Download Evolution of Brachyury proteins: identification of a novel regulatory

Available online at www.sciencedirect.com
R
Developmental Biology 260 (2003) 352–361
www.elsevier.com/locate/ydbio
Evolution of Brachyury proteins: identification of a novel regulatory
domain conserved within Bilateria
Sylvain Marcellini,a,b,1,* Ulrich Technau,c J.C. Smith,b and Patrick Lemairea,*
a
Laboratoire de Génétique et Physiologie du Développement, IBDM, CNRS/INSERM/Universite de la Méditerranée/AP de Marseille,
Parc Scientifique de Luminy, Case 907, F-13288, Marseille Cedex 9, France
b
Wellcome Trust/Cancer Research UK Institute of Cancer and Developmental Biology and Department of Zoology, University of Cambridge,
Tennis Court Road, Cambridge CB2 1QR, UK
c
Molekulare Zellbiologie, Zoologisches Institut, TU Darmstadt, Schnittspahnstrasse 10, D-64287 Darmstadt, Germany
Received for publication 28 October 2002, revised 28 March 2003, accepted 8 April 2003
Abstract
Orthologues of Brachyury, a subfamily of T-box transcription factors, specify distinct cell types in different metazoan phyla, suggesting
that the function of these genes has changed through the course of evolution. To investigate this evolutionary process, we have compared
the activities of Brachyury orthologues from all major phyla in a single cellular context, the pluripotent Xenopus laevis animal cap. In this
assay, an ancestral function is revealed: most orthologues, including the Hydra protein, mimic the action of endogenous Xenopus Brachyury,
in that they induce mesoderm but not endoderm. Orthologues from Drosophila and ascidians, however, display an additional derived
property, represented in our assay by the induction of endoderm. Misexpression of chimeric versions of Brachyury reveals that the
C-terminal half of the protein is important for the strength of the induced response but not for its specificity. In contrast, amino acids located
within the T-domain and in a short N-terminal peptide are involved in restricting the activity of Brachyury proteins to induction of mesoderm
and not endoderm. Possession of this N-terminal motif is correlated with early circumblastoporal expression of Brachyury orthologues. We
propose that restriction of Brachyury activity by this motif plays a conserved role in the control of Bilaterian gastrulation.
© 2003 Elsevier Science (USA). All rights reserved.
Keywords: Brachyury; Xenopus laevis; Hydra; Platynereis; Amphioxus; Drosophila; Ascidian; Vertebrate; Evolution; Mesoderm; Endoderm; Blastopore;
Gastrulation
Introduction
T-domain transcription factors play key roles during metazoan development and are implicated in several human congenital malformations (reviewed in Papaioannou and Silver,
1998; Smith, 1997). Orthologues of mouse Brachyury (or T)
(Herrmann et al., 1990) constitute a subfamily of T-domain
proteins with members in both diploblasts and triploblasts and
whose overall organisation has been conserved throughout
* Corresponding authors. Fax: ⫹33-4-9182-0682.
E-mail addresses: [email protected] (P. Lemaire), sm517@
hermes.cam.ac.uk (S. Marcellini).
1
Current address: Department of Zoology, University of Cambridge,
Downing Street, Cambridge CB2 3EJ, UK.
evolution. An N-terminal domain of variable length precedes a
highly conserved 180-amino-acid DNA-binding T-domain
(Kispert and Herrmann, 1993). The C-terminal half of the
protein has been shown to mediate transcriptional activation in
vertebrates (Conlon et al., 1996; Kispert et al., 1995) but is
only weakly conserved between phyla.
A consensual expression for Brachyury in the blastopore
and subsets of mesodermal cells in most metazoans has now
emerged (Technau, 2001). However, it is important to stress
that, in contrast to other evolutionary conserved gene families, such as Pax6 (Gehring and Ikeo, 1999) and Otx (Hirth
and Reichert, 1999), the Brachyury expression domain
shows great variability in different phyla. Thus, the gene is
expressed throughout the early mesoderm of vertebrate embryos (Schulte-Merker et al., 1992; Smith et al., 1991;
Wilkinson et al., 1990) but is restricted to the notochord of
0012-1606/03/$ – see front matter © 2003 Elsevier Science (USA). All rights reserved.
doi:10.1016/S0012-1606(03)00244-6
S. Marcellini et al. / Developmental Biology 260 (2003) 352–361
ascidians (Corbo et al., 1997; Yasuo and Satoh, 1993), the
presumptive endodermal gut of sea urchins (Gross and McClay, 2001), the ectodermal hindgut (Kispert et al., 1994)
and caudal visceral mesoderm of Drosophila (Kusch and
Reuter, 1999), the hind- and foregut of molluscs and annelids (Arendt et al., 2001; Lartillot et al., 2002), and the
gastrulating posterior pole of jellyfish (Spring et al., 2002).
In hydra and jellyfish polyps, expression is found in the
endoderm of the mouth anlage (Spring et al., 2002; Technau
and Bode, 1999). Hence, depending on the phylum,
Brachyury expression is found in ectoderm, mesoderm, or
endoderm.
Irrespective of the site of expression, loss-of-function
experiments in mouse, Xenopus, zebrafish, Drosophila,
ascidians, and sea urchins show that Brachyury is necessary for the specification of the cells in which it is
expressed (Conlon et al., 1996; Gross and McClay, 2001;
Kispert et al., 1994; Kusch and Reuter, 1999; Satou et al.,
2001; Schulte-Merker et al., 1994; Wilkinson et al.,
1990).
