Mechanisms of Development 96 (2000) 155±163
www.elsevier.com/locate/modo
The expression of nonchordate deuterostome Brachyury genes in the
ascidian Ciona embryo can promote the differentiation
of extra notochord cells
Gouki Satoh*, Yoshito Harada 1, Nori Satoh
Department of Zoology, Graduate School of Science, Kyoto University, Sakyo-ku, Kyoto 606-8502, Japan
Received 19 October 1999; received in revised form 8 May 2000; accepted 16 June 2000
Abstract
The notochord is a structure present in all chordates and its development requires the transcription of Brachyury. While previous studies
have shown that Brachyury is essential for notochord formation in vertebrate embryos, this gene is also expressed during the embryogenesis
of nonchordate deuterostomes, hemichordates and echinoderms. Here we report that nonchordate deuterostome Brachyury genes can trigger
the differentiation of extra notochord cells when these genes are ectopically expressed in ascidian embryos. The 2.6 kb upstream region of
fork head gene (Ci-fkh) of Ciona intestinalis promotes the tissue-speci®c expression of a reporter gene in endoderm, notochord and nerve
cord. By taking advantage of this promoter, we misexpressed the Brachyury gene of ascidian (Ci-Bra), cephalochordate amphioxus
(Am(Bb)Bra2), hemichordate acorn worm (PfBra), and echinoderm sea urchin (SpBra), in Ciona embryos. The misexpression of not
only the chordate Brachyury, but also the nonchordate deuterostome Brachyury, resulted in the transformation of presumptive endodermal
cells into notochord cells. This was con®rmed by in situ hybridization experiments using four different notochord-speci®c probes from Ciona
that have different temporal expression patterns. RT-PCR analyses indicated that Ci-Bra was not upregulated by the product of Am(Bb)Bra2,
PfBra or SpBra. In situ hybridization showed no ectopic expression of Ci-Bra in the manipulated embryos. These results suggest that the
introduction of nonchordate deuterostome Brachyury genes into ascidian embryos can trigger the differentiation of notochord cells in
ascidian embryos. Evolutionary alteration in the genetic circuitry, especially downstream of this transcription factor, seems critical for
the evolution of notochord and chordate body plan. q 2000 Elsevier Science Ireland Ltd. All rights reserved.
Keywords: Brachyury; Notochord differentiation; Ascidian; Amphioxus; Hemichordate; Echinoderm; Functional conservation; Chordate evolution
1. Introduction
The Brachyury (Bra) gene was ®rst discovered in mutagenized mice (reviewed by Herrmann and Kispert, 1994),
and the gene was cloned by Herrmann et al. (1990). Bra
homologs were later identi®ed in frogs (Smith et al., 1991),
zebra®sh (Schulte-Merker et al., 1992) and chick (Kispert et
al., 1995). It was later revealed that Brachyury encodes a
sequence-speci®c transcriptional activator that contains a T
DNA-binding domain (Kispert and Herrmann, 1993; MuÈller
and Herrmann, 1997; reviewed by Herrmann and Kispert,
1994; Smith, 1997, 1999; Papaioannou and Silver, 1998),
and that the expression of this gene is required for posterior
mesoderm formation and notochord differentiation in the
* Corresponding author. Tel.: 181-75-753-4095; fax: 181-75-705-1113.
E-mail address: [email protected] (G. Satoh).
1
Present address: Kuroda Chiromorphology Project, ERATO, Komaba
4-7-6, Meguro-ku, Tokyo 153-0041, Japan.
various vertebrates (Rashbass et al., 1991; Schulte-Merker
et al., 1994; Conlon et al., 1996).
Ascidians are members of the Urochordata, one of the
three basic taxonomic groups within the phylum Chordata
(vertebrates and cephalochordates are the other two), and
the ascidian tadpole is one of the most simpli®ed and primitive chordate body plans (reviewed by Satoh, 1994, 1999;
Satoh and Jeffery, 1995; Di Gregorio and Levine, 1998).
The tail of the ascidian tadpole contains a notochord
comprised of forty cells and the notochord cell lineage has
been previously described (Nishida, 1987). Previous studies
using two ascidian species, Halocynthia roretzi (Yasuo and
Satoh, 1993) and Ciona intestinalis (Corbo et al., 1997a)
revealed that Brachyury (designated as As-T for the former
species and Ci-Bra for the latter species) is expressed exclusively in the notochord precursor cells (Figs. 1 and 2B), and
that the timing of gene expression coincides with the clonal
restriction of the notochord lineages. In H. roretzi, notochord formation is induced at the 32-cell stage by signals
0925-4773/00/$ - see front matter q 2000 Elsevier Science Ireland Ltd. All rights reserved.
PII: S 0925-477 3(00)00395-6
156
G. Satoh et al. / Mechanisms of Development 96 (2000) 155±163
Fig. 1. (A) Alignment of amino acid sequences including T-domain of Brachyury gene products of ascidians (Ci-Bra), amphioxus (Am(Bb)Bra2), acorn worm
(PfBra) and sea urchin (SpBra). Asterisks indicate identical residues among the four, while dots indicate conservative changes. The T-domain is boxed. (B) A
molecular phylogenetic analysis based on comparison of amino acid sequences of the T-domain showing that the amphioxus Am(Bb)Bra2, ascidian Ci-Bra,
acorn worm PfBra, and sea urchin SpBra are members of the Brachyury subfamily of the T-box gene family. The tree was constructed by the neighbor-joining
method (Saitou and Nei, 1987). Branch length is proportional to the number of amino acid substitutions; the scale bar indicates 0.1 amino acid substitutions per
position in the sequence. The grouping of deuterostome Brachyury genes is supported in 100% bootstrap pseudoreplications (Felsenstein, 1985).
emanating from the adjacent endoderm (Nakatani and
Nishida, 1994). Overexpression of As-T results in notochord
formation without the requirement for this inductive event
at the 32-cell stage (Yasuo and Satoh, 1998). Ectopic
expression of As-T (Yasuo and Satoh, 1998) or Ci-Bra
(Takahashi et al., 1999a) causes the transformation of
presumptive endoderm cells into notochord cells. Therefore,
the ascidian Brachyury is a key regulatory gene responsible
for notochord formation.
