|
|
|
|
|
|
|
||||
| Home Help Feedback Subscriptions Archive Search Table of Contents | ||||
First published online 12 September 2007
doi: 10.1242/dev.007799
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
UMR 7009 CNRS, Université Pierre et Marie Curie (Paris 6) Observatoire Océanologique, 06230 Villefranche-sur-mer, France.
Author for correspondence (e-mail:
lepage{at}obs-vlfr.fr)
Accepted 31 July 2007
 |
SUMMARY |
|---|
 TOP SUMMARY INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key words: nodal, vg1, univin, GDFs, bZIP, Smad, SoxB1, Dorsal-ventral axis, Oral-aboral axis, Cis-regulatory analysis, Phylogenetic footprint, Gene regulatory network, Paracentrotus lividus
|
INTRODUCTION |
|---|
|
TOPSUMMARYINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
;
Duboc et al., 2004). In the
sea urchin embryo, expression of nodal is initiated around the
60-cell stage in most cells of the presumptive ectoderm and becomes rapidly
restricted to the presumptive ventral ectoderm by mechanisms that remain
unknown.
nodal is the earliest zygotic gene displaying a restricted
expression along the dorsal-ventral axis during sea urchin development and
Nodal function is absolutely required for establishment of dorsal-ventral
polarity. When translation of the nodal transcript is prevented,
specification of both the ventral and the dorsal ectoderm fails and most of
the ectoderm differentiates into a thickened ciliated ectoderm that represents
the default state in the absence of Nodal. Functional studies have shown that
overexpression of Nodal is sufficient to specify most ectodermal cells with a
ventral fate. Furthermore, rescue experiments revealed that Nodal-expressing
cells have a long-range organizing activity and are capable of restoring
dorsal-ventral polarity over the whole embryo
(Duboc et al., 2005;
Duboc et al., 2004). Nodal
activates a regulatory network of genes necessary for dorsal-ventral axis
formation, encoding key transcription factors such as Goosecoid, Brachyury and
FoxA, as well as signaling molecules such as Bmp2/4 and the Nodal antagonist
Lefty (Duboc and Lepage, 2006;
Duboc et al., 2005).
Therefore, the nodal gene is a crucial component of dorsal-ventral
axis formation in the sea urchin and understanding the molecular mechanisms
responsible for the initiation of nodal expression is important for
understanding how the body plan of this organism is established.
Classical bisection experiments on sea urchin embryos performed by
Hörstadius and more recently by Wilkrayamanake et al. and Yaguchi et al.
demonstrated that in the absence of vegetal signaling, animal-half explants
develop into neurogenic ectoderm that lacks dorsal-ventral polarity. However,
when these explants are recombined with vegetal blastomeres or when they are
treated with lithium, dorsal-ventral polarity and stomodeum formation are
rescued (Hörstadius,
1973; Wikramanayake et al.,
1995; Yaguchi et al.,
2006). Together with our previous finding that expression of
nodal also requires a functional Tcf/ß-catenin pathway
(Duboc et al., 2004), these
experiments strongly suggest that a ß-catenin-dependent vegetal signal is
required for induction of nodal expression in the animal
hemisphere.
Another molecular pathway implicated in the activation of nodal
expression is the p38 MAP kinase signaling pathway. p38 signaling is activated
around the 60-cell stage ubiquitously and is transiently inactivated on the
presumptive dorsal side. When p38 signaling is inhibited, nodal
expression is not initiated (Bradham and
McClay, 2006). The function of p38 during the establishment of
nodal expression is not known.
A possible link between p38 signaling and the transcriptional activation of
nodal is suggested by the known function of p38 in the
transcriptional responses downstream of oxidative stress
(Torres and Forman, 2003).
Experiments performed some years ago implicated oxidative gradients in the
formation of the dorsal-ventral axis of sea urchin embryos
(Czihak, 1971). Intriguingly,
these respiratory gradients, visualized by mitochondrial cytochrome oxidase
activity, prefigure the dorsal-ventral axis of the early embryo as early as
the 8-cell stage (Czihak,
1963). Even more enigmatic is the finding that orientation of the
dorsal-ventral axis can be biased by using respiratory inhibitors or by
culturing embryos in hypoxic conditions
(Child, 1948;
Coffman and Davidson, 2001;
Pease, 1941). Recent studies
reported that mitochondria are asymmetrically distributed in the egg of
Strongylocentrotus purpuratus and that microinjection of purified
mitochondria can bias orientation of the dorsal-ventral axis
(Coffman et al., 2004).
However, the links between the mitochondria, redox gradients, p38 signaling
and the transcriptional machinery responsible for initiating nodal
expression remain to be established. Since the nodal gene is a key
regulator of dorsal-ventral axis formation and the earliest zygotic gene
showing a restricted expression along this axis, it provides an excellent
entry point to dissect these relationships.
In this study, we identified the cis-regulatory elements of the sea urchin nodal gene and used them to dissect the regulatory interactions involved in the control of nodal expression.
|
MATERIALS AND METHODS |
|---|
|
TOPSUMMARYINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). A
consensus tree with a 50% cut-off value was derived from a 500 bootstrap
analysis using Mega 3.1
(http://www.megasoftware.net/).