The Brachyury proteins thus provide a powerful paradigm to study how evolutionarily conserved genes come
to fulfill different functions across the animal kingdom.
The surprising diversity of the roles of Brachyury in
metazoans, and the availability of full-length cDNAs for
orthologues from all major phyla, prompted us to test
whether these proteins have intrinsically different activities, or whether their different functions reflect changes
in their cellular contexts during evolution. We therefore
compared the effects of overexpression of Brachyury
orthologues in a constant cellular context, the Xenopus
animal cap.
Xenopus animal cap cells normally differentiate as epidermis, but they can be induced to form mesodermal,
endodermal, or neural tissue upon addition of extracellular
or intracellular factors (Horb and Thomsen, 1997; Hudson
et al., 1997; Wilson and Hemmati-Brivanlou, 1995). In particular, misexpression of Xenopus or zebrafish Brachyury
(Xbra or no tail) causes animal pole cells to form mesoderm
(Cunliffe and Smith, 1992; O’Reilly et al., 1995), while
another Xenopus T-box factor, VegT (which is not a member of the Brachyury family), induces both mesoderm and
endoderm (Conlon et al., 2001; Horb and Thomsen, 1997).
This observation indicates that the animal cap assay can
discriminate between the activities of T-box factors. Furthermore, this assay allows a more rapid screening of the
activities of the expressed genes than transgenesis in mouse,
C. elegans, or Drosophila, three model systems previously
used to test the functional conservation of pairs of orthologues (Acampora et al., 2001; Haun et al., 1998; Hunter
and Kenyon, 1995; Nagao et al., 1998). Our experiments
have focused on the induction of mesoderm and endoderm,
which are the two major sites of Brachyury expression in
metazoans.
353
Materials and methods
Embryo manipulations and injections
Embryos were in vitro fertilized, dejellied, and cultivated
as described previously (Darras et al., 1997). Synthetic
mRNA was prepared and injected at the animal pole of
two-cell embryos as in Darras et al. (1997). Animal caps
were dissected between stages 8.5 and 9.5 and cultured in
1⫻ MBS containing 50 ␮g/ml gentamycin until stage 32–35
(Nieuwkoop and Faber, 1967). The origin of constructs for
the injected Brachyury orthologues cDNAs was as follows:
T from Mus musculus (Herrmann et al., 1990), Xbra from
Xenopus laevis (Smith et al., 1991), ntl from Danio rerio
(Schulte-Merker et al., 1992), Am(Bb)Bra1 from the amphioxus species Branchiostoma belcheri (Terazawa and Satoh, 1997), As-T from Halocynthia roretzi (Yasuo and Satoh, 1994), CiBra from Ciona intestinalis (M. Levine and A.
Di Gregorio, unpublished result), SpBra from the sea urchin
Strongylocentrotus purpuratus (Peterson et al., 1999b), PfBra from the hemichordate Ptychodera flava (Tagawa et al.,
1998), Trg from Drosophila melanogaster (Kispert et al.,
1994), PdBra from the polychaete annelid Platynereis
dumerilii (Arendt et al., 2001), and HyBra1 from Hydra
vulgaris (Technau and Bode, 1999). Unless otherwise indicated, cDNAs were subcloned in the indicated sites of the
following Xenopus expression vectors: pBS-RN3: Trg
(HindIII–EcoRI), CiBra (NotI), As-T (EcoRI, obtained from
Yasuo and Satoh, 1998), T (EcoRI), AmphiBra1 (EcoRI–
NotI), SpBra (BglII–EcoRI, gift from N. Satoh); pCS2:
HyBra1 (BamHI–StuI), PfBra (EcoRI–XhoI), PdBra
(BamHI–XhoI); pSP64T: ntl (BglII), Xbra (EcoRI).
In situ hybridisation and immunohistochemistry
Whole-mount in situ hybridisation with the endodermin
probe was performed as described previously (Lemaire et
al., 1998). Embryos and explants were bleached in a methanol solution containing 10% hydrogen peroxide before
counting positive samples. Immunohistochemistry with the
12/101 monoclonal antibody recognising Xenopus somitic
muscle (Kintner and Brockes, 1984, obtained from the Developmental Studies Hybridoma Bank) was as in Hopwood
et al. (1992), using Magenta Phos (BioSynth) as an alkaline
phosphatase substrate.
Western blotting
A FLAG epitope coding sequence was added by PCR
after the last codon of Xbra and As-T. The sequences 5⬘ to
3⬘ of the reverse oligonucleotides used are: AsT-FLAG, ctt
gtc atc gtc gtc ctt gta gtc gac tga tgg tgg cgc aac g;
Xbra-FLAG, ctt gtc atc gtc gtc ctt gta gtc aag tct caa att ctg
taa at, where the sequence coding for the FLAG is underlined. The stop codon and the NotI site were provided by the
pGEM-T cloning vector (Promega). Only the C-terminal
354
S. Marcellini et al. / Developmental Biology 260 (2003) 352–361
Fig. 1. Quantitative analysis of endoderm and striated muscle induction in response to misexpression of different Brachyury orthologues. Histogram showing
the percentage of animal caps positive for somitic muscle (grey) or endoderm (black) upon misexpression of the indicated Brachyury orthologues or of the
positive control VegT. Above each bar is shown the total number of tested explants. The range of nanograms of injected mRNA per embryo is indicated.
Dt., Deuterostomians; Pt., Protostomians; Cn, Cnidarians; H.r., Halocynthia roretzi; C.i., Ciona intestinalis.
domain was amplified, and subsequently cloned in frame
after the T-box (XbaI–NotI), in the pBS-RN3 expression
vector (see the “Chimeric constructs” section for details).
