm induction (250 pM) (Kengaku & Okamoto, 1993), and it has been convincingly proved that bFGF at the former dose expresses exclusively neuronal markers and no mesodermal markers in Xenopus ectoderm cells (Kengaku & Okamoto, 1995; Lamb & Harland, 1995). Although the endogenous level of bFGF in the Halocynthia tadpole is not known, the threshold concentration of bFGF (50 pM) for neuronal induction in the present study may be slightly high in comparison with the case in Xenopus embryos. In the 9°C ASW where the ascidian embryos were cultured, the high ionic strength and low temperature tended to reduce the ionic binding forces between macromolecules, and thus a relatively low affinity of bFGF to its receptors might be expected. Subtypes of bFGF receptors might also differ between ascidian and Xenopus, further, the subtypes might also differ between neuron-precursor blastomeres and notochord-precursor blastomeres within an ascidian embryo, because 0·5 pM bFGF facilitates notochord differentiation in partial embryos from ascidian blastomeres (Nakatani et al. 1996). The effect of aFGF in the ascidian blastomeres may be ascribed to the cross-reactivity of aFGF with bFGF receptors (Gillespie et al. 1989).

The time window required for bFGF to induce neuronal differentiation coincided with that for cell contact with A4-1 blastomeres, between 8 and 14 h after fertilization. In the case of subtilisin, the required duration was considerably shorter, probably less than 1 h, during a period between 8 and 14 h after fertilization. This suggests that the neural induction caused by cell contact is more likely to be mediated by the action of bFGF or its family proteins than by that of proteases. The neuronal determination processes immediately after the induction phase represented by suppression of the irK channel currents were common among cells which were induced by bFGF, cell contact with A4-1 blastomeres and subtilisin. This suggests a common signal transduction step after the induction phase. Furthermore the neural induction activity of bFGF supports the notion that signal transduction through protein tyrosine kinases may play an important role in neural induction. Preliminary experiments showed that application of tyrosine kinase inhibitors blocked the neural induction by subtilisin and by cell contact (Inazawa et al. 1992). This indicated that subtilisin-induced neuronal differentiation may also be initiated by tyrosine kinase activity which is an essential downstream step in the bFGF signal transduction pathway. To explain the difference in the time requirement between bFGF and subtilisin treatments, we suggest that tyrosine kinases at the cytoplasmic side of bFGF receptors are constitutively activated when the extracellular domains of the receptors are truncated by subtilisin (a constitutively activated receptor tyrosine kinase mutant which lacks its extracellular domain as reported in Drosophila by Basler et al. 1991). Then, the tyrosine kinase constitutively activated by subtilisin at the beginning of the competent phase, around 8 h after fertilization, may be effective throughout the induction phase from 8 to 14 h after fertilization.

A recent report in Xenopus embryos (Launay et al. 1996) has attracted our attention. In that study, artificially expressed truncated FGF receptors that lacked tyrosine kinase activity inhibited the neural induction by noggin and by the endogenous inducer, indicating that it is the tyrosine kinase activities of bFGF receptors which are required for neural induction in Xenopus embryos. Assuming ascidian bFGF receptor genes being cloned, the same method would become available and the hypothesis could be tested to see if bFGF is the neural inducer common among vertebrates and protochordates.