In addition, the Brachyury gene has been cloned in two
cephalochordate species: AmBra1 and AmBra2 from Branchiostoma ¯oridae (Holland et al., 1995) and Am(Bb)Bra1
and Am(Bb)Bra2 from B. belcheri (Fig. 1) (Terazawa and
Satoh, 1997). The results suggest that Brachyury gene dupli-
G. Satoh et al. / Mechanisms of Development 96 (2000) 155±163
cated independently in the lineage leading to amphioxus,
and the expression pattern of these genes is the same
(Holland et al., 1995; Terazawa and Satoh, 1997). For
example the expression pattern of Am(Bb)Bra2 is similar
to that of vertebrate Brachyury (Fig. 2A), although function
of the amphioxus gene remains to be determined.
Furthermore, Brachyury has been cloned in several
nonchordate deuterostome species (Fig. 1): a hemichordate,
Ptychodera ¯ava and three echinoderms, Hemicentrotus
pulcherrimus and Strongylocentrotus purpuratus and Asterina pectinifera. In P. ¯ava, PfBra is ®rst expressed at the
gastrula stage at the site of invagination and later in early
tornaria development where the stomodeum invaginates
(Fig. 2C) (Tagawa et al., 1998). PfBra transcripts disappear
during late larval development but this gene is expressed
again in mesoderm cells of the adult proboscis, collar and in
the very posterior region of the trunk (Peterson et al.,
1999a). Sea urchin Brachyury genes have been cloned
from H. pulcherrimus (HpTa) (Harada et al., 1995) and S.
purpuratus (SpBra) (Peterson et al., 1999b). These genes are
both expressed in secondary mesenchyme founder cells of
gastrulae (Fig. 2D). When the expression of SpBra was
examined during later stages it was detected in the vestibule
of the adult rudiment and in the mesoderm of both the left
157
and right hydrocoels (Peterson et al., 1999b). In addition,
the star®sh Brachyury (ApBra) is expressed where the archenteron undergoes invagination and in the stomodeum invagination region of the gastrula and early bipinnaria larva
stages (Shoguchi et al., 1999). Therefore, based on expression patterns in these species, nonchordate deuterostome
Brachyury genes show an evolutionarily conserved expression pattern, although the function(s) of these genes remains
to be discovered.
As shown in Fig. 1B, a molecular phylogenetic analysis
supports the idea that these deuterostome Brachyury genes
are orthologous. Therefore, regarding molecular mechanisms underlying the evolution of chordate body plan, an
important question to be answered is whether the nonchordate deuterostome Brachyury has a developmental potential
to promote the differentiation of notochord cells when the
gene is misexpressed in chordate embryos. If it is the case, a
possible evolutionary origin of notochord may be investigated by comparative studies, between chordate and
nonchordate deuterostome embryos, of Brachyury-mediated
differentiation mechanisms, especially its downstream
genes. In the present study we addressed this question.
2. Results
2.1. Ectopic expression of the ascidian Ci-Bra caused the
differentiation of extra notochord cells in Ciona embryos
Fig. 2. Expression of deuterostome Brachyury genes. (A) The cephalochordate amphioxus Am(Bb)Bra2 is expressed in the notochord (not), tailbud
(tb) and paraxial mesoderm (mes) of the neurula (From Terazawa and
Satoh, 1997; with permission). (B) The urochordate ascidian Brachyury
(As-T) is expressed in the notochord precursor cells (not) of the 110-cell
embryo (Yasuo and Satoh, 1993; with permission). (C) The hemichordate
acorn worm PfBra is expressed in the archenteron invaginating region (ar)
and stomodeum invaginating region (st) of the gastrula (Tagawa et al.,
1998; with permission). (D) The echinoderm sea urchin Brachyury
(HpTa) is expressed in the secondary mesenchyme founder cells (smc) of
the gastrula (Harada et al., 1995; with permission).
To determine whether the nonchordate deuterostome
Brachyury gene can trigger the differentiation of extra notochord cells when it is misexpressed in chordate embryos, we
took advantage of the promoter of Ciona fork head gene (Cifkh) (Corbo et al., 1997b) as described below. The Ci-fkh
gene is expressed in the endoderm, endodermal strand, notochord, and ventral ependymal cells of the neural tube or
nerve cord. The endodermal tissue of an ascidian larva is
usually subdivided into endoderm situated in the trunk
region and endodermal strand located in the tail region.
Because this classi®cation is rather arbitrary, we refer to
the cells comprising these structures as `endoderm' hereafter. A 2.6 kb genomic DNA fragment located in the 5'
¯anking region of Ci-fkh gene was found to be suf®cient to
drive the tissue-speci®c expression of a lacZ reporter gene
in these tissues after these constructs were electroporated
into 1-cell embryos (data not shown) (Corbo et al., 1997b;
Takahashi et al., 1999a). The Ci-Bra gene can be misexpressed in larval tissues by attaching the Ci-Bra coding
sequence to the Ci-fkh promoter region (Ci-fkh/Ci-Bra)
(Takahashi et al., 1999a). The misexpression of this gene
caused an ectopic differentiation of notochord cells in the
presumptive endoderm region and less frequently in the
nerve cord region. The differentiation of normal and extra
notochord cells was assessed by in situ hybridization using
four probes speci®c for different notochord-speci®c genes
(see Section 4). Fig. 3A±E shows the results with the 210d
158
G. Satoh et al. / Mechanisms of Development 96 (2000) 155±163
gene probe and Fig. 3F±I shows the results with the 309h
gene probe.