Both constructs were fully sequenced prior to use. mRNA
was injected at the two-cell stage, and four animal hemispheres were dissected at stage 7 and frozen together. The
Western blotting procedure was performed as described
previously (Tada et al., 1997). Half of an animal hemisphere
equivalent was loaded in each lane. Protein detection was
performed with a 1/16,000 dilution of an anti-FLAG antibody HRP-conjugate (Abcam), followed by a revelation
with the ECL Western blotting detection system (Amersham Biosciences).
Chimeric constructs
XH and HX chimeric constructs (see Fig. 4A) were made
by using the overlapping-PCR technique. They were subcloned between the StuI and XbaI sites of the pCS2 expression vector. The sequences 5⬘ to 3⬘ of the oligonucleotides
used to fuse the cDNAs at the junction are: HX-forward, aa
gga tct gat tat aaa gac atc ttg gat g; HX-reverse, gtc ttt ata
atc aga tcc ttc ttt tgc; XH-forward, aa aga aat gat cac aaa gat
att tta tgc g; XH-reverse, atc ttt gtg atc att tct ttc ttt tgc atc.
For XA and AX (see Fig. 4B), a silent XbaI site was
introduced in As-T and Xbra, within nucleotides coding for the
FLD amino acids in the fully conserved “FAKAFLDAKER”
motif at the end of the T-domain. The sequences 5⬘ to 3⬘ of
the oligonucleotides used to modify the cDNAs at the junction are: As-T-forward, aac gca aag gag cgt cct gat; As-Treverse, a aaa tgc ctt agc aaa agg att; Xbra-forward, aac gca
aaa gaa aga aat gat tat; Xbra-reverse, a aaa tgc ttt ggc aaa tgg
att; the XbaI site is underlined. The domains were subsequently swapped by using restriction enzymes and subcloned between the EcoRI and NotI sites of the pBS-RN3
expression vector. The fusions of the upstream N-terminal
domains (see Fig. 5B) were performed with the overlapping
PCR technique, and subcloned as described for XA and AX.
The sequences 5⬘ to 3⬘ of the oligonucleotides used to fuse
the As-T N-terminal domain to the Xbra T-box: forward,
cca tct gac agc gaa gtg aag gtt agc ctg gag gag cgg; reverse,
ctc ctc cag gct aac ctt cac ttc gct gtc aga tgg cga; and to fuse
the Xbra N-terminal domain to the As-T T-box: forward,
ccc acc gag aag gag ctg aga ttg act ctg aat gat cgt; reverse,
atc att cag agt caa tct cag ctc ctt ctc ggt ggg gtc. Sequences
of all constructs were verified.
Results
Synthetic Brachyury mRNAs from all major phyla (Fig.
1) were microinjected into Xenopus embryos at the two-cell
stage, animal caps were explanted during blastula stages,
and were cultured in isolation until the equivalent of tailbud
stage (St. 32–35). To ensure that results were not affected
by dose-dependent effects of these T-box factors (O’Reilly
S. Marcellini et al. / Developmental Biology 260 (2003) 352–361
355
Fig. 2. The Brachyury proteins induce different types of terminal differentiation in the Xenopus laevis animal cap assay. Dissected animal caps were cultured
until the equivalent of the tailbud stage and assayed for the presence of endoderm (dark blue staining) and subsequently for somitic muscle (purple staining).
Results are shown for a control embryo (A), and for animal caps dissected from uninjected embryos (B), or from embryos injected with the following doses
of synthetic mRNAs: (C) 0.2 ng of VegT mRNA; (D) 1.7 ng of Xbra mRNA; (E) 3.3 ng of Xbra mRNA; (F) 3.4 ng of ntl mRNA; (G) 1 ng of T mRNA;
(H) 1 ng of AmBra1 mRNA; (I) 0.2 ng of As-T mRNA; (J) 0.06 ng of CiBra mRNA; (K) 1 ng of SpBra mRNA; (L) 1 ng of PfBra mRNA; (M) 1 ng of
PdBra mRNA; (N) 0.3 ng of Trg mRNA; (O) 6 ng of HyBra1 mRNA. Scale bars represent 400 ␮m in (A), and 200 ␮m in (B–O).
et al., 1995), a range of mRNA concentrations for each
Brachyury orthologue was injected, from an amount which
had no detectable effect, to a dose at which cell viability was
reduced. Induction of endoderm was revealed by expression
of the pan-endodermal gene endodermin (Sasai et al., 1996)
and induction of striated muscle, the major mesodermal cell
type, was detected by immunocytochemistry using the antibody 12/101 (Kintner and Brockes, 1984) (Fig. 2A).
Control uninjected animal caps formed atypical epidermis and showed no staining (Figs. 1 and 2B). As reported
previously, animal caps dissected from embryos injected
with low doses of Xbra mRNA developed a vesicular morphology indicative of the presence of ventral mesoderm,
such as mesothelial smooth muscle and mesenchyme, as
well as small patches of striated muscle (Cunliffe and
Smith, 1992; O’Reilly et al., 1995) (Fig. 2D). Injection of
higher doses of Xbra mRNA led to the induction of more
muscle-positive cells and no ventral mesoderm was detectable (Fig. 2E) (O’Reilly et al., 1995). Endoderm was never
detected in response to Xbra (Figs. 1, and 2D and E). In
contrast to Brachyury, and as reported by Horb and Thomsen (1997), injection of VegT mRNA induced mesoderm at
low concentrations and endoderm at higher doses (Figs. 1
and 2C, and data not shown).