Numbers above branches represent posterior probabilities, calculated from this
consensus. Linkage analysis was performed by searching the zebrafish (http://www.ensembl.org/Danio_rerio/index.html) and sea urchin (http://www.ncbi.nlm.nih.gov/genome/guide/sea_urchin/index.html) genomes for dvr1/bmp2 and for univin/bmp2/4.
In situ hybridization and quantitative (Q) PCR
The nodal and univin probes have been described
previously (Duboc et al.,
2004; Lapraz et al.,
2006). For QPCR, total RNA was prepared from 100 embryos using
Trizol (Invitrogen) and reverse transcribed using the TaqmanR PCR Kit (Applied
Biosystems) after first removing DNA with DNaseI. Cycling was performed using
a Lightcycler 480 (Roche) and Fast Start SYBR Green PCR Kit (Roche) with
optimized primer pairs. Relative quantification of nodal expression
between experimental samples and controls was obtained by subtracting the
sample Ct (threshold cycle) from the control Ct and using ubiquitin
mRNA as an internal control.
Treatments, constructs, RNA and morpholino injections
Treatment with 10 µM SB431542 was performed as described previously
(Duboc et al., 2005).
Dominant-negative dnTcf mRNA was used at 500 µg/ml and
univin mRNA at 800 µg/ml.
The specificity of the Nodal morpholino used in this study has been
demonstrated previously (Duboc et al.,
2004). Two oligonucleotides, directed against different regions of
the SoxB1 5' UTR were used and produced similar phenotypes including
radialization, absence of spicules and gut as described previously
(Kenny et al., 2003).
Similarly, two different morpholino oligonucleotides directed against the
univin transcript produced identical phenotypes. A single morpholino
directed against the first eight codons of the
alk4/5/7 transcript produced a phenotype extremely
consistent with the presumed role of the protein encoded by this transcript as
a receptor for Nodal. The specificity of this morpholino was demonstrated by
performing a rescue experiment. Sequences of morpholino oligonucleotides
are:
NodalMo, 5'-ACTTTGCGACTTTAGCTAATGATGC;
UnivinMo1, 5'-ACGTCCATATTTAGCTCGTGTTTGT;
UnivinMo2, 5'-GTTAAACTCACCTTTCTAAACTCAC;
SoxBMo1, 5'-GACAGTCTCTTTGAAATTAGACGAC;
SoxB1Mo2, 5'-GAAATAAAGCCAAAGTCTTTTGATG; and
Alk4/5/7Mo, 5'-TAAGTATAGCACGTTCCAATGCCAT.
All injections were repeated three times and for each experiment 50-100 embryos were analyzed. Only representative phenotypes present in at least 80% of the injected embryos are presented.
Isolation of BAC clones
Paracentrotus lividus and Lytechinus variegatus BAC
libraries were screened with a radioactive probe corresponding to the 5'
UTR of nodal and ten positive clones were further characterized by
pulse-field electrophoresis, PCR and restriction analysis. 15 kb of upstream
sequence were obtained for the longest L. variegatus clone (120 kb)
and for the Paracentrotus (60 kb) clone by subcloning and sequencing
restriction fragments. The entire P. lividus BAC clone was
subsequently sequenced by the Marine Genomics Europe technology platform.
Sequence from the Strongylocentrotus nodal locus was obtained from
the Sea Urchin Genome Assembly.
Comparison of genomic sequences
Sequence comparisons were performed with the Vista platform
(http://genome.lbl.gov/vista/index.shtml).
The window size used varied between 50 and 100 bp, with a window of 50 bp with
75% conservation.
Reporter constructs and mutagenesis
A GFP construct containing the Endo16 basal promoter, EpGFP
(Arnone et al., 1997), was used
for spatial expression analysis. For quantitative analyses, the luciferase
expression vector pGL3 Basic was modified by introducing the endo16
promoter in front of the luciferase start site (henceforth referred to as
EpGluc). The relevant conserved regions were PCR amplified from the
Paracentrotus BAC clone using the Long Expand PCR System (Roche) and
appropriate primers (see Table
1). Each relevant fragment was introduced into the appropriate
vector using standard molecular biology techniques and each construct was
verified by sequencing. The conserved predicted binding sites for Smad,
homeodomain, TCF, Oct, Sox and bZIP factors were identified using TransFac and
MatInspector software (Genomatix). Primers containing between four and eight
mutated bases were generated for each of these sites (see
Table 1). Mutations were
introduced by PCR using Pfx polymerase (Invitrogen). All mutations were
confirmed by restriction digestion and sequencing.
|
). In the case of GFP reporters, the linearized construct was
injected at 5 ng/µl together with carrier DNA (HindIII-digested
sea urchin genomic DNA) at 30 ng/µl. In the case of luciferase reporters,
the linearized plasmid was injected at 5 ng/µl, together with pRL-TK
(Renilla luciferase) or Endo16-Renilla DNA at 7.5 ng/µl and
carrier DNA at 30 ng/µl. For each experiment, more than 50 embryos were
mounted and the position of GFP-expressing cells within individual embryos was
scored. For luciferase analysis, between 100 and 150 injected embryos were
collected and the level of luciferase and Renilla determined with the Dual
Luciferase Kit (Promega) according to the manufacturer's instructions using a
GloMax luminometer with an integration of 10 seconds. The level of luciferase
activity was normalized to the level of Renilla activity for each experiment.