The ascidian larval notochord consists of exactly 40 cells
aligned in a single row of cells along the midline of the tail
(Fig. 3A). The differentiation of extra notochord cells was
easily assessed by observing additional cells that expressed
notochord-speci®c genes (Fig. 3B,F). In addition, it was
observed in approximately a half of the embryos that the
extra notochord cells were aligned into short rows to form a
partial notochord (Fig. 3B,F). These ectopic notochords
were situated in the endodermal region and at lower
frequencies in the nerve cord region (Fig. 3B). As summarized in Table 1, in approximately 80% of the electroporated
embryos, there were extra notochord cells, con®rming a
previous study (Takahashi et al., 1999a). These results indicate that presumptive endoderm and nerve chord cells can
be converted into notochord cells when the Ci-Bra gene is
misexpressed in these cells.
2.2. Misexpression of the amphioxus Am(Bb)Bra2 caused
the differentiation of extra notochords in Ciona embryos
The Am(Bb)Bra2 gene was misexpressed in ascidian
embryos by the electroporation of the Ci-fkh/Am(Bb)Bra2
construct into Ciona eggs. The misexpression of this
construct triggered the differentiation of extra notochord
cells within the ventral region of the embryos and the formation of short notochords was evident in about a half of the
embryo (Fig. 3C,G). A total of 143 eggs that were electroporated with Am(Bb)Bra2 were examined for the expression
of four different notochord-speci®c genes after their development into tailbud-stage embryos (Table 1). Differentiation of extra notochord cells was detected in approximately
82% of the embryos. Therefore, similar to the results using
Ci-Bra, the misexpression of Am(Bb)Bra2 in the ascidian
embryonic endodermal region triggered not only the differentiation of notochord cells but also the formation of a
morphologically normal notochord.
2.3. Misexpression of the hemichordate PfBra gene caused
the differentiation of extra notochord cells in Ciona embryos
Similar to Ci-Bra and Am(Bb)Bra2, we misexpressed the
hemichordate acorn worm PfBra gene in ascidian embryos
by electroporation of Ci-fkh/PfBra constructs into Ciona
eggs. The in situ hybridization shown in Fig. 4A shows
the misexpression of PfBra in the endodermal region of
the embryo. The expression of PfBra in endoderm cells
converted them into extra notochord cells (Fig. 3D,H). In
a few cases, short morphologically normal notochords were
evident (Fig. 3D). In contrast to the high frequency of extra
notochord cell formation by electroporation of Ci-Bra and
Am(Bb)Bra2 constructs, the frequency of extra notochord
Fig. 3. Differentiation of extra notochord cells by misexpression of chordate
and nonchordate deuterostome Brachyury genes in Ciona embryos visualized by whole-mount in situ hybridization with 210d probe (A±E) and
309h probe (F±I). (A) Formation of notochord in a control C. intestinalis
tailbud embryo. Asterisks indicate the notochord of Ciona embryo proper.
Tr, trunk; Tl, tail of the embryo. Scale bar, 100 mm for all panels. (B,F)
Differentiation of extra notochord cells (black arrowheads) caused by an
ectopic expression of the ascidian Ci-Bra with the aid of Ci-fkh/Ci-Bra
vector electroporation. (C,G) Differentiation of extra notochord cells
(black arrowheads) caused by misexpression of Am(Bb)Bra2. (D,H) Differentiation of extra notochord cells (black arrowheads) caused by misexpression of PfBra. (E,I) Differentiation of extra notochord cells (black
arrowheads) caused by misexpression of SpBra.
G. Satoh et al. / Mechanisms of Development 96 (2000) 155±163
159
Table 1
Induction of ectopic notochord cells in ascidian embryos injected with Brachyury gene from four deuterostomes
No. of embryos with ectopic notochord cells / No. of embryos injected
Brachyury gene
Probes
Ascidian (Ciona intestinalis)
Ci-Bra
Amphioxus (Branchiostoma berlcheri)
Am(Bb)Bra2
Acorn worm (Ptychodera ¯ava)
PfBra
Sea urchin (Strongylocentrotus purpuratus)
SpBra
210d
309h
502a
28/32 (87.5%)
34/48 (70.8%)
12/18 (66.7%)
74/98 (75.5%)
32/44 (72.7%)
59/66 (89.4%)
26/33 (78.8%)
117/143 (81.8%)
13/54 (24.1%)
12/41 (29.3%)
11/38 (28.9%)
36/133 (27.1%)
9/28 (32.1%)
8/29 (27.6%)
10/44 (22.7%)
27/101 (26.7%)
cell formation using PfBra constructs was lower (only 27%
as shown in Table 1).
Although the frequency was low, it is concluded that the
hemichordate Brachyury gene can trigger the differentiation
of extra notochord cells when expressed in ascidian
embryos.
2.4. The misexpression of the echinoderm SpBra gene
promoted the differentiation of extra notochord cells in
Ciona embryos
We also examined the expression of an echinoderm sea
urchin Brachyury gene (SpBra) in ascidian embryos. Electroporation of the fusion gene construct Ci-fkh/SpBra
resulted in the misexpression of SpBra in endoderm cells
of the ascidian embryo (Fig. 4B). As is evident in Fig. 3E,I,
notochord-speci®c genes were expressed in the extra notochord cells of the embryo. However, in contrast to Ci-Bra,
Am(Bb)Bra2 and PfBra genes, embryos in which the SpBra
gene was misexpressed rarely developed a notochord rod
(Fig. 3E). The frequency in the occurrence of extra notochord cells triggered by SpBra expression was similar to
that of PfBra expression (Table 1).