All Brachyury orthologues tested in this assay induced
striated muscle in a dose-dependent fashion (Figs. 1 and 2),
although the percentage of muscle-positive caps at the highest nontoxic dose varied, with the diploblast Brachyury,
HyBra1, displaying the lowest muscle-inducing activity
(11.5% of analysed explants) (Figs. 1 and 2O).
Although all Brachyury orthologues were able to induce
mesoderm, most were unable to induce endoderm, thus
mimicking the endogenous Xbra activity (Figs. 1 and
2D–H, K–M, and O). The low endoderm induction activity
displayed by the mouse and hemichordate orthologues was
not reproducible, and was therefore considered negligible.
By contrast, the low muscle induction observed upon
misexpression of HyBra1 was obtained consistently. A randomization t test showed that the percentage of muscle
induction by HyBra1 was significantly higher than the
endoderm induction by T and PfBra (P ⬍ 0.005). To our
surprise, however, we found that Brachyury orthologues
from two ascidians and from Drosophila melanogaster possessed strong endoderm-inducing activity in Xenopus animal caps (Figs. 1, and 2I, J, and N).
The endoderm-inducing activity observed with As-T,
CiBra, and Trg but not with the other orthologues could
reflect qualitatively different activities. Alternatively, it
could reflect a quantitative difference in the amounts of
proteins produced in our system. For example, in the case of
the T-box factor VegT, induction of endoderm needs a
higher level of expression than mesoderm (Horb and Thomsen, 1997). To discriminate between these two hypotheses,
we analysed by Western blot the amounts of protein produced in Xenopus animal caps following injection of Xbra
and As-T mRNAs, chosen here as models for the two
356
S. Marcellini et al. / Developmental Biology 260 (2003) 352–361
classes of proteins. For this, we introduced a FLAG tag at
the extreme C terminus of the Xbra and As-T proteins.
mRNAs coding for the tagged and wild-type proteins behaved in a same way in our test (not shown). Using an
antibody against the tag, we then compared the amount of
protein produced in Xenopus embryos after injection of
three doses of mRNA ranging from 0.27 to 2.5 ng. At equal
amounts of mRNA injected, around 10 times less AsTFLAG protein was detected (Fig. 3A, compare the lanes
corresponding to 0.27 ng of Xbra-FLAG mRNA with 2.5 ng
of AsT-FLAG mRNA). Thus, compared with Xbra, As-T is
either poorly translated or unstable following mRNA injection. As in addition, endoderm induction with As-T is
achieved at low mRNA concentrations (Fig. 1), our results
suggest that very low amounts of As-T are sufficient to
induce endoderm, while high amounts of Xbra only induce
mesoderm.
These results raise the possibility that, unlike what was
observed for VegT (Horb and Thomsen, 1997), endoderm
induction is obtained in response to low amounts of
brachyury proteins, while mesoderm is obtained at all doses.
We therefore tested whether decreasing the amount of injected Xbra-FLAG mRNA could lead to endoderm induction. Injection of 250 pg led to a strong response, the caps
adopting a morphology indicative of the presence of mesoderm (compare Figs. 3B and 2E). At 10 pg, ventral mesoderm was morphologically detected in some explants, while
others fully or partly differentiated into epidermis (Fig. 3C).
Finally, injection of 1 or 3 pg of Xbra-FLAG mRNA did not
divert the animal caps from their epidermal morphology
(compare Fig. 3D and E with F). In none of these cases was
endodermin detected (Fig. 3B–E). In contrast, when we
injected 10 or 30 pg of AsT-FLAG mRNA, which should
produce an amount of protein comparable to the injection of
1–3 pg of Xbra-FLAG, a strong endodermin induction was
observed (Fig. 3G and H). Therefore, overexpressing
amounts of Xbra-FLAG protein equivalent or higher to the
amount of AsT-FLAG required for endoderm induction can
only lead to mesoderm induction. We conclude from these
results that Brachyury proteins inducing only mesoderm
bear qualitatively different properties from ascidian and
Drosophila orthologues, which induce both mesoderm and
endoderm. The fact that ascidians and insects are phylogenetically distant within the metazoans suggests that the
ability to induce endoderm reflects a derived character acquired independently in both lineages.
The qualitatively and quantitatively different activities of
Brachyury orthologues in the animal cap assay inspired us
to search for domains of the proteins responsible for these
differences, thus shedding light on the structural changes
that accompanied Brachyury evolution. We first wished to
know what is responsible for the poor efficiency of HyBra1
to induce striated muscle when compared with other orthologues. We swapped the C-terminal domains of Xbra and
HyBra1, two orthologues which show the same specificity
in terms of germ layer induction but different strengths of
induction (Fig. 1). The resulting chimeras, XH and HX (Fig.
4A), were analysed as described above. Consistently with
the activities of the parent proteins, induction of endoderm
was never observed with these chimeras (data not shown).
However, we note that the muscle-inducing activity of the
HX protein is substantially more pronounced than that of
HyBra1 and similar to that of Xbra (Fig. 4A, lanes 2, 3, and
5). Conversely, the XH chimera showed no muscle-induction activity (Fig. 4A, lane 4). Previous data indicate that the
C-terminal domain of Brachyury functions as an activator of
transcription (Conlon et al., 1996; Kispert et al., 1995) and
the present results suggest that the C terminus of HyBra1
behaves as a weaker transcriptional activator than its Xenopus counterpart.
