修士学位論文
題名Repression of the transcription of扱gene by Noda1
signa1ing in the d−eve1opment ofαb刀∂∠批θ曲加ムh
カタユウレイボヤの発生における放遺伝子のNoda1
signa1ingによる転写抑制（英文）
指導教授 西駕 秀俊 教授
平成 25 年 1 月 ユO 目 提出
首都大学東京大学院
理工学研究科
学修番号
生命科学
11881319
専攻
氏 名 末吉 美佐
学位論文要旨（修士（理学））
論文著者名 末吉 美佐 論文題名1Repressionofthetranscriptionof伽genebyNoda1signa1inginthedeve1opmentof
Cゴ。〃αゴnτe8が〃α方∫．
（邦題） 力タユウレイボヤの発生における伽遺伝子のNoda1signa1ingによる転写抑制 （英文）
本文
左右非対称な形態形成は、脊椎動物をはじめとする多くの動物の体制の特徴である。左 右非対称性の成立には、マウス胚でのノード流などの対称性を破る機構、そして、左右非 対称に発現するシグナル伝達因子Noda1、その下流の転写因子遺伝子P炊2からなる遺伝子 発現のカスケードが重要だとされている。しかし、非対称なNoda1の発現と非対称な形態 形成をつなぐ機構についての情報は乏しい。ホヤは脊索動物門・尾索動物亜門に属し、そ の幼生は脊椎動物と類似した体制と発生様式を持っている。ホヤの」種カタユウレイボヤ C1o舳肋鮒加α脆は、ゲノム情報が利用出来ること、そしてレポーター遺伝子の導入も容易 であることから、転写調節領域の解析に適している。そのカタユウレイボヤでは、尾芽胚 期の左側表皮において、Noda1，FoxH，Pitxの遺伝子発現のカスケードが脊椎動物と同様に 働いていることが、当研究室の先行研究によって明らかにされている（Yoshida and Saiga，
2009）。さらに左側のNoda1シグナリングは、尾芽胚期に脳胞右側に生じる眼点（光受容器 官）の形成に関与することが明らかにされている。眼点の形成にはRetina1homeobox遺伝
子柵の右側での発現が必須である（D，anie11o et a1．，2004）。Noda1シグナリングは、伽の転写 を左側で抑制し、伽の右側特異的な発現に関わっている（Y1oshida and Saiga，2012）。伽の発
現制御機構については、伽遺伝子の上流3kbの領域が内在性の発現を再現すること、伽の 転写には上流590〜453bpに存在するC1ox／Onecut結合酉己列が重要であることが明らかにされ ている（D Anie11o eta1．，2004；2011）。しかし、ルの転写がどのようにしてNoda1シグナリング によって抑制されるのかは不明である。本研究では、柵の上流解析を行い、その発現を負 に制御する領域を同定することにより、Noda1による伽の転写抑制機構を明らかにするこ
とを目的とした。
カタユウレイボヤ受精卵へのエレクトロポレーション法による遺伝子導入には、受精卵 の卵膜を除去する必要がある。カタユウレイボヤと異なる目に属するマボヤHαZ00γn〃α 707物1では、神経胚期に胚が卵膜内で回転すること、回転が止まって卵膜と接触した側が 左側となることが報告されている（Nishide et a1．，2012）。また、カタユウレイボヤでも、
めに、卵膜内で発生した胚と、受精直後に卵膜を除去して発生させた胚について、尾芽胚 期にwho1e mount in situ hybridization（WISH）を行って伽とMo〃mRNAの発現パターンを 比較した。その結果、卵膜内で発生した胚では、ほぼ全ての胚で伽は正中に対して右側特 異的に、Mo〃は左側特異的に発現するのに対し、卵膜除去胚では伽、Mo〃が左右両側で 発現、伽が左側で発現、NOaα1が右側で発現するなど、発現パターンに乱れが生じること が分かった。
そこで、伽の発現を負に制御する領域を同定するために、遺伝子の上流域をGFP遺伝子 につないだレポーターコンストラクトを卵膜を除去した受精卵に導入し、Noda1強制発現 下とNoda1シグナル阻害下で、GFPレポーターの発現レベルを比較することにした。より 具体的には、脳胞全体でNoda1を強制発現するコンストラクトCra1bp−Noda1をレポーターコ
ンストラクトと共に導入した胚と、さらにNoda1シグナル受容体阻害剤であるSB431542を 加えた胚との間で、GFPmRNAをWISHで検出した時の染色強度を定性的に比較した。も し伽遺伝子の発現を負に制御する領域がレポーターコンストラクトに含まれるならば、
GFPレポーターの発現はNoda1強制発現下では抑制されNoda1シグナル阻害胚では抑制され ないため、両者の発現レベルには差があり、負に制御する領域が含まれないときには、両 者の発現レベルに差がなくなると期待される。
最初に、上流3kbpを含むGFPレポーターコンストラクト内に伽の発現を抑制する領域が 存在するかどうかを調べた。その結果、Noda1強制発現コンストラクトと上流3kbpレポー ターコンストラクトを共導入した強制発現胚と、阻害剤によるNoda1シグナル阻害胚の GFP mRNAのレベルには、差があることが分かった。同様の実験を上流0．7kbpコンストラ
クトについて行ったところ、GFPmRNAのレベルに差が観察された。従って、Noda1シグナ リングに応答し、伽の発現を抑制する領域は、上流0．7kbp内に存在すると考えられる。鮒 遺伝子の上流0．7kbの中には、正の発現調節を行うことが分かっているC1oxとOnecut転写因 子の結合配列があり、2つの結合配列の間にTG耶／BMPシグナル伝達因子Smadの結合配列 が存在する。このSmad結合酉己列に変異を導入すると、Noda1強制発現胚でのGFP mRNAの レベルは、シグナル阻害胚のものに近づいた。このことから、伽の転写の抑制にはC1ox／
Onecut間に存在するSmad結合酉己列が重要であることが示唆された。
Repression of the transcription of RM gene by Nodal signaling in the development of Ciona intestinalis.