The transformation of presumptive endoderm cells into
notochord cells was considerably lower in PfBra or SpBra
electroporated embryos compared to embryos containing
Ci-Bra or Am(Bb)Bra2 constructs. This may have been
caused by a lower ef®ciency in the incorporation of SpBra
or PfBra constructs into embryos or by an inef®cient trans-
Total
lation of PfBra and SpBra transcripts. We examined this
possibility by whole-mount in situ hybridization, and as
shown in Table 2, the frequency of embryos with PfBra
and SpBra expression was 84 and 75%, which was comparable to that of the Ci-Bra control (89%). This suggests that
the low ef®ciency of transformation was not caused by a low
rate of incorporation of the constructs. These results also
suggest that almost all of the embryos successfully electroporated with Ci-Bra exhibited transformations of cell fates,
because the ef®ciency of the construct incorporation (Table
2) was nearly the same as that of transformation (Table 1).
2.5. Ci-Bra expression was not upregulated in embryos
electroporated with Am(Bb)Bra2, PfBra or SpBra
As mentioned above, electroporation of not only cephalochordate Am(Bb)Bra2 but also hemichordate PfBra and
echinoderm SpBra resulted in ectopic differentiation of
notochord cells. The minimal length of the promoter
required for notochord-speci®c expression of As-T and CiBra was characterized (Corbo et al., 1997a; Takahashi et al.,
1999b). A sequence of approximately 290 bp upstream to
the transcription start site of the As-T gene contains a putative Brachyury protein binding motif (Takahashi et al.,
1999b). Deletion of this motif from (2290)As-T/lacZ
construct resulted in the loss of reporter gene expression,
suggesting that As-T is upregulated by the gene product
Table 2
Ef®ciency of electroporation of Brachyury cDNA
Brachyury gene
Probes
No. of embryos with
Brachyury expression
No. of embryos injected
Fig. 4. Occurrence of mRNA of PfBra (A) or SpBra (B) in the endoderm
cells of Ciona embryos (arrowheads). Embryos developed from eggs electroporated with Ci-fkh/PfBra (A) or Ci-fkh/SpBra (B) construct were examined with PfBra or SpBra probe. Scale bar, 100 mm for all panels.
Ascidian
Ci-Bra
Amphioxus
Am(Bb)Bra2
Acorn worm
PfBra
Sea urchin
SpBra
Ci-Bra
93/105 (88.6%)
Not determined
PfBra
101/120 (84.2%)
SpBra
74/98 (75.5%)
160
G. Satoh et al. / Mechanisms of Development 96 (2000) 155±163
itself. In contrast, the promoter/enhancer region of the CiBra gene approximately 450 bp in length does not contain a
Brachyury protein binding motif (Corbo et al., 1997a).
However, the possibility remained that the electroporation
of exogenous Am(Bb)Bra2, PfBra or SpBra may cause the
upregulation of the Ci-Bra gene, which in turn transforms
presumptive endoderm cells to notochord cells.
To test this possibility, we examined the amount of CiBra mRNA in the experimental embryos using RT-PCR
(see Section 4). As shown in Fig. 5, 18 cycles of RT-PCR
demonstrated the presence of large quantities of Ci-Bra
mRNA in Ci-Bra cDNA electroporated embryos. However,
the amount of Ci-Bra mRNA in embryos electroporated
with Am(Bb)Bra2, PfBra or SpBra remained at the same
level as in the control uninjected embryos. Ampli®cation
after 30 cycles yielded similar results (see Fig. 5).
We also performed in situ hybridization analysis to determine whether the expression of Ci-Bra is ectopically
induced in embryos that were misexpressed with Am(Bb)Bra2, PfBra or SpBra. Ectopic expression of Ci-Bra was not
observed in these embryos (data not shown). All of these
results suggest that the misexpression of exogenous Brachyury genes does not induce the upregulation and/or ectopic
expression of the Ci-Bra gene. Therefore, the transformation of presumptive endoderm cells into notochord cells in
the experimental embryos was caused by the activity of the
electroporated Brachyury genes.
3. Discussion
The Brachyury gene is essential for posterior mesoderm
formation and notochord differentiation in vertebrate
embryos (reviewed by Herrmann and Kispert, 1994;
Smith, 1997) as well as for notochord differentiation in
urochordate ascidian embryos (Yasuo and Satoh, 1998;
Takahashi et al., 1999a). The Brachyury gene has been
identi®ed in all of the deuterostome groups, including
cephalochordates (Holland et al., 1995; Terazawa and
Satoh, 1997), hemichordates (Tagawa et al., 1998; Peterson
et al., 1999a) and echinoderms (Harada et al., 1995; Peterson et al., 1999b; Shoguchi et al., 1999), and in certain
Fig. 5. RT-PCR analysis of the quantity of Ci-Bra transcripts in embryos
developed from eggs electroporated with Ci-Bra, Am(Bb)Bra2, PfBra or
SpBra. This method allows direct comparison of the quantity by the band
intensity. Ampli®cations by 18 cycles (upper) and 30 cycles (lower) show
that the quantity of Ci-Bra transcripts is similar between control embryos
and embryos misexpressed with Am(Bb)Bra2, PfBra or SpBra.
protostomes such as insects (Kispert et al., 1994; Singer et
al., 1996). Furthermore, a recent characterization of hydra
homolog of Brachyury suggests that the gene was present
prior to the diversi®cation of diploblastic and triploblastic
animals, and its occurrence is considerably deep in the
evolutionary history of metazoans (Technau and Bode,
1999). In the present study, we focused on the functional
conservation of Brachyury genes in chordate and nonchordate deuterostomes, and we have shown that the Brachyury
gene of nonchordate deuterostomes has the potential to
promote the differentiation of notochord cells when this
gene is misexpressed in ascidian embryos. This suggests
that the activity of the nonchordate deuterostome Brachyury
gene which encodes a transcription factor required for notochord formation is conserved. Future studies will determine
if the functional conservational Brachyury gene can be
extended back to protostomes and diploblasts.