All experiments were repeated three times using separate batches of
embryos.
|
|
|
RESULTS |
|---|
|
TOPSUMMARYINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
; Yuh et al.,
2002). BAC recombinant clones containing the nodal gene
were isolated from P. lividus and L. variegatus. The
sequence of the S. purpuratus nodal locus was obtained in silico from
the sea urchin genome sequence. We scanned the genomic sequences of the three
nodal genes for short conserved motifs with the Vista program using
as parameters a 70% identity in a 50 bp window. In addition to regions
corresponding to the coding sequences, this comparison identified three highly
conserved non-coding regions (Fig.
1A,B). These consist of a 700 bp region just 5' of the
transcriptional start site, a 650 bp region within the intron, and a 600 bp
region corresponding to the 3' UTR
(Fig. 1B,C).
|
In order to determine which areas of the conserved 5' and intronic sequences contained sites involved in the transcriptional activation and/or spatial repression of nodal expression, we performed a deletion analysis (Fig. 1C,D). A fragment containing the first 350 bp of the 5' proximal region did not drive detectable expression of the GFP reporter gene (Fig. 1D). By contrast, a fragment containing the last 300 bp of the proximal region efficiently drove expression of the reporter gene (Fig. 1D). Finally, a deletion that removed 600 bp of the proximal region, leaving only the last 50 bp and the predicted TATA box, did not show any expression (data not shown). This indicates that cis-regulatory elements driving nodal expression are located within a 300 bp region just upstream of the transcriptional start site.
Deletion analysis of the intron region showed that neither the first half nor the last 464 bp of the intron is able to drive GFP expression, whereas a small conserved region of 350 bp is capable of driving strong expression (Fig. 1D). These results indicate that regulatory elements sufficient to drive nodal expression reside in a short region just upstream of the first exon (R-module) and in a small, highly conserved region within the intron.
Spatial regulatory activity of the conserved nodal cis-regulatory sequences
In order to determine whether the identified regulatory modules are
sufficient for the proper spatial expression of nodal on the oral
side, we examined the spatial expression profiles of GFP reporter genes driven
by the R-module (the 300 bp 5' of exon 1) or by various regions derived
from the intron (Fig. 2A and
Table 2).
|
Taken together, these data suggest that both the R-module and the intron contain regulatory elements involved in the spatial restriction of nodal and that the combinatorial activity of these elements may restrict nodal expression to the oral territory. However, the relatively high level of ectopic expression of the R-module- and intron-driven reporter constructs suggests that additional repressor elements might be involved in the spatial control of nodal expression.
Deletion analysis of the intron
Deletion analysis of the intron revealed that it contains a central module
involved in global activation (the A-element), flanked by two restriction
elements (R1 and R2) involved in spatial restriction
(Fig. 2A,B). The A-element
drives robust GFP expression evenly throughout the embryonic territories
(overall: oral 66%, ectopic 66%), suggesting that this element responds to
broadly distributed factors that regulate nodal expression. Embryos
injected with an R1+A-element GFP construct displayed GFP expression similar
to embryos injected with the full intron GFP reporter (overall: 72% oral, 52%
ectopic). Embryos injected with an R2+A-module GFP construct also showed
preferential expression of GFP in the oral territory (overall: 83% oral, 36%
ectopic). Both of these constructs showed a decrease in the overall aboral and
endoderm ectopic expression and a concomitant rise in the oral expression
compared with the A-module. The effect of the R1-module on the spatial
restriction of the reporter is modest (R1+A, 14% aboral decrease, 8% endoderm
decrease), suggesting that it acts as a weak repressor element. The R2-module
is more effective than R1 in restricting the spatial expression of the
reporter (R2+A, 28% aboral decrease, 20% endoderm decrease), suggesting that
R2 can exert negative control of nodal expression
(Table 2). The extent of this
effect is in keeping with other negative-regulatory spatial elements
characterized in several sea urchin genes
(Minokawa et al., 2005;
Ransick and Davidson, 2006).
Thus, the 3' half of the intron sequence contains regulatory elements
that are able to respond to a positive regulator(s) that is globally
expressed, as well as elements that are necessary to repress nodal
expression in the endoderm.
The proximal R-module and the intron both contain Smad binding sites involved in autoregulation by Nodal
Studies on the regulation of nodal gene expression in vertebrates
have shown that regulatory elements upstream of the first exon control
activation of nodal and that elements within the introns contain
binding sites that are necessary for the autoregulation by Smad2/3
(Norris et al., 2002;
Osada et al., 2000). To
examine the regulatory architecture of the sea urchin nodal gene, we
compared the kinetics of luciferase expression driven by the R-module and by
the intron (Fig. 3A). When the
R-module was fused to a luciferase reporter construct carrying the
endo16 basal promoter (EpGluc), it activated luciferase expression as
early as the 60-cell stage (Fig.