A Thesis
Submitted for the degree ofMaster of Science Tokyo Metropolitan University
By
Misa Sueyoshi
Department ofBiological Sciences, Graduate School of Science an Tokyo Metropolitan University
d Engineering,
2013
Contents
Acknowledgements
Abstract 4 Introduction 6
Materials and Methods
Results 13
Discussion 19 Reference 23 Figures 26
Supplemental figure
3
35 9
Acknowledgements
I thank Professor Hidetoshi Saiga of Tokyo Metropolitan University for his guidance and encouragement throughout the course ofthis study. Thanks are also due to Professors Kimiko Fukuda and Naohito Takatori, and Dr. Keita Yoshida for their comments and suggestions. I am gratefu1 to all members of the Laboratory of Developmental Programs for valuable advice.
Abstract
Left-right asymmetry is a fundamental character for chordates' body plan. It is well-known that Nodal signaling plays an essential role in establisimeRt ofthe left-right
asymmetry; Nodal, encoding a secreted signaling protein, is induced to express asymmetrically by so-called symmetry-breaking event such as nodal flow in the node of mouse embryos, and in turn, left-sided Nodal signaling induces left-side specific gene
expression such as a transcription factor gene Pitx, followed by asymmetric
morphogenesis. However, the mechanism that connects the asymmetrical Nodal signaling and asymmetrical morphogenesis remains largely unlmown.
Ascidians, belonging to the subphylum Urochordata, are regarded as the closest relative of vertebrates (Putnam et al., 2008). Their larvae, consisting of about 2,600
cells, share a characteristic body plan of chordates with a hollow dorsal neural tube, a
notochord and paraxial mesoderm. Left-right asymmetry can also be seen in the simple ascidian development; formation ofthe ocellus, a light-sensing organ, is an example of left-right asymmetric development in the ascidian, Ciona intestinalis. In C. intestinalis,
Nodal signaling on the left side leads to asymmetrical formation of ocellus during tailbud stages. The ocellus is located on the right side of the trunk, the head of an ascidian tadpole-shaped larva. For its formation, it has been reported that the function of
Ci-Rx, the retinal homeobox gene is required and that transcription of Ci-Rx is positively regulated by transcription factors, Clox and Onecut (D'aniello et al.,2006;
2011). Moreover, it has been shown that Nodal signaling on the left side represses the
transcription of Ci-Rx, which otherwise would take place on both sides, leading to the
formation of the ocellus on both sides (Yoshida and Saiga, 2011). However, it is unlmown how the transcription of Ci-Rx is repressed by the Nodal signaling on the left
side.