We next applied this approach to address what determines the qualitatively different activities of Xbra, which
induces exclusively mesoderm, and As-T, which induces
both mesoderm and endoderm. Fig. 4B shows the structure
of the chimeric proteins XA and AX. Overexpression of the
XA chimera in animal pole regions caused efficient muscle
induction, but endoderm was never observed, even following injection of large amounts of mRNA (Fig. 4B, lanes
4 – 6). In contrast, overexpression of AX led to both muscle
and endoderm formation (Fig. 4B, lanes 8 and 9). The
activities of XA and AX are therefore similar to those of
Xbra and As-T, respectively (Fig. 4B), indicating that the
principal determinants of Brachyury specificity are located
in the N-terminal half of the protein, which contains the
T-domain. The different specificities of VegT and Xbra also
reside within the N-terminal halves of the proteins (Conlon
et al., 2001).
The N-terminal half of Brachyury proteins contains both
the T-domain and a more N-terminal domain of variable
length. Previous work (Conlon et al., 2001) has proposed
that lysine 149 of the Xbra T-domain might influence the
different specificities of VegT and Brachyury, but this
amino acid is conserved in all known Brachyury orthologues (not shown) and can therefore not explain our results.
Indeed, comparison of Brachyury T-domains revealed that
all amino acids contacting DNA or involved in dimerisation
are strictly conserved (Muller and Herrmann, 1997). More
generally, we could not identify any residue within the
T-domain which might confer on Brachyury the ability to
induce (or not induce) endoderm.
In contrast, alignment of the short upstream N-terminal
domain of Brachyury orthologues revealed a motif (KxxQxxxxHLLxAVxxEMxxGSEKGDPTER) that is conserved
in most deuterostomes, lophotrochozoans, and the phylogenetically enigmatic chaetognaths (Fig. 5A), suggesting that
these amino acids were present in Urbilateria, the common
protostome deuterostome ancestor. Loss of this region correlates with the ability of Brachyury to induce endoderm in
the animal cap assay: it is present neither in the outgroup
member VegT, nor in ascidian or Drosophila Brachyury.
Only HyBra1, which lacks this motif, yet is unable to induce
endoderm, represents an exception to this observation.
S. Marcellini et al. / Developmental Biology 260 (2003) 352–361
357
358
S. Marcellini et al. / Developmental Biology 260 (2003) 352–361
Fig. 4. C-terminal domain swapping experiments. In (A) and (B), the chimeric constructs are drawn as boxes where numbers refer to the amino acid sequence
used from either Xbra, HyBra1, or As-T. (A) Histograms summarising four independent experiments, and indicating the percentage of muscle positive
explants (grey) dissected from uninjected embryos (lane 1) or upon misexpression of synthetic mRNA coding for Xbra (lane 2, 3.3 ng), HyBra1 (lane 3, 2.5–5
ng), XH (lane 4, 2– 6 ng), and HX (lane 5, 0.5– 4 ng). Above each bar is shown the total number of tested explants. (B) Histograms summarising five
independent experiments. The grey and black bars indicate the percentage of muscle and endoderm positive caps, respectively. Numbers above these bars
indicate the total number of explants assayed for both markers. Results are shown for animal caps dissected from uninjected embryos (lane 1), or dissected
from embryos injected with: 1.7 ng of Xbra mRNA (lane 2); 0.2 ng of As-T mRNA (lane 3); the XA mRNA (lane 4, 0.02– 0.05 ng; lane 5, 0.1 ng; lane 6,
0.3–1 ng); and the AX mRNA (lane 7, 0.02 ng; lane 8, 0.09 – 0.125 ng; lane 9, 0.25–1 ng).
To investigate the significance of this motif, we constructed three new chimeras in which this small N-terminal
domain was swapped between Xbra and As-T, thereby creating the constructs XAX, AXX, and XAA (Fig. 5B). All
three constructs induced endoderm in Xenopus animal caps
(Fig. 5C and D), revealing that the short N-terminal domain
of As-T (AXX chimera) and its T-domain (XAX chimera)
are both sufficient to confer endoderm-inducing activity to
Xbra (Fig. 5C and D). These two domains thus harbour key
determinants of the inducing specificity of Brachyury proteins. The failure of HyBra1 to induce endoderm in the
absence of the N-terminal domain may be due to the characteristics of its T-domain.
Discussion
In summary, the results presented here reveal both ancestral and derived functions of Brachyury proteins and
point to novel structural determinants of their specificity of
action.
An efficient mesoderm induction in Xenopus cells was
observed upon misexpression of all bilaterian orthologues.
This may reflect a general role for brachyury in mesoderm
specification across phyla. Such a hypothesis would be
consistent with the expression of Brachyury in at least
subsets of mesodermal cells in vertebrates (Schulte-Merker
et al., 1992; Smith et al., 1991; Wilkinson et al., 1990),
ascidians (Corbo et al., 1997; Yasuo and Satoh, 1993),
echinoderms (Peterson et al., 1999b), hemichordates (Peterson et al., 1999a), annelids (Arendt et al., 2001), molluscs
(Lartillot et al., 2002), and insects (Kusch and Reuter,
1999). Therefore, our observation could be directly linked
to an ancestral function of Brachyury in mesoderm specification. Interestingly, mesoderm induction was also observed, albeit at lower frequency, with a diploblast orthologue, Hybra1. This may suggest that the acquisition by
Brachyury of properties, subsequently used to generate mesodermal fates in bilaterians, predated the emergence of the
mesoderm.