3A) and levels of expression rose until the mesenchyme blastula
stage, similar to endogenous nodal activation
(Duboc et al., 2004). However,
the intron region did not activate luciferase expression until the very early
blastula stage and the level of expression was reduced when compared with
R-module expression at this stage. The activity of the intron-driven
luciferase reporter continued to rise, with a sharp increase between early
blastula and mesenchyme stages, perhaps indicating the time when
autoregulation by Nodal signaling is strongly influencing the promoter
(Fig. 3A). These data suggest
that the R-module contains transcription factor binding sites necessary for
the activation of nodal around the 60-cell stage, whereas the
function of the intronic modules might be to respond to autoregulation by
Nodal signaling. We reasoned that mutating the predicted Smad binding sites
present within these regions should uncover which module is involved in
autoregulation, as the mutated construct would no longer be able to respond to
Nodal autoregulation. We mutated all four Smad binding sites present within
the R-module and all 12 sites present within the intron
(Fig. 4 and see Fig. S1 in the
supplementary material). In both cases, mutation of the Smad binding sites
caused a sharp decrease of the activity of the reporters at the hatched
blastula stage. However, the R-module Smad mutant still retained 35% of its
original expression when compared with the normal R-module, whereas the intron
Smad mutant had only 3% residual expression compared with the normal intronic
region (Fig. 3B). Similarly,
injection of the Nodal morpholino or treatments with the Alk4/5/7 receptor
inhibitor SB431542 (Inman et al.,
2002), which should both inhibit autoregulation by Nodal, severely
decreased the activity of the intronic reporter gene, whereas it only
moderately affected the activity of the R-module reporter
(Fig. 3B). Taken together,
these data suggest that both the R-module and the intron act as autoregulatory
elements. However, when compared with the intron, the R-module still drove
relatively high levels of luciferase expression in the absence of
autoregulatory Smad binding sites, suggesting that it contains other
transcription factor binding sites necessary for full nodal
expression.
|
).
This analysis indicated that the R-module contains multiple candidate binding
sites for several families of transcription factors. We then mutated 17
conserved motifs within the R-module, including the predicted binding sites as
well as other well conserved regions and analyzed the effect of each mutation
alone, or in combination with other mutations, on the activity of the
luciferase and GFP reporters at blastula stages
(Fig. 4). These functional
tests allowed the identification of six motifs which, when mutated, caused a
strong decrease in reporter gene activity at the hatched blastula stage,
suggesting that these sites are necessary for activation of nodal
expression. Each of the sites found in the systematic screen corresponds to a
predicted transcription factor binding site
(Fig. 4A-C). Mutation of all
three predicted homeodomain factor binding sites decreased luciferase
expression to about 5% of that of control constructs
(Fig. 4C). Similarly, a
mutation that maps to one of the predicted bZIP binding sites caused a 20-fold
reduction of the activity of the reporter
(Fig. 4C). Interestingly, the
sequence of this motif, GCCGATTCAT, resembles the antioxidant responsive
element (ARE), GCNNNGTCAY, which mediates transcriptional regulation of
antioxidant proteins (Rushmore et al.,
1991). These data suggest that both homeodomain and bZIP
transcriptional regulators are crucial for the activation and/or maintenance
of nodal expression. In addition to homeodomain and bZIP, three
mutations produced strong effects on the activity of the reporter. These
mapped to a motif, TACAAAAGA, that resembles a Tcf binding site (A/TA/TCAAAG)
(Giese et al., 1991;
van de Wetering and Clevers,
1992), to a motif, AACAAT, which fits the consensus Sox binding
site (AACAAT) (van Beest et al.,
2000) and to a motif, ATGCTAAA, that resembles the Oct1 consensus
binding site (ATGCAAA) (Jin and Li,
2001). Mutation of the TCF-like, Sox or Oct sites decreased the
activity of the R-module-driven reporter gene to 
64%, 40% and 30%,
respectively, of its original value (Fig.
4C).
Interestingly, we identified a site which, when mutated, consistently
resulted in an almost 3-fold stimulation of the expression of the reporter
gene (Fig. 4D). The sequence of
this motif, CAACGGT, fits the consensus binding site sequence for Myb
(YAACG/TG) (Luscher and Eisenman,
1990). The Myb transcription factor is known to act as a repressor
and a global regulator of chromatin structure
(Coffman et al., 1997;
Lipsick, 2004). To determine
whether this site is involved in spatial restriction of nodal, we
introduced the mutation into the R-module GFP construct and assayed spatial
expression at the early pluteus stage (Fig.
4E). Indeed, the percentage of embryos displaying restricted
expression of GFP in the oral territory decreased from 65% to 47% when this
site was mutated. Furthermore, when the Myb site was mutated, the percentage
of embryos expressing GFP in aboral and endoderm also increased compared with
the wild-type R-module (4% and 20% increase, respectively). These variations
in the spatial expression of reporter genes are somewhat modest, but they are
of the same order of magnitude as those caused by mutating important
transcription factor binding sites in other spatially regulated sea urchin
genes (Minokawa et al., 2005;
Ransick and Davidson, 2006).
Thus, the Myb-like binding site is likely to bind a regulatory factor that
represses nodal expression in the aboral and endomesodermal
territory. Alternatively, the function of this repressor site might be to
control the level of nodal expression in the oral ectoderm.
Initiation of nodal expression requires TCF, SoxB1 and early TGF-ß signaling
Since the R-module contains binding sites for Sox and Tcf and because both
factors are expressed maternally and are broadly distributed in the early
embryo (Huang et al., 2000;
Kenny et al., 2003), we
examined whether Tcf and SoxB1 function is required for expression of
endogenous nodal and of the nodal reporter genes. Consistent
with previous studies (Kenny et al.,
2003), we found that interfering with SoxB1 function using
antisense morpholino oligonucleotides severely affected dorsal-ventral
patterning, causing embryos to develop with a strongly radialized phenotype
(data not shown). Examination of endogenous nodal expression in the
SoxB1 morpholino-injected embryos at early blastula stages revealed that
nodal expression was abolished
(Fig. 5A). Also, the activity
of the R-module-driven reporter gene decreased to 18% of its normal value in
embryos injected with the SoxB1 morpholino at the hatched blastula stage
(Fig. 5D). Similarly,
co-injection of RNA encoding a dominant-negative (dn) version of Tcf reduced
the activity of the reporter gene to 11% of its original value
(Fig. 5D). We conclude that TCF
and SoxB1 function is essential for nodal expression, possibly
through direct binding to the R-module.