In the present thesis, how the repression of Ci-Rx gene transcription is exerted
by the Nodal signaling was addressed. For this, I canied out GFP reporter assay with upstream region ofthe Ci-Rx gene using C. intestinalis embryos. Since sidedness ofthe
Ci-Rx expression is known to be perturbed by dechorionation that is an inevitable '
process for the electroporation upon reporter assay in C. intestlnalis, first I established a
'
method to identify negative transcriptional regulatory region of the Ci-Rx gene using
Nodal overexpressed embryos and Nodal signaling-inhibited embryos, according to the following assumptions. If the negative transcriptional regulatory region is present in an
Rx reporter construct, the expression of the reporter will be repressed in Nodal overexpressed embryos but not in Nodal signaling-inhibited embryos. On the contrary, if it is absent, levels of the expression of the reporter will be the same or similar to each
other between the two experimental groups of embryos. Using this method, I identified that Ci-Rx O.7k upstream region as the negative transcriptional regulatory region of Ci-
Rx. I further found out that a binding site for the TGFP/BMP intercellular signal transducer Smad is located at O.4k upstream within the region and that the Smad binding site is likely involved in the repression of Ci-Rx by Nodal signaling.
Introduction
Left--righi asymmetry is a fundamental feature of vertebrates and invertebrates (Brown and Wolpert, 1990; Levin, 2005). In the establishnent of left--right asymmetry,
Nodal signaling plays an essential role, regulating a homeodomain transcription factor,
Pitx2. Nodal-Pitx gene expression cascade is also essential for left-right asymmetric
morphogenesis, such as unidirectional looping of the heart and gut in vertebrate development (Capdevila et al., 2000; Hamada et al., 2002; Shiratori et al., 2001). The
left--right asymmetric Nodal-Pitx cascade can also be found in deuterostome animals, such as ascidians, amphioxus and sea urchins (Boorman and Shimeld, 2002a; Duboc and Lepage, 2008; Morokuma et al., 2002; Yoshida and Saiga, 2008; Yu et al., 2002), and even in snails (Grande and Patel, 2009). However, it is largely unclear how left-
right asymmetric morphology is established by the asymmetric Nodal signaling and what are the downstream target genes ofNodal signaling that leads to the asymmetric morphogenesis.
Ascidians belong to the subphylum Urochordata and are regarded as the closest relative of vertebrates (Putnam et al., 2008). Their larvae share a characteristic body
plan of chordates with a hollow dorsal neural tube, a notochord and paraxial mesoderm.
In an ascidian species, Ciona intestinalis, which has recently become a popular model
animal, plenty of genome information of its compact genome is available and the electroporation method to easily introduce a reporter construct has been established.
Therefore, C. intestinalis is suitable for the analysis of transcriptional regulatory mechanisms, including those conceming left-right asymmetry.
Left--right asymmetry is apparent in the morphogenesis of C. intestinalis development, which includes unidirectional coiling of the elongating tail of the tailbud
embryo in the chorion, unidirectional looping of the alimentary tract in juveniles, and
right-sided positioning of the ocellus pigment cell in the sensory vesicle (SV), the anterior most part of the larval central nervous system (CNS). The ocellus is a light-
sensing organ, consisting of a pigmented cell, three lens cells and about 20 photoreceptor cells, located on the right side of SV (Nicol and Meinertzhagen, 1991).
During tailbud stages, the cells that have potential to express retinal homeobox gene are
once aligned on the anterior dorsal midline of the trunk- and then they move laterally.
The cell that moves rightwards but not leftwards becomes the ocelius pigment precursor
to express Ci-RJc and, together with the neighboring cells, eventually forms the ocellus.
The other cell does not express Rx, and hence does not form the ocellus, which is suggested to be exerted by the left side specific Nodal signaling (Yoshida and Saiga, 2011).
Rx gene, belonging to the paired-like homeobox gene family, is known to play a critical role in the eye development in vertebrates (Mather et al., 1997). In C.
intestinalis, it has been reported that the ocellus development is depend on Ci-Rx gene
function and suggested that the transcription of Ci-Rx is under control of the transcription factors, Onecut and Clox (D'aniello et al., 2006; 2011). As stated above,
the transcription of Ci-Rx should be repressed by the Nodal signaling on the left side,
but the mechanism how the transcription of Ci-Rx is repressed remains unknown.