As shown in the present study, electroporation of not only
cephalochordate Am(Bb)Bra2 but also hemichordate PfBra
and echinoderm SpBra caused the differentiation of notochord cells in ascidian embryos. However, the ef®ciency of
the notochord cell transformation was signi®cantly lower in
PfBra or SpBra electroporated embryos. In addition, misexpression of PfBra or SpBra rarely caused the formation of a
morphologically normal notochord rod. Misexpression of
the three exogenous Brachyury genes did not promote the
upregulation and/or ectopic expression of Ci-Bra, and therefore the transformation of presumptive endoderm cells into
notochord cells in experimental embryos is thought to be
due to the activity of the exogenous Brachyury genes, not
caused indirectly by transcribing endogenous Ci-Bra genes.
Fig. 1A shows an alignment of amino acid residues including the T domain of Brachyury gene products from representative species belonging to cephalochordates, ascidians,
hemichordates and echinoderms. The amino acid sequence
similarities between the gene products of Ci-Bra and
Am(Bb)Bra2, between those of Ci-Bra and PfBra, and
between those of Ci-Bra and SpBra are 60.3, 61.6, and
59.2%, respectively. Therefore, it appears that differences
in the amino acid sequences among the four animal groups
is not signi®cant enough to explain differences in the ability
of the different Brachyury genes to promote extra notochord
cell development, although there may be signi®cant threedimensional con®guration differences between these
proteins. The Brachyury protein consists of a T DNA-binding domain, usually located in the N-terminal half, and the
transcription activation domain in the C-terminal half (e.g.
Conlon et al., 1996). Therefore, it is possible to attribute
functional differences in Brachyury genes to differences in
the transcription activation domain.
Alternatively, this difference may be due to differences in
the trans and cis binding af®nities between the gene
products of Am(Bb)Bra2, PfBra and SpBra, and Ci-Bra
target downstream genes or co-factors. Namely, it has
been revealed that Ci-Bra activates directly or indirectly
nearly 40 target genes which are expressed in the notochord
G. Satoh et al. / Mechanisms of Development 96 (2000) 155±163
cells (Takahashi et al., 1999a) and that these Ci-Bra downstream genes appear to be involved in the formation of the
architecture of notochord (Hotta et al., 2000). If the af®nity
of hemichordate PfBra and sea urchin SpBra to bind ciselements of Ci-Bra downstream genes or co-factors is low,
and therefore PfBra and SpBra might activate target genes
in some cells but not in other cells, it would not promote the
formation of morphologically normal notochord in which
40 cells are aligned in a single ®le. In contrast, if the
Am(Bb)Bra2 downstream genes that are involved in the
formation of notochord are conserved among urochordates
and cephalochordates, the Am(Bb)Bra2 gene product might
have high af®nity to bind to cis-elements of Ci-Bra target
genes simultaneously, which in turn results in the formation
of morphologically normal notochord. As discussed below,
studies of Brachyury target genes are critical to reveal the
entire genetic mechanism required for making a notochord.
The developmental mechanisms underlying the evolutionary pathway from advanced invertebrates through
primitive chordates to vertebrates has been a subject of
extensive investigation and vigorous discussion for more
than a century (see Gee, 1996). The phylum Chordata
consists of three subphyla, Urochordata (tunicates), Cephalochordata (amphioxus) and Vertebrata (e.g. Brusca and
Brusca, 1990). These groups exhibit a notochord, a dorsal
nerve cord, and pharyngeal gill slits, which are hallmarks of
the chordate body plan. In addition, chordates are categorized as a superphyletic group of deuterostomes, together
with two other invertebrate groups, hemichordates and echinoderms, because they share several features, including
radial cleavage, the fate of the blastopore that does not
form a mouth, an enterocoelic coelom, and a tripartite
body plan (Brusca and Brusca, 1990; Willmer, 1990; Nielsen, 1995). Recent cladistic studies as well as molecular
phylogenetic studies support the idea that echinoderms,
hemichordates and chordates share a common ancestor
and form the monophyletic group of deuterostomes (Schaeffer, 1987; Wada and Satoh, 1994; Cameron et al., 2000).
Therefore, chordates originated from a common ancestor
more than 550 million years ago by organizing the morphological features characteristical this animal group as
mentioned above. Because the notochord is the most prominent feature of chordates, an elucidation of developmental
mechanisms underlying the notochord formation is essential
to understanding the evolution of chordate body plan (Satoh
and Jeffery, 1995; Di Gregorio and Levine, 1998).