We had shown previously that maintenance of nodal expression
strongly depends on an autoregulatory loop and that in the absence of Nodal
signaling or following overexpression of Lefty protein, nodal
expression is lost at late blastula stages
(Duboc et al., 2005;
Duboc et al., 2004). To
further test whether nodal expression also requires TGF-ß
signaling and to determine the period when autoregulation becomes important
for maintenance of nodal, we microinjected morpholino
oligonucleotides directed against nodal or
alk4/5/7, which encodes a candidate Nodal type I
receptor (Lapraz et al.,
2006), and analyzed the temporal and spatial expression of
nodal by in situ hybridization and QPCR
(Fig. 5B). Strikingly,
nodal expression was barely detectable in the MoNodal-injected
embryos at the early 64/128-cell stage
(Fig. 5Bf-h) and completely
absent starting at the early blastula stage
(Fig. 5Bi-j). In the
MoAlk4/5/7-injected embryos, nodal expression was not detectable at
any stage (Fig. 5Bk-o).
Consistent with these observations, MoNodal injection or SB431542 treatment
caused a 3- to 4-fold decrease of the activity of the luciferase reporter gene
(Fig. 5D). These results
suggest that a Nodal-dependent autoregulatory loop is active very early and
its integrity is crucial to maintain nodal expression.
Univin, a maternally expressed TGF-ß, is the sea urchin ortholog of Vg1 and acts upstream of nodal expression
During the experiments described above, we noticed that nodal
expression was more effectively downregulated in embryos treated with SB431542
or microinjected with MoAlk4/5/7 than in embryos injected with the Nodal
morpholino. This suggested that another early-acting TGF-ß signal might
participate in the regulation of nodal expression. Univin is a good
candidate for this additional early-acting TGF-ß signal required for
nodal expression as it is an abundant, ubiquitously expressed
maternal transcript and because its zygotic expression pattern encompasses
that of nodal (Lapraz et al.,
2006).
We tested whether Univin is the early signal required for nodal expression. Following injection of the Univin morpholino, nodal transcripts could not be detected by in situ hybridization (Fig. 5Bp-t), although a residual level of nodal transcripts could be detected by QPCR (Fig. 5E). Injection of the Univin morpholino also caused a reduction in the activity of the R-module-driven luciferase reporter gene at the hatched blastula stage (Fig. 5D). However, as in the presence of the Alk4/5/7 inhibitor or in the absence of Smad binding sites, the activity of the R-module was not completely abolished, suggesting that initial activation of nodal is achieved by transcription factors that act in parallel with TGF-ß signaling. Furthermore, unlike nodal, which requires Nodal and Alk4/5/7 signaling early to be maintained, univin expression was found to be independent of Alk4/5/7 signaling (Fig. 5C) or nodal expression (data not shown), consistent with the idea that Univin acts very early to regulate nodal expression. Finally, in an attempt to link the zygotic expression of univin with the activity of maternal transcription factors, we examined the dependence of univin expression on maternal Wnt/ß-catenin signaling and SoxB1. We found that zygotic expression of univin, like zygotic expression of nodal, critically requires TCF and SoxB1 function (Fig. 5A). Taken together, these results show that Univin, a maternally deposited TGF-ß, is required early for nodal expression during sea urchin development and that both univin and nodal expression require Tcf and SoxB1 function.
|
).
|
The striking effects on nodal expression resulting from
overexpression or downregulation of Univin prompted us to re-examine the
phylogenetic relationships between this TGF-ß and Nodal. Previous
phylogenetic comparisons indicated that Univin is most closely related to BMPs
and Vg1 (Stenzel et al.,
1994), whereas more recent comparisons indicated a close
evolutionary relationship with Gdf factors
(Lapraz et al., 2006). To
precisely determine the orthology relationship of Univin, we performed a
phylogenetic analysis using a set of sequences that included several
vg1 members (Fig. 8A).
This analysis confirmed that Univin is more closely related to Vg1 from
Xenopus, Dvr1 from zebrafish, and to Gdf1 and Gdf3 from mouse, than
to BMPs. This strong phylogenetic relationship is further supported by genomic
linkage data (Fig. 8B). In the
zebrafish genome, the dvr1 gene is located 8 kb from bmp2a.