In the present thesis, I addressed how the transcription of Ci-Rx is repressed by
Nodal signaling using reporter assay. Since right-sided expression of Ci-Rx and left-
sided expression ofNodal are perturbed by dechorionation, the necessary procedure for introducing reporter construct into C. intestinalis embryos, I designed an experimental
method that allows identification of the negative regulatory region by comparison ofthe
levels of Rx reporter gene expression between in Nodaloverexpressed (OE) and in Nodal signaling-inhibited (SI) embryos. I found out that O.7kbp upstream region from the translation start site of the Ci-Rx gene negatively regulates Ci-Rx transcription, in
which it seems to be important that a Smad transcription factor binding site located at
O.4kbp upstream from the translation start site of the Ci-Rx gene. These findings revealed a part of the mechanism connecting the left-sided Nodal signaling to the left-
right asymmetrical ocellus formation.
Materials and methods
Ascidians
Adults C. intestinalis were provided by the Maizuru Fisheries Research Station of Kyoto University and Misaki Marine Biological Station of University of Tokyo through the National Bio-Resource Project (NBRP) of the MEXT, Japan. Eggs and
sperm were obtained surgically from the gonoducts. After insemination, eggs were dechorionated, electroporated and cultured in filtered natural seawater at 180C. Embryos
were allowed to develop and fixed in 40/oPFA.
Preparation of reporter constructs
To create the Ci-Rx upstream-GFP reporter constructs, various genomic
upstream regions of Ci-libc were isolated by PCR using C. Intestinalis genomic DNA as
template and inserted into the Sall/Smal sites of the pPD46.21-GFP vector. This had been prepared by inserting GFP coding sequence excised from the plasmid GFP.RN3 (Zernicka-Goetz et al., 1996) into the Kpnl/EcoRI sites of the pPD46.21 vector, which is a variant ofpPD1.27 (Fire et al. 1990).
To create the internal control construct Pitx Dl module-GFP, first, the genomic
region of Ci-Piix Dl module (Christeaen et al., 2010) was amplified by PCR from C.
intestinalis genomic DNA and inserted into the Hindlll/Pstl site of pDP46.21-GFP vector. Second, the basal promoter region (233bp in length) of C intestinalis Forkhead gene (Harafuji et al., 2002) was inserted into the Sall/BamHI sites.
Making ofthe Cl-Nodal over-expression construct, Cralbp-Nodal, was described previously (Yoshida and Saiga, 2011).
Electroporation of reporter constructs '
Electroporation of reporter constructs into fertilized eggs was carried out essentialiy as described previously (Corbo et al., 1997). Fifteen min after insemination,
eggs were treated with 10 mi of seawater containing 10/o sodium thioglycolate and O.05
O/o Actinase E added freshly with 400 pl of IN NaOH for 5 min to remove the chorion.
Immediately after the treatment, the eggs were washed in seawater twice and in the solution consisting of 9 volume of O.693 M marmitol and lvolume of seawater (M/S solution) twice, and 150 pl of the egg suspension was transferred into a 2 mm-width cuvette, followed by immediate mixing with 210 pt1 of O.693 M mannitol solution and 40 pl TE containing 60 ptg of DNAs, including reporter (Rx us-GFP), Nodal expression (Cralbp-Nodal) and internal control (Pitx Dl module-GFP) plasmid DNAs in equal molar ratio. Electroporation was carried out with either decrement pulse or square pulse
applied under the following conditions; decrement pulse with the voltage of 25 V, capacitance of200 ptE resistance of ooS;l!, or one square pulse with the voltage of30 V
for 10msec. After electroporation, the eggs were cultured in natural seawater at 18-200C.
Preparation and fixation of Nodal inhibited (SI) embryos
.
over-expressmg (OE) and Nodal signaling
Embryos electroporated with Rx upstream-GFP reporter, Cralbp-Nodal, and Pitx Dl module-GFP constructs were divided into two groups at the Reurula stage. One group of embryos (SI embryos) was cultured in seawater added with 1/1000 volume of 5mM SB431542 to give final concentration of 5ptM, and the other group of embryos (OE embryos) was cultured in sea water containing O.1 O/o DMSO. Embryos of both groups were allowed to develop to the mid tailbud stag and fixed with the mixture consisting of40/oPFA, O.5M NaCl, O.IM MOPS pH7.5 in seawater at 40C overnight.