In addition to this shared mesoderm-inducing activity,
most bilaterian orthologues are expressed throughout the
circumference of the blastopore (Technau, 2001), a property
that perfectly correlates with the presence of the N-terminal
peptide and with the ability to induce solely mesoderm in
Xenopus animal caps. This circumblastoporal expression
Fig. 3. Equal protein amounts of Xbra and As-T proteins trigger different biological responses. (A) Embryos were injected animally at the two-cell stage with
the indicated amount of synthetic mRNA coding either for the Xbra-FLAG or the AsT-FLAG protein. At stage 7, animal hemispheres where dissected and
processed by Western blotting with an anti-FLAG antibody. (B–H) Pictures of stage 35 animal caps assayed by in situ hybridisation for the presence of the
endodermin transcript. The embryos were injected with 250 pg (B), 10 pg (C), 3 pg (D), or 1 pg (E) of the Xbra-FLAG mRNA; or with 30 pg (G) and 10
pg (H) of the AsT-FLAG mRNA. (F) Explants dissected from uninjected embryos. Arrowheads and arrow in (C) show animal caps fully or partly
differentiated as epidermis, respectively. The scale bar in (F) represents 400 ␮m in (B–H).
S. Marcellini et al. / Developmental Biology 260 (2003) 352–361
359
Fig. 5. Endodermal determinants are contained both within the T-box and within its upstream domain. (A) Sequence alignment of the N-terminal amino acids
of Brachyury orthologues and VegT. For clarity, the first 37 amino acids of Drosophila (D.m.) Brachyury have been omitted. H.r., Halocynthia roretzi; C.i.,
Ciona intestinalis; C.s., Ciona savignyi; the mollusc sequence is from Patella vulgata (Lartillot et al., 2002), and the chaetognath sequence is from
Paraspadella gotoi (Takada et al., 2002). Conserved residues are shaded. The arrow shows the site of fusion used in the chimeras. The tree on the left indicates
the phylogenetic relationships between the animals from which these orthologues have been isolated. The blue and red colours refer, respectively, to an
induction of mesoderm only, or of both mesoderm and endoderm, in the animal cap assay. The branch leading to the chaetognaths is dotted to indicate that
their exact phylogenetic position is unclear. (B) Drawings of the chimeras. The numbers around A and X represent the amino acid sequence position used
from As-T and Xbra, respectively. The T-domains are shaded grey. (C) The histograms summarise three independent experiments, and show the percentages
of endoderm positive caps. The numbers of assayed samples are noted above each bar. Injected doses of mRNA are: Xbra, 2 ng; As-T, 0.06 – 0.1 ng; XAX,
0.1– 0.3 ng; AXX, 0.1– 0.7 ng; and XAA 0.03 ng. (D) Pictures of explants assayed for the presence of endoderm, and dissected from embryos injected with
2 ng of Xbra mRNA, 0.1 ng of XAX mRNA, 0.2 ng of AXX mRNA, and 0.03 ng of XAA mRNA. The scale bar represents 200 ␮m.
has been independently lost in insects (Kispert et al., 1994)
and tunicates (Bassham and Postlethwait, 2000; Corbo et
al., 1997; Yasuo and Satoh, 1993), animals for which
Brachyury orthologues have lost the N-terminal peptide and
can induce both endoderm and mesoderm in our assay. This
strong correlation suggests that bilaterian Brachyury proteins share an additional ancestral function in the formation
of the blastopore, and that this function is linked to the
presence of the N-terminal peptide. The blastopore has a
key function in the gastrulation movements of all metazoa
but the cell types it gives rise to, in a Brachyury-dependent
manner, vary (Conlon et al., 1996; Gross and McClay,
2001; Kispert et al., 1994; Kusch and Reuter, 1999; Satou et
al., 2001; Schulte-Merker et al., 1994; Wilkinson et al.,
1990). Thus, the different roles of Brachyury orthologues
across phyla may reflect evolutionary changes in the fate of
the blastopore, rather than alterations in the activity of the
Brachyury proteins. A role in the specification of a type of
structure independently of the cell fates it generates has also
been proposed for Distal-less, a gene patterning many body
wall outgrowths in Bilateria (Panganiban and Rubenstein,
2002).
We propose that Brachyury plays an important role in
blastopore formation, and this is consistent with the observation that much of the blastopore of Xenopus goes on to
form muscle, the tissue we monitor in our animal cap assay.
In this regard, it may be significant that endodermal and
mesodermal cells of Xenopus have different migratory prop-
360
S. Marcellini et al. / Developmental Biology 260 (2003) 352–361
erties (Wacker et al., 1998). To ensure the correct behaviour
of the blastopore, it may therefore be important to restrict its
fate to mesoderm, a function in which the conserved Nterminal peptide and T-domain of Brachyury play a crucial
role.
Insects and tunicates have a derived mode of gastrulation. The formation of their blastopores is in large part
independent of Brachyury, as this gene is only expressed in
a minority of blastoporal cells (Bassham and Postlethwait,
2000; Corbo et al., 1997; Kispert et al., 1994; Yasuo and
Satoh, 1993). This may have led to a reduction of the
selective pressure on part of the activity of Brachyury and
allowed loss of the N-terminal domain by the accumulation
of independent mutations. The nature of these mutations is
unlikely to have been dictated by strong constraints as the N
termini of related species, such as Ciona intestinalis and
Ciona savignyi, show very poor sequence similarity (Fig.
5A). Interestingly, proteins which possess the conserved
N-terminal motif and the ability only to induce mesoderm in
our assay are able to mimic, in ascidians and Drosophila,
the activity of endogenous Brachyury (Kusch et al., 2002;
Satoh et al., 2000). Thus, loss of the N terminus probably
reflects a loss of selective pressure on this motif rather than
an obligate requirement for its disappearance.