	REFERENCES

Top Abstract Introduction Methods Results Discussion References

Asashima, M., Nakano, H., Shimada, K., Kinoshita, K., Ishii, K., Shibai, H. & Ueno, N. (1990). Mesodermal induction in early amphibian embryos by activin A (erythroid differentiation factor). Roux's Archives of Developmental Biology 198, 330-335.
Basler, K., Christen, B. & Hafen, E. (1991). Ligand-independent activation of the sevenless receptor tyrosine kinase changes the fate of cells in the developing Drosophila eye. Cell 64, 1069-1081	[Medline]
Chomczynski, P. & Sacchi, N. (1987). Single-step method for RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Analytical Biochemistry 162, 156-159	[Medline]
Conklin, E. G. (1905). The organization and cell-lineage of the ascidian egg. Journal of Academy of Natural Science Philadelphia 2nd series 13, 1-119.
Gillespie, L. L., Paterno, G. D. & Slack, J. M. W. (1989). Analysis of competence: receptors for fibroblast growth factor in early Xenopus embryos. Development 106, 203-208	[Abstract]
Hirano, T., Takahashi, K. & Yamashita, N. (1984). Determination of excitability types in blastomeres of the cleavage-arrested but differentiated embryos of an ascidian. The Journal of Physiology 347, 301-325	[Abstract]
Hori, A., Sasada, R., Matsutani, E., Naito, K., Sakura, Y., Fujita, T. & Kozai, Y. (1991). Suppression of solid tumor growth by immunoneutralizing monoclonal antibody against human basic fibroblast growth factor. Cancer Research 51, 6180-6184	[Medline]
Hunter, T. & Cooper, J. A. (1981). Epidermal growth factor induces rapid tyrosine phosphorylation of proteins in A431 human tumor cells. Cell 24, 741-752	[Medline]
Inazawa, T., Okamura, Y. & Takahashi, K. (1994). Basic fibroblast growth factor causes neural induction on ascidian ectodermal blastomeres in the same manner as a natural inducer. Japanese Journal of Physiology 44, suppl. 1, S158.
Inazawa, T., Shigemoto, T. & Takahashi, K. (1992). Effect of bFGF and protein tyrosine kinase inhibitors on the differentiation of ascidian ectodermal blastomeres. Japanese Journal of Physiology 42 (suppl.), S120.
Kengaku, M. & Okamoto, H. (1993). Basic fibroblast growth factor induces differentiation of neural tube and neural crest lineages of cultured ectoderm cells from Xenopus gastrula. Development 119, 1067-1078	[Abstract]
Kengaku, M. & Okamoto, H. (1995). bFGF as a possible morphogen for anteroposterior axis of the central nervous system in Xenopus. Development 121, 3121-3130	[Abstract]
Klein, R., Jing, S., Nanduri, V., O'Rourke, E. & Barbacid, M. (1991). The trk proto-oncogene encodes a receptor for nerve growth factor. Cell 65, 189-197	[Medline]
Knöchel, W., Born, J., Hoppe, P., Loppnow-Blinde, B., Tiedemann, H. & Tiedemann, H. (1987). Mesoderm inducing factors. Their possible relationship to heparin-binding growth factors and transforming growth factor-b. Naturwissenschaften 74, 604-606.	[Medline]
Lamb, T. M. & Harland, R. M. (1995). Fibroblast growth factor is a direct neural inducer, which combined with noggin generates anterior-posterior neural pattern. Development 121, 3627-3636	[Abstract]
Launay, C., Fromentoux, V., Shi, D. L. & Boucaut, J. C. (1996). A truncated FGF receptor blocks neural induction by endogenous Xenopus inducers. Development 122, 869-880	[Abstract]
Lee, P. L., Johnson, D. E., Cousens, L. S., Fried, V. A. & Williams, L. T. (1989). Purification and complementary DNA cloning of a receptor for basic fibroblast growth factor. Science 245, 57-60	[Medline]
Makabe, K. W., Fujiwara, S., Saiga, H. & Satoh, N. (1990). Specific expression of myosin heavy chain gene in muscle lineage cells of the ascidian embryo. Roux's Archives of Developmental Biology 199, 307-313.
Nakatani, Y., Yasuo, H., Satoh, N. & Nishida, H. (1996). Basic fibroblast growth factor induces notochord formation and the expression of As-T, a Brachyury homolog, during ascidian embryogenesis. Development 122, 2023-2031	[Medline]
Nishida, H. & Satoh, N. (1983). Cell lineage analysis in ascidian embryos by intracellular injection of a tracer enzyme. I. Up to the eight-cell stage. Developmental Biology 99, 382-394	[Medline]
Okada, T., Hirano, H., Takahashi, K. & Okamura, Y. (1997). Distinct neuronal lineages of the ascidian embryo revealed by expression of a sodium channel gene. Developmental Biology 190, 257-272	[Medline]
Okado, H. & Takahashi, K. (1988). A simple 'neural induction' model with two interacting cleavage-arrested ascidian blastomeres. Proceedings of the National Academy of Sciences of the USA 85, 6197-6201	[Medline]
Okado, H. & Takahashi, K. (1990a). Differentiation of membrane excitability in isolated cleavage-arrested blastomeres from early ascidian embryos. The Journal of Physiology 427, 583-602	[Abstract]
Okado, H. & Takahashi, K. (1990b). Induced neural-type differentiation in the cleavage-arrested blastomere isolated from early ascidian embryos. The Journal of Physiology 427, 603-623	[Abstract]
Okado, H. & Takahashi, K. (1993). Neural differentiation in cleavage-arrested ascidian blastomeres induced by a proteolytic enzyme. The Journal of Physiology 463, 269-290	[Abstract]
Okamura, Y., Okado, H. & Takahashi, K. (1993). The ascidian embryo as a prototype of vertebrate neurogenesis. BioEssays 15, 723-730	[Medline]
Okamura, Y., Ono, F., Okagaki, R., Chong, J. A. & Mandel, G. (1994). Neural expression of a sodium channel gene requires cell-specific interactions. Neuron 13, 937-948	[Medline]
Okamura, Y. & Shidara, M. (1987). Kinetic differences between Na channels in the egg and in the neurally differentiated blastomere in the tunicate. Proceedings of the National Academy of Sciences of the USA 84, 8702-8706	[Medline]
Okamura, Y. & Shidara, M. (1990a). Changes in sodium channels during neural differentiation in the isolated blastomere of the ascidian embryo. The Journal of Physiology 431, 39-74	[Abstract]
Okamura, Y. & Shidara, M. (1990b). Inactivation kinetics of the sodium channel in the egg and the isolated, neurally differentiated blastomere of the ascidian. The Journal of Physiology 431, 75-102	[Abstract]
Okamura, Y. & Takahashi, K. (1993). Neural induction suppress early expression of the inward-rectifier K+ channel in the ascidian blastomere. The Journal of Physiology 463, 245-268	[Abstract]
Reverberi, G. & Minganti, A. (1946). Fenomeni di evocazione nello sviluppo dell'uovo di Ascidie. Risultati dell'indagine sperimentale sull'uovo di Ascidiella aspersa e di Ascidiella malaca allo stadio di 8 blastomeri. Pubblicazioni della Stazione Zoologica di Napoli 20, 199-252.
Rosa, F., Roberts, A. B., Danielpour, D., Dart, L. L., Sporn, M. B. & Dawid, I. B. (1988). Mesoderm induction in amphibians: The role of TGF-2-like factors. Science 239, 783-785	[Medline]
Saitoe, M., Inazawa, T. & Takahashi, K. (1996). Neuronal expression in cleavage-arrested ascidian blastomeres requires gap junctional uncoupling from neighbouring cells. The Journal of Physiology 491, 825-842	[Abstract]
Schlessinger, J. & Ullrich, A. (1992). Growth factor signaling by receptor tyrosine kinases. Neuron 9, 383-391	[Medline]
Slack, J. M., Darlington, B. G., Heath, J. K. & Godsave, S. F. (1987). Mesoderm induction in early Xenopus embryos by heparin-binding growth factors. Nature 326, 197-200	[Medline]
Slack, J. M. W. (1994). The inducer that never was. Nature 369, 279-280	[Medline]
Slack, J. M. W. & Tannahill, D. (1992). Mechanism of anteroposterior axis specification in vertebrates: Lessons from the amphibians. Development 114, 285-302	[Abstract]
Spemann, H. & Mangold, H. (1924). Über Induktion von Embryonenanlagen durch Implantation artfremder Organisatoren. Archiv für Mikrosckopishe Anatomie und Entwicklungsmechanik 100, 599-638.
Weiss, P. (1939). Principles of Development. Henry Holt and Company, New York.