VVhole-mount in situ hybridization
Whole-mount in situ hybridization (WISH) was carried out using DIG labeled RNA probes as described previously (Ikuta and Saiga, 2007; Ikuta et al., 2004). The probe RNA was added to the hybridization buffer to give a final concentration of O.5yg/
ml for every experiment. RNA probes for Ci--Cralbp, and Ci-0necut were prepared using plasmid DNA obtained from C. intestinalis gene collection release 1 (http://
www.ghost.zool.kyoto-u.ac.jp/indexrl.html). DNA fragment for probe synthesis of Ci- Rx and Ci-Nodal were obtained through RT-PCR using total RNA prepared from tailbud stage embryos, and were cloned into the EcoRI/Clal sites and Spel/Xhol sites of the pBluescript KS+ vector, respectively (Yoshida and Saiga, 2008, 2011). A DNA fragment for probe synthesis of GFP was excised from the plasmid GFP.RN3 (Zemicka-Goetz et al., 1996) and cloned into the Kpnl/EcoRI site ofthe pBluescript KS+ vector.
Primers
Primers used were listed as below.
Rx-up3000-sall Rx us1834 sall Rx usM88 sall Rx us77t sall Rx +13 smal
'
,Rx uSo.7ks'mi dMT
• F-
Rx usO.7ksmadMT -R
PitxDl F PitxDl R
5'-CCGTCGACTTTTCCAATCGTGTT-3' 5'-CAGTCGACACAACCAAACCATAC-3' 5'-CAGTCGACACGTATTGTTAGATT-3' 5'-TATGCCGTGTTTTAGTCTATTAC-3'
' 5'-
'TTCCCGGGTGTCTGTACTCATCTTCAAGGA-3' 5'-ATTAATAGAAGGAGTATAGCAGTGCAT-3'
5'"-ATGCACTGCTATACTCCTTCTATTAATTAT-3' 5'--CCAAGCTTAGCGGCGACATTTTACT-3'
5'-AACTGCAGAGCCTACACCGAAArATAAACA-3'
Results
Asymmetric expression of Ci-I?Lx and Ci-Nodat is perturbed in dechorionated embryos
In the reporter assay using embryos of C. intestinalis, dechorionation that is to
remove the chorion (egg coat), is necessary to introduce a reporter construct into fertilized eggs by electroporation. It is known that the chorion plays an important role
for determination of left-right asymmetric gene expression and morphogenesis in ascidian development. For example, when eggs were dechorionated and allowed to
develop, the left-sided expression of Ci-Pitx at the- tailbud becomes randomized (Yoshida and Saiga, 2008). Simiiarly, the iooping of the gut, which should take place iR
the right to the branchial basket in thejuvenile (when viewed frontally) in the normal
development, becomes perturbed by dechorionation of the fertilized egg (unpublished observation). Therefore, it is expected that dechorionation affect on the expression of
Ci-Rx, which has been reported to be expressed in the right sensory vesicle (SV) at the
tailbud stage (D'aniello et al., 2006) and Ci-Nodal, which is expressed in the left epidermis and SV at the tailbud stage (Yoshida and Saiga, 2011). Hence, prior to the transcriptional analysis of Ci-Rx, I re-examined the expression of Ci-Rx and examined
effects of the dechorionation on the endogenous expression of Ci-Rx as well as Ci- Nodal at the tailbud stage.
In normal developing embryos with the chorion, Ci-Rx mRNA was detected by WISH in the prospective SV region bilaterally at the early tailbud stage. The expression
became down regulated on the left side as the development proceeded. In the mid tailbud stage, Ci-Rx.mRNA was detected only on the right side of SV as previously
reported (Fig. IA and Bs D'aniello et al., 2006). Also, Cl-Nodal mRNA was detected in the left epidermis at early and mid tailbud stages as previously reported (Fig. IC and D;
Yoshida and Saiga, 2011). By contrast, in the embryos which had been dechorionated after insemination, though the asymmetric expression of Ci-Nodal was observed at the early tailbud stage, asymmetric expression of Ci-Rx as well as Ci-Nodal was perturbed at the mid taiibud stage. As shown in Fig. 1, dechorionated mid tailbud stage embryos exhibited abiated (Fig. IE and H), bilateral (Fig. IF and I), and less often reversed (Fig.