The present results emphasize the critical role played by
the Brachyury gene and genes located downstream of
Brachyury in notochord development and demonstrate that
the Brachyury gene present in nonchordate deuterostomes
can be introduced into ascidian embryos and convert endoderm cells into notochord cells. Although the function of
Brachyury genes in nonchordate deuterostomes is not yet
understood, the present results support the idea that Brachyury genes have been co-opted for various functions during
evolution. When the different groups of deuterostomes
161
originated from a common ancestor, probably prior to or
during Cambrian explosion, the nonchordate deuterostome
Brachyury gene might act to induce the invagination in the
embryos and to form the adult mesoderm, and the same gene
functioned to form the notochord in the lineage leading to
chordates. We have already isolated cDNA clones for nearly
40 notochord-speci®c genes that are activated by Ci-Bra
(Takahashi et al., 1999a; Hotta et al., 1999, 2000). These
potential Ci-Bra downstream genes appear to encode a
broad spectrum of divergent proteins that are associated
with ascidian notochord formation. The same approach
towards the identi®cation of downstream genes of PfBra
or SpBra may allow us to compare Brachyury downstream
genes and/or Brachyury genetic circuits of various chordate
and nonchordate deuterostome species to help us understand
chordate evolution.
4. Experimental procedures
4.1. Construction of plasmid vectors
Plasmid vector Ci-fkh/lacZ was constructed by fusion of
about 2.6 kb 5' ¯anking region of Ci-fkh (the fork head/
HNF-3b gene of Ciona intestinalis) with a reporter gene
lacZ (Corbo et al., 1997b). Plasmid vectors Ci-fkh/Ci-Bra,
Ci-fkh/Am(Bb)Bra2, Ci-fkh/PfBra and Ci-fkh/SpBra were
constructs in which lacZ of Ci-fkh/lacZ was replaced by
cDNAs for Ci-Bra (Corbo et al., 1997a), Am(Bb)Bra2
(Terazawa and Satoh, 1997), PfBra (Tagawa et al., 1998),
and SpBra (Peterson et al., 1999b), respectively.
4.2. Ascidian eggs and electroporation of plasmid vectors
Eggs and sperm of C. intestinalis were obtained surgically from the gonoduct. After insemination, eggs were
dechorionated by immersing them in seawater that
contained 1% sodium thioglycorate (Wako Pure Chem.
Ind., Ltd., Osaka, Japan) and 0.05% actinase E (Kaken
Pharm. Co., Ltd., Tokyo, Japan). The plasmid vectors
were electroporated as described (Corbo et al., 1997a). Electroporated embryos were allowed to develop at 188C in
agar-coated dishes with Millipore-®ltered seawater
(MFSW) containing 50 mg/ml streptomycin sulfate. They
were ®xed at the tailbud stage for in situ hybridization.
4.3. Whole-mount in situ hybridization
cDNAs for four Ciona notochord-speci®c genes with
different temporal expression patterns were used to monitor
ectopic differentiation of notochord cells in Ciona embryos.
The 210d (Ci-Noto4) gene begins to be expressed at the
neural plate stage, and the expression is speci®c to notochord cells. The cDNA consists of 1472 nucleotides with a
predicted polypeptide of 307 amino acids having very weak
sequence similarity to myb-like protein (Hotta et al., 2000).
The 309h gene is expressed ®rst at the neurula stage, and its
162
G. Satoh et al. / Mechanisms of Development 96 (2000) 155±163
expression is speci®c to notochord cells. Determination of
the partial nucleotide sequence of this cDNA suggests that
309h encodes for a sulfate transporter protein (Hotta et al.,
unpublished). The 502a gene begins to be expressed at the
early tailbud stage and is expressed only in notochord
progenitor cells. Determination of the partial nucleotide
sequence of this cDNA suggests that 502a encodes for a
methionine adenosyltransferase protein (Hotta et al., unpublished). The 403e gene encodes for a Prickle LIM domain
protein (Hotta et al., 2000). This gene begins to be expressed
as early as the gastrula stage, about 2 h behind Ci-Bra
expression at the 64-cell stage. Because this transcript is
detected ®rst in notochord cells up to the neural plate
stage, but later its expression is also evident in the posterior
part of the central nervous system, we did not include data
obtained with this probe in Table 1. cDNAs for all four
genes were used to prepare DIG-labeled RNA probes
(Roche Diagnostics, Tokyo, Japan).
In situ hybridization was carried out essentially according
to the method described by Corbo et al. (1997a). Brie¯y,
embryos were ®xed in 4% paraformaldehyde in 0.1 M
MOPS buffer (pH 7.5), 0.5 M NaCl. After being thoroughly
washed with PBT (phosphate-buffered saline containing
0.1% Tween 20), the embryos were treated with 2 mg/ml
proteinase K (Sigma, St. Louis, MO, USA) in PBT for 30
min at 378C, then post-®xed with 2% paraformaldehyde in
PBT for 30 min at room temperature. After a 1.5-h period of
prehybridization at 428C, the embryos were allowed to
hybridize with the DIG-labeled probes at a concentration
of 1 mg/ml for at least 18 h at 428C. After hybridization,
the embryos were washed ten times for 15 min each with the
hybridization solution at 508C. Thereafter, the samples were
incubated for 2 h with 500 ml anti-DIG-alkaline phosphatase
conjugate, and color was developed according to the Roche
protocol. After dehydration, some of the specimens were
cleared by placing them in a 100% xylene. The samples
were mounted on glass slides with entellan neu (Merck,
Darmstadt, Germany) and photographed.
4.4. Quantitative reverse transcription-polymerase chain
reaction (RT-PCR)
RT-PCR was carried out according to the method
described by Satou (1999). Twenty embryos were lysed in
200 ml of GTC solution (4 M guanidinium thiocyanate, 50
mM Tris±HCl pH 7.5, 10 mM EDTA, 2% sarkosyl, 1%
mercaptoethanol) and the lysates were stored at 2808C.