Acknowledgements

Special thanks go to Dr Harumasa Okamoto at the National Institute of Bioscience and Human Technology, for his valuable advice on this study. Thanks go to Professor Yoshiaki Kidokoro at Gunma University and Dr Naohide Yamashita at University of Tokyo for their useful comments on this manuscript. This study was partly supported by a grant-in aid for scientific research from the Ministry of Education, Science, Sports and Culture, Japan.

Corresponding author

K. Takahashi: Department of Medical Physiology, Meiji College of Pharmacy, Nozawa 1-35-23, Setagaya-ku, Tokyo 154-0003, Japan.

Email: kunitaro{at}my-pharm.ac.jp

This article has been cited by other articles:

				M. Tanaka-Kunishima, K. Takahashi, and F. Watanabe Cell contact induces multiple types of electrical excitability from ascidian two-cell embryos that are cleavage arrested and contain all cell fate determinants Am J Physiol Regulatory Integrative Comp Physiol, November 1, 2007; 293(5): R1976 - R1996. [Abstract] [Full Text] [PDF]

				C. D. Stern Neural induction: old problem, new findings, yet more questions Development, May 1, 2005; 132(9): 2007 - 2021. [Abstract] [Full Text] [PDF]

				S. Wada and H. Saiga HrzicN, a new Zic family gene of ascidians, plays essential roles in the neural tube and notochord development Development, March 14, 2003; 129(24): 5597 - 5608. [Abstract] [Full Text] [PDF]

				C. Hudson, S. Darras, D. Caillol, H. Yasuo, and P. Lemaire A conserved role for the MEK signalling pathway in neural tissue specification and posteriorisation in the invertebrate chordate, the ascidian Ciona intestinalis Development, January 1, 2003; 130(1): 147 - 159. [Abstract] [Full Text] [PDF]

				F. Ono, Y. Katsuyama, K. Nakajo, and Y. Okamura Subfamily-Specific Posttranscriptional Mechanism Underlies K+ Channel Expression in a Developing Neuronal Blastomere J. Neurosci., August 15, 1999; 19(16): 6874 - 6886. [Abstract] [Full Text] [PDF]

				M. Tanaka-Kunishima and K. Takahashi Cleavage-arrested cell triplets from ascidian embryo differentiate into three cell types depending on cell combination and contact timing J. Physiol., April 1, 2002; 540(1): 153 - 176. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Inazawa, T. Articles by Takahashi, K. Search for Related Content PubMed PubMed Citation Articles by Inazawa, T. Articles by Takahashi, K.