Similarly, the univin transcription unit lies only 32 kb from
bmp2/4. The synteny of
univin/bmp2/4 and dvr1/bmp2a in
the sea urchin and zebrafish genomes strongly suggests that these two genes
evolved by gene duplication before emergence of the chordates.
|
DISCUSSION |
|---|
|
TOPSUMMARYINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
;
Kelly et al., 2000),
Xenopus (Agius et al.,
2000; Hyde and Old,
2000; Rex et al.,
2002; Schohl and Fagotto,
2003; Takahashi et al.,
2000; Xanthos et al.,
2002; Yang et al.,
2002) and mouse (Ben-Haim et
al., 2006; Chazaud and
Rossant, 2006; Huelsken et
al., 2000). Studies performed in Xenopus and mouse
identified the regulatory elements responsible for regulation of
nodal genes and, in the case of Xnr1 and Xnr3,
Wnt-responsive elements containing consensus binding sites for the Lef-1/Tcf
proteins have been identified (Hyde and
Old, 2000; Kofron et al.,
1999; McKendry et al.,
1997; Osada et al.,
2000). Inspection of the sea urchin nodal promoter
detected one such motif and mutational analysis indicated that it is indeed
essential for high-level promoter activity. Furthermore, inhibition of Tcf
function severely affected expression of a nodal reporter gene. Thus,
expression of nodal (Duboc et
al., 2004) and Univin (this work) requires Tcf activity, further
reinforcing the idea that in the sea urchin, as in vertebrates, initiation of
nodal expression critically relies on prior activation of the
Wnt/ß-catenin pathway. In the future, it will be important to determine
whether the Tcf/ß-catenin complex binds directly to the nodal
promoter, or if a relay molecule produced at the level of the vegetal pole
mediates this effect.
|
; Zhang et al.,
2005; Zhang et al.,
2004; Zhang et al.,
2003). In Xenopus, the SoxB1 family member Sox3 is
expressed maternally in the animal hemisphere, where it restricts mesendoderm
induction (Penzel et al.,
1997). Studies in Xenopus and zebrafish have also shown
that Sox3 acts as a transcriptional repressor of Xnr5
(Zhang et al., 2003). Based on
the conservation of Sox3 function in vertebrates, Zhang et al. proposed that
SoxB1 proteins might fulfil a phylogenetically conserved role in regulating
cell fate specification along the animal-vegetal axis through the regulation
of nodal-like genes.
During sea urchin early development, SoxB1 is expressed in the presumptive
animal hemisphere, where it prevents the early ß-catenin-dependent
vegetal signaling necessary for specification of the mesendoderm
(Kenny et al., 1999;
Kenny et al., 2003). Our
finding that SoxB1 is required for expression of both nodal and
univin, confirms that SoxB1 factors do indeed play a phylogenetically
conserved role as regulators of nodal expression in deuterostomes.
However, it appears that the function of SoxB1 factors has diverged in the two
phyla as they serve as positive regulators of nodal in the sea urchin
but act as repressors of nodal expression in vertebrates.
|
|
|
). Conversely, when Vg1 signaling is blocked, or in embryos
depleted of endogenous vg1 transcript, endomesoderm development and
formation of the organizer fail (Birsoy et
al., 2006; Joseph and Melton,
1998). Transcripts encoding Dvr1, the zebrafish ortholog of Vg1,
are also deposited maternally and are distributed throughout the
dorsal-ventral axis (Dohrmann et al.,
1996; Helde and Grunwald,
1993). Mature Dvr1 is a potent inducer of dorsal mesoderm, but
loss-of-function experiments have not been performed to demonstrate that Dvr1
is required for axis formation. A chick ortholog of vg1 has also been
characterized (Seleiro et al.,
1996; Shah et al.,
1997). Overexpression of chick Vg1 is capable of initiating
formation of an ectopic embryonic axis and organizer by cooperating with
canonical Wnt signals, such as Wnt1 and Wnt8, to induce nodal
(Bertocchini et al., 2004;
Skromne and Stern, 2001;
Skromne and Stern, 2002). The
mouse genome contains two genes, Gdf1 and Gdf3, that are
closely related to vg1 (Jones et
al., 1992; Lee,
1991), and functional experiments demonstrated that they act in a
Nodal signaling pathway and share several properties with Nodal factors
(Andersson et al., 2006;
Chen et al., 2006;
Cheng et al., 2003). For
example, Gdf1 and Gdf3, like nodal, can induce an
ectopic embryonic axis when overexpressed. Gdf1 and Gdf3 also form a complex
with activin receptors, require the EGF-CFC coreceptor cripto (also known as
cryptic) and, like Nodal, their activity is inhibited by Lefty. Furthermore,
embryos mutant for Gdf3 frequently display abnormal expression of
Nodal, indicating that Gdf factors participate in the regulation of
Nodal expression in the mouse
(Chen et al., 2006). Taken
together, these experiments suggest that Gdf1 and Gdf3 are indeed functional
homologs of Vg1 in the mouse (Shen,
2007).
Examination of the recently sequenced sea urchin genome revealed that
univin is the only gene related to Gdf1/3
(Lapraz et al., 2006). The
functional relationship between Univin and other TGF-ß proteins, such as
Vg1 and Nodal, escaped attention in previous studies partly because both GDFs
and Univin were initially described as related to BMPs. The subsequent finding
that univin lies only a few kb from Bmp2/4 in the sea urchin
genome further reinforced the idea that univin and bmp2/4
were closely related and recently duplicated genes
(Lapraz et al., 2006).
However, the functional analysis of Univin led us to reinvestigate the
relationships between Univin and other TGF-ß proteins and to discover
that the function of Univin is much more closely related to that of Nodal than
of Bmp2/4. Overexpression of Univin induces ectopic expression of
nodal in the aboral ectoderm, whereas blocking translation of
univin transcripts prevents initiation of nodal expression.
univin is expressed maternally throughout the dorsal-ventral axis,
then zygotically in a large belt of cells surrounding the equatorial region
starting at the early blastula stage
(Lapraz et al., 2006;
Stenzel et al., 1994).