1G and J) expression of Ci-Rx and Ci-Nodal. Therefore, in order to do the reporter assay
to find out the negative transcriptional regulatory region of Cl-Rx, I had to design the
experimental conditions that allow the reporter assay without perturbation due to the dechorionation.
Designing of the assay to allow the identification of negative transcriptional regulatory region of Ci-Rrc in response to the Nodal signaling
In order to overcome the perturbation due to dechorionation on Ci-Rx expression that leads to its occasional loss and to successfu11y detect the negative regulatory region
of Ci-Rx in response to the Nodal signaling, I decided to compare the Rx reporter expression between in Nodal-overexpressed (OE) embryos and in Nodal signaling- inhibited (SI) embryos. For Nodal over-expression, the construct, Cralbp-Nodal, driving
Ci-Nodal cDNA under the control ofthe 3kb upstream region of Ci-Cralbp, was used to express Ci-Nodal and repress Ci-Rx in the whole SV (Fig. 2B; Yoshida and Saiga, 2011). For Nodal signaling inhibition, SB431542 that is an inhibitor of ALK4/5/8 (Inman et al., 2002) was applied to the OE embryos. In SB431542 treated embryos, Ci-
Rx mRNA is expressed bilaterally in the SV at the mid tailbud stage (Yoshida and Saiga
2011). Similarly, in SB431542-treated OE embryos, expression of Ci-Rx was retained
bilateral.,(Fig. 2C).
On this reporter assay, what one can expect is as follows. When the Rx reporter construct includes a negative transcriptional regulatory region, levels of Rx reporter
gene expression will be different between Nodal OE and Nodal SI embryos; the latter will be higher than the former, because it is expected as such that the reporter expression must be repressed by Nodal signaling in OE but not SI embryos. By contrast,
when the Rx reporter construct does not include the negative regulatory region, levels of
the reporter gene expression wiH be the same or simiiar between in OE and SI embryos.
In addition, there was one another problem in the present reporter assay. That is, when
electroporated into a fertilized egg, molecules of a reporter construct are not necessarily
spread over the cytoplasm uniformly, and hence within later stage embryos, some cells
may receive enough number of molecules to detect the reporter expression and others may not. In ordinary reporter assay, this would not be a serious problem. In this reporter
assay, however, this will be a problem, since one could not discriminate the loss of the
reporter expression due to the repression by Nodal signaling from the loss ofexpression
due to insufficient distribution of the reporter construct upon electroporation. To solve
this problem and to successfu11y discriminate Rx reporter-repressed embryos, Pitx Dl module-GFP reporter was co-introduced as an internal control (Fig. 2A). This reporter
construct drives the expresSion of GFP at the anterior neural boundary, which is very
'
close to the endogenous expression domain of Ci-Rx and more importantly, the '
expression driven by Pitx Dl module is not affected at all by the Nodal signaling
(Christiaen et al., 2007; Yoshida and Saiga, 2008). By introducing Pitx Dl module-GFP reporter together with the Rx reporter construct, expression levels as estimated from the
staining intensity upon WISH ofthe reporter GFP mR[NA can be evaluated qualitatively by comparing to expression levels ofthe internal control.
Deletion analysis of Ci-RAc upstream regions
I performed deletion analysis of upstream region of Ci-Rx by using GFP reporter constructs in combination with the Pitx Dl module-GFP reporter. Previously, it was reported that 2.9kbp upstream region was able to reproduce the endogenous expression of Ci-Rx at the tailbud and larva stages (D'anieilo et al.,2006). I prepared four reporter
constructs, which were us3k-GFP, us2k-GFP, uslk-GFP and usO.7k-GFP constructs,
harboring 3kb, 2kb, lkb and 0.7kb upstream region, respectively, and 13 bp from the first nucleotide A ofthe translation start site (Fig. 3A).
First, us3k-GFP reporter was introduced into fertilized eggs and the resulting
embryos were examined by WISH at the mid tailbud stage. This construct was capable of driving expression in the dorsal trunk, though laterality must be perturbed by dechorionation prior to electroporation (Fig. 3B; Supplemental fig. 1). Then, expression
levels of the GFP reporter us 3k-GFP were compared between in OE and SI embryos.