Following phenol-chloroform extraction and isopropanol
precipitation, RNA was treated with 1 unit of RNase-free
DNase (Life Technologies, Inc., Rockville, MD, USA) for
15 min at room temperature and then with 0.2 mg/ml proteinase K for 30 min at 378C. After annealing with 10 pmol
oligo (dT), the total RNA was incubated with 200 units of
Superscript II reverse transcriptase (Life Technologies, Inc.)
for 50 min at 428C in total volume of 20 ml. One-tenth of the
reverse-transcribed cDNA was used as template for the
subsequent polymerase chain reaction (PCR). PCRs were
performed in total volume of 50 ml containing 0.2 mM
dNTPs, 1.5 mM MgCl2, 0.05 ml [ 32P]-dCTP, 100 pmol of
each primer, 1£ Taq buffer, and 1.25 units of Taq DNA
polymerase (Toyobo, Osaka, Japan). Fifteen microliters of
the reaction products was resolved on 6% non-denaturing
polyacrylamide gels and subjected to autoradiography. The
number of cycles of PCR reactions within the quantitative
range was determined empirically. The primers used in the
present study and the number of cycles were as follows:
Ci-Bra, 5'-GAACCCCACGTTTCGTCA-3' and 5'GAATGCGACGCAACGACT-3', 18 and 30 cycles.
Acknowledgements
We thank Michael Levine for Ci-fkh cassette and critical
reading of the manuscript, Eric H. Davidson for SpBra
cDNA, Kuni Tagawa for PfBra cDNA, Kohji Hotta and
Hiroki Takahashi for cDNAs for Ciona notochord-speci®c
genes, Yutaka Satou for technical suggestion, and William
R. Bates for comments on the manuscript. We thank all staff
members of Marine BioSource Education Center of Tohoku
University for their help in collection of C. intestinalis. This
research was supported by a Grant-in-Aid for Specially
Promoted Research (07102012) from the Monbusho,
Japan to NS and a grant from Human Frontier Science
Program (RG212/1997). GS was a Predoctoral Fellow and
YH was a Postdoctoral Fellow of JSPS, with the Monbusho
research grants 3577 and 8910, respectively.
References
Brusca, R.C., Brusca, G.J., 1990. Invertebrates, Sinauer Associates, Inc.
Sunderland, MA.
Cameron, C.B., Garey, J.R., Swalla, B.J., 2000. Evolution of the chordate
body plan: new insights from phylogenetic analyses of deuterostome
phyla. Proc. Natl. Acad. Sci. USA 97, 4469±4474.
Conlon, F.L., Sedgwick, S.G., Weston, K.M., Smith, J.C., 1996. Inhibition
of Xbra transcription activation causes defects in mesodermal patterning and reveals autoregulation of Xbra in dorsal mesoderm. Development 122, 2427±2435.
Corbo, J.C., Levine, M., Zeller, R.W., 1997a. Characterization of a notochord-speci®c enhancer from the Brachyury promoter region of the
ascidian, Ciona intestinalis. Development 124, 589±602.
Corbo, J.C., Erives, A., Di Gregorio, A., Chang, A., Levine, M., 1997b.
Dorsoventral patterning of the vertebrate neural tube is conserved in a
protochordate. Development 124, 2335±2344.
Di Gregorio, A., Levine, M., 1998. Ascidian embryogenesis and the origins
of the chordate body plan. Curr. Opin. Genet. Dev. 8, 457±463.
Felsenstein, J., 1985. Con®dence limits on phylogenies: an approach using
the bootstrap. Evolution 39, 783±791.
Gee, H., 1996. Before the Backbone. Views on the Origin of the Vertebrates, Chapman & Hall, London.
Harada, Y., Yasuo, H., Satoh, N., 1995. A sea urchin homologue of the
chordate Brachyury (T) gene is expressed in the secondary mesenchyme
founder cells. Development 121, 2747±2754.
Herrmann, B.G., Kispert, A., 1994. The T genes in embryogenesis. Trends
Genet. 10, 280±286.
Herrmann, B.G., Labeit, S., Poustka, A., King, T.R., Lehrach, H., 1990.
G. Satoh et al. / Mechanisms of Development 96 (2000) 155±163
Cloning of the T gene required in mesoderm formation in the mouse.
Nature 343, 617±622.
Holland, P.W.H., Koschorz, B., Holland, L.Z., Herrmann, B.G., 1995.
Conservation of Brachyury (T) genes in amphioxus and vertebrates:
developmental and evolutionary implications. Development 121,
4283±4291.
Hotta, K., Takahashi, H., Erives, A., Levine, M., Satoh, N., 1999. Temporal
expression patterns of 39 Brachyury-downstream genes associated with
notochord formation in the Ciona intestinalis embryo. Dev. Growth
Differ. 41, 657±664.
Hotta, K., Takahashi, H., Asakura, T., Saitoh, B., Takatori, N., Satou, Y.,
Satoh, N., 2000. Characterization of Brachyury-downstream notochord
genes in the Ciona intestinalis embryo. Dev. Biol. in press.
Kispert, A., Herrmann, B.G., 1993. The Brachyury gene encodes a novel
DNA binding protein. EMBO J. 12, 3211±3220.
Kispert, A., Herrmann, B.G., Leptin, M., Reuter, R., 1994. Homologs of the
mouse Brachyury gene are involved in the speci®cation of posterior
terminal structures in Drosophila, Tribolium, and Locusta. Genes
Dev. 8, 2137±2150.
Kispert, A., Koschorz, B., Herrmann, B.G., 1995. The T protein encoded by
Brachyury is a tissue-speci®c transcription factor. EMBO J. 14, 4763±
4772.
MuÈller, C.W., Herrmann, B.G., 1997. Crystallographic structure of the T
domain-DNA complex of the Brachyury transcription factor. Nature
389, 884±888.