Therefore, in the sea urchin, as in vertebrates, the territories expressing
univin/vg1 and nodal/xnr are overlapping,
consistent with the finding that Univin and Vg1 factors act upstream of
nodal/Xnr expression
(Fig. 9). Furthermore, the
linkage between Univin and Bmp2/4 suggests that one gene derived from the
other by gene duplication. Finally, the strong regulatory interaction between
Univin and nodal suggests that an ancestral function of Vg1/Univin
might have been to regulate expression of nodal genes and that this
regulatory interaction may have been an evolutionary conserved early step in
the establishment of the dorsal-ventral axis of deuterostomes.
Homeodomain factors, bZIP, Oct and Myb: candidate regulators of nodal expression
SoxB1, TCF and Smads have already been implicated in patterning of the
ectoderm along the animal-vegetal axis of the sea urchin embryo
(Huang et al., 2000;
Kenny et al., 2003;
Yaguchi et al., 2007). In this
study, we showed that these maternal factors are also required for
dorsal-ventral axis formation upstream of nodal expression. In
addition, we identified several binding sites for potential novel regulators
of nodal expression including activating binding sites for
homeodomain, Oct and bZIP factors, along with a binding site for a repressor,
possibly Myb (Coffman et al.,
1997). It is intriguing to note that several of the candidate
transcription factors predicted to bind to this promoter are known to be
regulated by redox signaling, including bZIP
(Liu et al., 2005), Oct
(Guo et al., 2004;
Zheng et al., 2003) and Myb
(Bergholtz et al., 2001;
Brendeford et al., 1997;
Myrset et al., 1993). In
particular, a wealth of data is available on the role of bZIP transcription
factors as sensors of redox signaling downstream of MAP kinases
(Amoutzias et al., 2006;
Liu et al., 2005). It is
therefore tempting to speculate that in the sea urchin embryo, bZIP, Oct and
Myb act downstream of p38 and redox gradients to regulate nodal
expression (Fig. 10). Further
studies will be required to identify and characterize these transcription
factors. These studies are warranted because nodal is, to our
knowledge, the earliest gene displaying a restricted expression along the
dorsal-ventral axis in the sea urchin embryo. Analysing the regulatory circuit
driving nodal expression might therefore help to understand how
maternal information is integrated at the level of the promoter sequence of
regulatory genes to specify the secondary axis of polarity of the embryo.
Supplementary material
Supplementary material for this article is available at
http://dev.biologists.org/cgi/content/full/134/20/3649/DC1
|
ACKNOWLEDGMENTS |
|---|
|
Footnotes |
|---|
|
REFERENCES |
|---|
|
TOPSUMMARYINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Agius, E., Oelgeschlager, M., Wessely, O., Kemp, C. and De Robertis, E. M. (2000). Endodermal Nodal-related signals and mesoderm induction in Xenopus. Development 127,1173 -1183.[Abstract]
Amoutzias, G. D., Bornberg-Bauer, E., Oliver, S. G. and Robertson, D. L. (2006). Reduction/oxidation-phosphorylation control of DNA binding in the bZIP dimerization network. BMC Genomics 7,107 .[CrossRef][Medline]
Andersson, O., Reissmann, E., Jornvall, H. and Ibanez, C. F. (2006). Synergistic interaction between Gdf1 and Nodal during anterior axis development. Dev. Biol. 293,370 -381.[CrossRef][Medline]
Arnone, M. I., Dmochowski, I. J. and Gache, C. (2004). Using reporter genes to study cis-regulatory elements. Methods Cell Biol. 74,621 -652.[Medline]
Arnone, M. I., Bogarad, L. D., Collazo, A., Kirchhamer, C. V., Cameron, R. A., Rast, J. P., Gregorians, A. and Davidson, E. H. (1997). Green Fluorescent Protein in the sea urchin: new experimental approaches to transcriptional regulatory analysis in embryos and larvae. Development 124,4649 -4659.[Abstract]
Bellipanni, G., Varga, M., Maegawa, S., Imai, Y., Kelly, C.,
Myers, A. P., Chu, F., Talbot, W. S. and Weinberg, E. S.
(2006). Essential and opposing roles of zebrafish beta-catenins
in the formation of dorsal axial structures and neurectoderm.
Development 133,1299
-1309.
Ben-Haim, N., Lu, C., Guzman-Ayala, M., Pescatore, L., Mesnard, D., Bischofberger, M., Naef, F., Robertson, E. J. and Constam, D. B. (2006). The nodal precursor acting via activin receptors induces mesoderm by maintaining a source of its convertases and BMP4. Dev. Cell 11,313 -323.[CrossRef][Medline]
Bergholtz, S., Andersen, T. O., Andersson, K. B., Borrebaek, J.,
Luscher, B. and Gabrielsen, O. S. (2001). The highly
conserved DNA-binding domains of A-, B- and c-Myb differ with respect to
DNA-binding, phosphorylation and redox properties. Nucleic Acids
Res. 29,3546
-3556.
Bertocchini, F., Skromne, I., Wolpert, L. and Stern, C. D.
(2004). Determination of embryonic polarity in a regulative
system: evidence for endogenous inhibitors acting sequentially during
primitive streak formation in the chick embryo.
Development 131,3381
-3390.