As shown in Fig. 3C, expression of GFP reporter in Nodal OE but not in SI embryos was largely repressed. In the shortest construct us O.7k-GFP, when expression levels of
the GFP reporter us3k-GFP were compared between in OE and SI embryos, expression level was higher in SI embryos than in OE embryos (Fig. 3D), though the difference in
the expression levels between in SI and in OE was less pronounced in comparison of
usO.7k-GFP with us3k-GFP. These results regulatory region of Ci-Rx in response to No the O.7kbp upstream region.
suggest that a negative transcriptional dal signaling seems to be localized within
Smad binding site in the e.7kbp upstream region may be involved in the negative transcriptional regulation in response to the Nodal signaling
I searched for a putative transcription factor binding site(s) that acts negatively
on the transcription of Ci-Rx in response to the Nodal signaling in the 0.7kbp upstream
regionusingsoftware,TFsearch(lteq;t;4pt2suci2xg4pltgstt//bb / h/db/TFSEARCHJht1)and TESS (http://www.cbil.upenn.edu/cgi-bin/tess/tess), and found a putative Smad binding
site (Smad BS) localized at the 473bp upstream. This position is within a highly conserved region (603 to 407bp upstream region) between C. intestinalis and its closely
related species, C. savignyi, and is in between two transcription factor binding sites for
Clox and Onecut, which have been shown required for the transcription of Ci-Rx in the ocellus development (D'aniello et al., 2011). I focused on the Smad BS that transduces
TGFP/BMP intercellular signaling (Massague and Wotton, 2000; 2005).
To examine the importance of the Smad BS in repression of Ci-Rx transcription,
the Smad BS was mutated from AGAC to GAGT in the reporter construct usO.7k-GFP, which resulted in the reporter construct, us O.7k-muSmad-GFP(Fig3A). Then, the
mutated construct was introduced into embryos and examined for the expression level between Nodal OE and SI embryos at the mid tailbud stage. It was found that the reporter expression level was higher in Nodal SI embryos than in OE embryos (Fig.3E).
However, when the levels of the reporter expression exhibited by us O.7k-muSmad-
GFP were compared with those exhibited by us O.7k-GFP reporter, both of the expression levels in Nodal OE and SI embryos were higher in Smad BS mutated
construct (Fig. 3D and E). This means that in the absence of the Smad BS, expression of
Ci-Rx is facilitated, suggesting that, though the extent of the repression seems to be modest, the Smad BS is involved in the negative transcriptional regulation of Ci-Rx in response to the Nodal signaling.
Discussion
In the present thesis, I studied negative transcription regulatory mechanism of Ci-Rx exerted by Nodal signaling. First, I revealed that expression of Ci-Rx as well as
Ci-Nodal was perturbed by the dechorionation, which is the necessary procedure for the
reporter assay using C. intestinalis embryos. Second, in order to identify negative transcriptional regulatory region by reporter assay, I established a method, in which
reporter expression is compared between Nodal over-expressed (OE) and Nodal
signaling-inhibited (SI) embryos with reference to internal control construct, Pitx Dl
module-GFP. By using this method, 0.7kb upstream region from the translation start site
is likely able to negatively reguiate the transcription of Ci-libc in response to Nodal
signaling. It was further suggested that a Smad binding site present within the O.7kb
upstream region may be involved in transcriptional repression of Ci-Rx. This result connects the part of asymmetrical Nodal expression and asymmetrical morphogenesis (Fig.4). However, there are some points to be unsolved.
Which Smad regulates Ci-Rbe transcription
In the present study, it was not clarified which Smad binds to the Smad BS in O.7kb upstream region. In C intestinalis, 5 Smad genes are present in the genome,
Smad7/5, Smad2/3a, Smad2/3b, Smad4, and Smad6/7, according.to Ciona cDNA
database Ghost (htt :// host.zool.k oto-u.ac.' /c i-binl b2/ browse/}shD (Hino et al.,
2003). Smad gene family can be classified into three types based on their function, Co-
Smad, I-Smad, and R-Smad (Massague and Wotton, 2000; 2005). Co-Smad, Smad4,
acts as a common partner for R-smads, and I-Smad, Smad6/7, acts as a R-Smad
inhibitor in the cytoplasm. R-Smads, including Smadl, 2, 3, 5, 8 and 9, act as a signal
transducer and can regulate downstream gene expression. (Massague and Wotton, 2000;
2005).