Finally, the very similar target DNA sequence preferences of Xenopus VegT and Brachyury (Conlon et al.,
2001), coupled to the absence of amino acid changes at
positions contacting DNA in endoderm-inducing proteins,
suggests that the intrinsic DNA binding specificity of
Brachyury orthologues is not a key determinant of their
activity. Rather, it is more likely that most Bilaterian
Brachyury proteins interact with cofactors via their N-terminal and T-domains which, directly or indirectly, confer
inductive specificity in the animal cap assay. A similar
situation has recently been shown for Tbx5, which interacts
via its N terminus and T-domain with the homeodomain
protein Nkx2.5 during vertebrate heart development (Hiroi
et al., 2001). We predict that the cofactors of Brachyury will
be evolutionary conserved and that their identification will
provide important insights into the core mechanisms of
Bilaterian gastrulation.
Acknowledgments
cDNA clones were kindly provided by B.G. Herrmann
(T); M.-A.J. O’Reilly, and S. Schulte-Merker (ntl), N. Satoh
(Am(Bb)Bra1, As-T, SpBra, PfBra), M. Levine, and A. Di
Gregorio (CiBra), T. Kusch (Trg), and D. Arendt (PdBra).
The endodermin probe plasmid was kindly provided by Y.
Sasai. The 12/101 antibody developed by J.P. Brockes was
obtained from the Developmental Studies Hybridoma Bank
developed under the auspices of the NICHD and maintained
by The University of Iowa, Department of Biological Sciences, Iowa City, IA 52242. We are grateful to D. Caillol,
H. Yasuo, S. Darras, L. Kodjabachian, M. Manuel, E.
Delaune, and V. Bertrand for very helpful advice. S.M. was
supported by a French MENRT and by a PhD student
fellowship from the Association pour la Recherche sur le
Cancer. This work was supported by grants in aid from the
CNRS (to P.L.), from the D. F. G. (to U.T.), and by an EU
TMR grant, number ERBFMRX-CT98-0216 (to J.C.S. and
P.L.).
References
Acampora, D., Boyl, P.P., Signore, M., Martinez-Barbera, J.P., Ilengo, C.,
Puelles, E., Annino, A., Reichert, H., Corte, G., Simeone, A., 2001.
OTD/OTX2 functional equivalence depends on 5⬘ and 3⬘ UTR-mediated control of Otx2 mRNA for nucleo-cytoplasmic export and epiblast-restricted translation. Development 128, 4801– 4813.
Arendt, D., Technau, U., Wittbrodt, J., 2001. Evolution of the bilaterian
larval foregut. Nature 409, 81– 85.
Bassham, S., Postlethwait, J., 2000. Brachyury (T) expression in embryos
of a larvacean urochordate, Oikopleura dioica, and the ancestral role of
T. Dev. Biol. 220, 322–332.
Conlon, F.L., Fairclough, L., Price, B.M., Casey, E.S., Smith, J.C., 2001.
Determinants of T box protein specificity. Development 128, 3749 –
3758.
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., 1997. Characterization of a notochord-specific enhancer from the Brachyury promoter region of the
ascidian, Ciona intestinalis. Development 124, 589 – 602.
Cunliffe, V., Smith, J.C., 1992. Ectopic mesoderm formation in Xenopus
embryos caused by widespread expression of a Brachyury homologue.
Nature 358, 427– 430.
Darras, S., Marikawa, Y., Elinson, R.P., Lemaire, P., 1997. Animal and
vegetal pole cells of early Xenopus embryos respond differently to
maternal dorsal determinants: implications for the patterning of the
organiser. Development 124, 4275– 4286.
Gehring, W.J., Ikeo, K., 1999. Pax 6: mastering eye morphogenesis and
eye evolution. Trends Genet. 15, 371–377.
Gross, J.M., McClay, D.R., 2001. The role of Brachyury (T) during
gastrulation movements in the sea urchin Lytechinus variegatus. Dev.
Biol. 239, 132–147.
Haun, C., Alexander, J., Stainier, D.Y., Okkema, P.G., 1998. Rescue of
Caenorhabditis elegans pharyngeal development by a vertebrate heart
specification gene. Proc. Natl. Acad. Sci. USA 95, 5072–5075.
Herrmann, B.G., Labeit, S., Poustka, A., King, T.R., Lehrach, H., 1990.
Cloning of the T gene required in mesoderm formation in the mouse.
Nature 343, 617– 622.
Hiroi, Y., Kudoh, S., Monzen, K., Ikeda, Y., Yazaki, Y., Nagai, R.,
Komuro, I., 2001. Tbx5 associates with Nkx2-5 and synergistically
promotes cardiomyocyte differentiation. Nat. Genet. 28, 276 –280.
Hirth, F., Reichert, H., 1999. Conserved genetic programs in insect and
mammalian brain development. Bioessays 21, 677– 684.
Hopwood, N.D., Pluck, A., Gurdon, J.B., Dilworth, S.M., 1992. Expression
of XMyoD protein in early Xenopus laevis embryos. Development 114,
31–38.
Horb, M.E., Thomsen, G.H., 1997. A vegetally localized T-box transcription factor in Xenopus eggs specifies mesoderm and endoderm and is
essential for embryonic mesoderm formation. Development 124,
1689 –1698.
Hudson, C., Clements, D., Friday, R.V., Stott, D., Woodland, H.R., 1997.
Xsox17alpha and -beta mediate endoderm formation in Xenopus. Cell
91, 397– 405.