Nakatani, Y., Nishida, H., 1994. Induction of notochord during ascidian
embryogenesis. Dev. Biol. 166, 289±299.
Nielsen, C., 1995. Animal Evolution. Interrelationships of the Living Phyla,
Oxford Univ. Press, Oxford.
Nishida, H., 1987. Cell lineage analysis in ascidian embryos by intracellular
injection of a tracer enzyme. III. Up to the tissue restricted stage. Dev.
Biol. 121, 526±541.
Papaioannou, V.E., Silver, L.M., 1998. The T-box gene family. BioEssays
20, 9±19.
Peterson, K.J., Cameron, R.A., Tagawa, K., Satoh, N., Davidson, E.H.,
1999a. A comparative molecular approach to mesodermal patterning
in basal deuterostomes: the expression pattern of Brachyury in the
enteropneust hemichordate Ptychodera ¯ava. Development 126, 85±95.
Peterson, K.J., Harada, Y., Cameron, R.A., Davidson, E.H., 1999b. Expression pattern of Brachyury and Not in the sea urchin: comparative implications for the origins of mesoderm in the basal deuterostomes. Dev.
Biol. 207, 419±431.
Rashbass, P., Cooke, L.A., Herrmann, B.G., Beddington, R.S.P., 1991. A
cell autonomous function of Brachyury in T/T embryonic stem cell
chimaeras. Nature 353, 348±351.
Saitou, N., Nei, M., 1987. The neighbor-joining method: a new method for
reconstructing phylogenetic trees. Mol. Biol. Evol. 4, 406±425.
Satoh, N., 1994. Developmental Biology of Ascidians, Cambridge Univ.
Press, New York.
Satoh, N., 1999. Cell fate determination in the ascidian embryo. In Cell
Lineage and Fate Determination. In: Moody, S.A. (Ed.). Academic
Press, pp. 59±74.
163
Satoh, N., Jeffery, W.R., 1995. Chasing tails in ascidians: developmental
insights into the origin and evolution of chordates. Trends Genet. 11,
354±359.
Satou, Y., 1999. posterior end mark 3 (pem-3), an ascidian maternally
expressed gene with localized mRNA encodes a protein with Caenorhabditis elegans MEX-3-like KH domains. Dev. Biol. 212, 337±350.
Schaeffer, B., 1987. Deuterostome monophyly and phylogeny. Evolutionary Biology 21, 179±235.
Schulte-Merker, S., Ho, R.K., Herrmann, B.G., NuÈsslein-Volhard, C., 1992.
The protein product of the zebra®sh homologue of the mouse T gene is
expressed in nuclei of the germ ring and the notochord of the early
embryo. Development 116, 1021±1032.
Schulte-Merker, S., van Eeden, F.J.M., Halpern, M.E., Kimmel, C.B.,
NuÈsslein-Volhard, C., 1994. no tail (ntl) is the zebra®sh homologue
of the mouse T (Brachyury) gene. Development 120, 1009±1015.
Shoguchi, E., Satoh, N., Maruyama, Y.K., 1999. Pattern of Brachyury gene
expression in star®sh embryos resembles that of hemichordate embryos
but not of sea urchin embryos. Mech. Dev. 82, 185±189.
Singer, J.B., Harbecke, R., Kusch, T., Reuter, R., Lengyel, J.A., 1996.
Drosophila brachyenteron regulates gene activity and morphogenesis
in the gut. Development 122, 3707±3718.
Smith, J., 1997. Brachyury and the T-box genes. Curr. Opin. Genet. Dev. 7,
474±480.
Smith, J., 1999. T-box genes: what they do and how they do it. Trends
Genet. 15, 154±158.
Smith, J.C., Price, B.M.J., Green, J.B.A., Weigel, D., Herrmann, B.G.,
1991. Expression of a Xenopus homolog of Brachyury (T ) is an
immediate-early response to mesoderm induction. Cell 67, 79±87.
Tagawa, K., Humphreys, T., Satoh, N., 1998. Novel pattern of Brachyury
gene expression in hemichordate embryos. Mech. Dev. 75, 139±143.
Takahashi, H., Hotta, K., Erives, A., Di Gregorio, Zeller, A., Levine, R.W.,
Satoh, N., 1999a. Brachyury downstream notochord differentiation in
the ascidian embryo. Genes Dev. 13, 1519±1523.
Takahashi, H., Mitani, Y., Satoh, G., Satoh, N., 1999b. Evolutionary alterations of the minimal promoter for notochord-speci®c Brachyury expression in ascidian embryos. Development 126, 3725±3734.
Technau, U., Bode, H.R., 1999. HyBra1, a Brachyury homologue, acts
during head formation in Hydra. Development 126, 999±1010.
Terazawa, K., Satoh, N., 1997. Formation of the chordamesoderm in the
amphioxus embryo: analysis with Brachyury and fork head/HNF-3
genes. Dev. Genes Evol. 207, 1±11.
Wada, H., Satoh, N., 1994. Details of the evolutionary history from invertebrates to vertebrates, as deduced from the sequences of 18S rDNA.
Proc. Natl. Acad. Sci. USA 91, 1801±1804.
Willmer, P., 1990. Invertebrate Relationships. Patterns in Animal Evolution, Cambridge Univ. Press, Cambridge.
Yasuo, H., Satoh, N., 1993. Function of vertebrate T gene. Nature 364, 582±
583.
Yasuo, H., Satoh, N., 1998. Conservation of the developmental role of
Brachyury in notochord formation in a urochordate, the ascidian Halocynthia roretzi. Dev. Biol. 200, 158±170.