Birsoy, B., Kofron, M., Schaible, K., Wylie, C. and Heasman,
J. (2006). Vg 1 is an essential signaling molecule in Xenopus
development. Development
133, 15-20.
Bradham, C. A. and McClay, D. R. (2006). p38
MAPK is essential for secondary axis specification and patterning in sea
urchin embryos. Development
133, 21-32.
Brendeford, E. M., Myrset, A. H., Hegvold, A. B., Lundin, M. and
Gabrielsen, O. S. (1997). Oncogenic point mutations induce
altered conformation, redox sensitivity, and DNA binding in the minimal DNA
binding domain of avian myeloblastosis virus v-Myb. J. Biol.
Chem. 272,4436
-4443.
Cartharius, K., Frech, K., Grote, K., Klocke, B., Haltmeier, M.,
Klingenhoff, A., Frisch, M., Bayerlein, M. and Werner, T.
(2005). MatInspector and beyond: promoter analysis based on
transcription factor binding sites. Bioinformatics
21,2933
-2942.
Chazaud, C. and Rossant, J. (2006). Disruption
of early proximodistal patterning and AVE formation in Apc mutants.
Development 133,3379
-3387.
Chen, C., Ware, S. M., Sato, A., Houston-Hawkins, D. E., Habas,
R., Matzuk, M. M., Shen, M. M. and Brown, C. W. (2006). The
Vg1-related protein Gdf3 acts in a Nodal signaling pathway in the
pre-gastrulation mouse embryo. Development
133,319
-329.
Cheng, S. K., Olale, F., Bennett, J. T., Brivanlou, A. H. and
Schier, A. F. (2003). EGF-CFC proteins are essential
coreceptors for the TGF-beta signals Vg1 and GDF1. Genes
Dev. 17,31
-36.
Child, C. M. (1948). Exogastrulation by Sodium Azide and other inhibiting conditions in Strongylocentrotus purpuratus. J. Exp. Zool. 107,1 -38.[CrossRef]
Coffman, J. A. and Davidson, E. H. (2001). Oral-aboral axis specification in the sea urchin embryo. I. Axis entrainment by respiratory asymmetry. Dev. Biol. 230, 18-28.[CrossRef][Medline]
Coffman, J. A., Kirchhamer, C. V., Harrington, M. G. and Davidson, E. H. (1997). SpMyb functions as an intramodular repressor to regulate spatial expression of CyIIIa in sea urchin embryos. Development 124,4717 -4727.[Abstract]
Coffman, J. A., McCarthy, J. J., Dickey-Sims, C. and Robertson, A. J. (2004). Oral-aboral axis specification in the sea urchin embryo II. Mitochondrial distribution and redox state contribute to establishing polarity in Strongylocentrotus purpuratus. Dev. Biol. 273,160 -171.[CrossRef][Medline]
Czihak, G. (1963). Entwicklungsphysiologische Untersuchungen an Echininiden (Verteilung und bedeutung der Cytochomoxydase). Whilhem Roux's Arch. Entwickl. Mech. Org. 154,272 -292.[CrossRef]
Czihak, G. (1971). Echinoids. In Experimental Embryology of Marine Invertebrates (ed. G. Reverberi), pp. 363-482. Amsterdam: North-Holland.
Dohrmann, C. E., Kessler, D. S. and Melton, D. A. (1996). Induction of axial mesoderm by zDVR-1, the zebrafish orthologue of Xenopus Vg1. Dev. Biol. 175,108 -117.[CrossRef][Medline]
Duboc, V. and Lepage, T. (2006). A conserved role for the nodal signaling pathway in the establishment of dorso-ventral and left-right axes in deuterostomes. J. Exp. Zoolog. B Mol. Dev. Evol. doi: 10.1002/jez.b.21121.
Duboc, V., Rottinger, E., Besnardeau, L. and Lepage, T. (2004). Nodal and BMP2/4 signaling organizes the oral-aboral axis of the sea urchin embryo. Dev. Cell 6, 397-410.[CrossRef][Medline]
Duboc, V., Rottinger, E., Besnardeau, L., Lapraz, F. and Lepage, T. (2005). Left-right asymmetry in the sea urchin embryo is regulated by Nodal signalling on the right side. Dev. Cell 8,1 -12.[Medline]
Giese, K., Amsterdam, A. and Grosschedl, R.
(1991). DNA-binding properties of the HMG domain of the
lymphoid-specific transcriptional regulator LEF-1. Genes
Dev. 5,2567
-2578.
Guindon, S., Lethiec, F., Duroux, P. and Gascuel, O.
(2005). PHYML Online - a web server for fast maximum
likelihood-based phylogenetic inference. Nucleic Acids
Res. 33,W557
-W559.
Guo, Y., Einhorn, L., Kelley, M., Hirota, K., Yodoi, J.,
Reinbold, R., Scholer, H., Ramsey, H. and Hromas, R. (2004).
Redox regulation of the embryonic stem cell transcription factor oct-4 by
thioredoxin. Stem Cells
22,259
-264.
Helde, K. A. and Grunwald, D. J. (1993). The DVR-1 (Vg1) transcript of zebrafish is maternally supplied and distributed throughout the embryo. Dev. Biol. 159,418 -426.[CrossRef][Medline]
Hörstadius, S. (1973). Experimental Embryology of Echinoderms. Oxford: Clarendon Pres