S. Marcellini et al. / Developmental Biology 260 (2003) 352–361
Hunter, C.P., Kenyon, C., 1995. Specification of anteroposterior cell fates
in Caenorhabditis elegans by Drosophila Hox proteins. Nature 377,
229 –232.
Kintner, C.R., Brockes, J.P., 1984. Monoclonal antibodies identify blastemal cells derived from dedifferentiating limb regeneration. Nature
308, 67– 69.
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 specification 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-specific transcription factor. EMBO J. 14,
4763– 4772.
Kusch, T., Reuter, R., 1999. Functions for Drosophila brachyenteron and
forkhead in mesoderm specification and cell signalling. Development
126, 3991– 4003.
Kusch, T., Storck, T., Walldorf, U., Reuter, R., 2002. Brachyury proteins
regulate target genes through modular binding sites in a cooperative
fashion. Genes Dev. 16, 518 –529.
Lartillot, N., Lespinet, O., Vervoort, M., Adoutte, A., 2002. Expression
pattern of Brachyury in the mollusc Patella vulgata suggests a conserved role in the establishment of the AP axis in Bilateria. Development 129, 1411–1421.
Lemaire, P., Darras, S., Caillol, D., Kodjabachian, L., 1998. A role for the
vegetally expressed Xenopus gene Mix. 1 in endoderm formation and
in the restriction of mesoderm to the marginal zone. Development 125,
2371–2380.
Muller, C.W., Herrmann, B.G., 1997. Crystallographic structure of the T
domain-DNA complex of the Brachyury transcription factor. Nature
389, 884 – 888.
Nagao, T., Leuzinger, S., Acampora, D., Simeone, A., Finkelstein, R.,
Reichert, H., Furukubo-Tokunaga, K., 1998. Developmental rescue of
Drosophila cephalic defects by the human Otx genes. Proc. Natl. Acad.
Sci. USA 95, 3737–3742.
Nieuwkoop, P.D., Faber, J., 1967. Normal table of Xenopus laevis. North
Holland, Amsterdam.
O’Reilly, M.A., Smith, J.C., Cunliffe, V., 1995. Patterning of the mesoderm in Xenopus: dose-dependent and synergistic effects of Brachyury
and Pintallavis. Development 121, 1351–1359.
Panganiban, G., Rubenstein, J.L., 2002. Developmental functions of the
Distal-less/Dlx homeobox genes. Development 129, 4371– 4386.
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 flava. 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.
361
Sasai, Y., Lu, B., Piccolo, S., De Robertis, E.M., 1996. Endoderm induction by the organizer-secreted factors chordin and noggin in Xenopus
animal caps. EMBO J. 15, 4547– 4555.
Satoh, G., Harada, Y., Satoh, N., 2000. The expression of nonchordate
deuterostome Brachyury genes in the ascidian Ciona embryo can promote the differentiation of extra notochord cells. Mech. Dev. 96,
155–163.
Satou, Y., Imai, K.S., Satoh, N., 2001. Action of morpholinos in Ciona
embryos. Genesis 30, 103–106.
Schulte-Merker, S., Ho, R.K., Herrmann, B.G., Nusslein-Volhard, C.,
1992. The protein product of the zebrafish 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., Halpern, M.E., Kimmel, C.B.,
Nusslein-Volhard, C., 1994. no tail (ntl) is the zebrafish homologue of
the mouse T (Brachyury) gene. Development 120, 1009 –1015.
Smith, J., 1997. Brachyury and the T-box genes. Curr. Opin. Genet. Dev.
7, 474 – 480.
Smith, J.C., Price, B.M., Green, J.B., Weigel, D., Herrmann, B.G., 1991.
Expression of a Xenopus homolog of Brachyury (T) is an immediateearly response to mesoderm induction. Cell 67, 79 – 87.
Spring, J., Yanze, N., Josch, C., Middel, A.M., Winninger, B., Schmid, V.,
2002. Conservation of brachyury, mef2, and snail in the myogenic
lineage of jellyfish: a connection to the mesoderm of bilateria. Dev.
Biol. 244, 372–384.
Tada, M., O’Reilly, M.A., Smith, J.C., 1997. Analysis of competence and
of Brachyury autoinduction by use of hormone-inducible Xbra. Development 124, 2225–2234.
Tagawa, K., Humphreys, T., Satoh, N., 1998. Novel pattern of Brachyury
gene expression in hemichordate embryos. Mech. Dev. 75, 139 –143.
Takada, N., Goto, T., Satoh, N., 2002. Expression pattern of the Brachyury
gene in the arrow worm paraspadella gotoi (chaetognatha). Genesis 32,
240 –245.
Technau, U., 2001. Brachyury, the blastopore and the evolution of the
mesoderm. Bioessays 23, 788 –794.
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.
Wacker, S., Brodbeck, A., Lemaire, P., Niehrs, C., Winklbauer, R., 1998.
Patterns and control of cell motility in the Xenopus gastrula. Development 125, 1931–1942.
Wilkinson, D.G., Bhatt, S., Herrmann, B.G., 1990. Expression pattern of
the mouse T gene and its role in mesoderm formation. Nature 343,
657– 659.
Wilson, P.A., Hemmati-Brivanlou, A., 1995. Induction of epidermis and
inhibition of neural fate by Bmp-4. Nature 376, 331–333.
Yasuo, H., Satoh, N., 1993. Function of vertebrate T gene. Nature 364,
582.
Yasuo, H., Satoh, N., 1994. An ascidian homolog of the mouse Brachyury
(T) gene is expressed exclusively in notochord cells at the fate restricted stage. Dev. Growth Differ. 36, 9 –18.