Accordong to the data bases of C. intestinalis, transcripts of Smadl/5 mRNA are detected at the neurula and early tailbud stages in the endoderm but not in the nervous
system. Smad2/3a seems to be expressed from cleavage and gastrula stage, but not to be expressed at the neurula and tailbud stage. Only Smad2/3b is expressed in the muscle at
the neurula stage and in the nervous system at the early and mid tailbud stages.
Considering the expression pattern of Ci-Rx in normal developmeRt, which is bilateral at the early tailbud stage and right-sided at the mid tailbud stage, the repression of Ci-Rx
expression on the left side likely starts after the early tailbud stage. Considering these, it
is suggested that Smad 2/3b is a likely candidate for the negative transcriptional regulator of Ci-Rx in the left SV.
How putative Smad negatively regulates Ci-Rx transcription
Sufficient DNA binding affinity of Smad transcription factor is accomplished by
coordination with Smad-interacting proteins like FoxH (Zawel et al.,1998). In C.
intestinalis, Nodal signaling regulates transcription of Pitx via FoxH (Yoshida and Saiga, 2008). In the present study, however, genomic upstream region of Ci-Rx contains
only single FoxH binding site localized at the 2.6k upstream from the translation start
site. In the future study, importance ofthis site should be addressed.
One interesting possibility may be the location ofthe Smad BS in O.7kb region.
It has been reported that Ci-Rx is regulated by two transcription factors, Onecut and Clox, and that the Onecut but not Clox binding site (BS) in the O.7kb region is essential
for the expression of Ci-Rx in the ocellus (D'aniello et al., 2011). The Onecut BS is localized in the close vicinity of 20bp upstream from the Smad BS. It is possible that
Smad binding may interfere with Onecut binding, which leads to the negative
transcriptional regulation of Ci-Rx.
Upstream region may also be important for negative transcriptiona! regu!ation of Ci-Rnc
When compared us3k with usO.7k in OE and SI embryos, the extent of negative transcriptional regulation is much evident in us3k. This suggests that the putative Ci-Rx
positive and negative regulatory sites may also be present between 3k and O.7k upstream region. In fact, in this region, nine Smad BSs as well as two Otx BS, known as
a positive regulator of Rx in Xenopus (Danno et al., 2008). Thus, it is suggested that
though O.7kb upstream region seems to include essential transcription factor BSs for transducing repressive Nodal signaling, such BSs must be present in other region than the O.7kb upstream region to exert efficient repression of Ci-Rx.
Another point to be addressed further
I found out that expression patterns of us 3k-GFP reporter and endogenous Rx expression in response to the Nodal signaling inhibition were fairly different. In SI embryos, transcripts of endogenous Rx were detected bilaterally in most embryos (Fig.
2B). By contrast, in SI embryos, transcripts of GFP were detected most often on the midline ofthe trunk (Supplemental Fig.1). At present there is no plausible explanation,
but this point should be addressed by determining which cells express Ci-Rx and/or reporter more precisely.
Implication of Nodal signaling for the asymmetric morphogenesis in C. intestinatis In this thesis, I revealed possibility that Nodal signaling directly interact with
asymmetrical morphogenesis by repressing transcription of Ci-Rx, the master gene of ocellus formation. Yoshida and Saiga (2011) suggest that Nodal signaling involves the
asymmetric SV morphogenesis (Yoshida and Saiga, 2011). Additionally, Mita et al.
(2010) identified genes of type IV collagenl/3/5, laminin-a5, and Prickle as targets positively regulated by Nodal signaling and genes of glypican and il-protocadherin- like as negative targets of Nodal signaling by microarray analysis (Mita et al.,2010).
Thus, it is highly likely that asymmetrical morphogenesis by Nodal signaling involves controlling not only transcription of Ci-Rx and Ci-Piix but also these non transcription
factor genes.
References
Boorman, C., J, Shimeld, S,M., 2002, The evolution of Ieft-right asynrmetry in chordates.Bioessays24, 1004-1O11.
Brown, N.A., Wolpert, L., 1990, The development ofhandedness in left/right asymmetry. Development 109, 1-9.
Capdevila,J.,Vogan,K.J.,Tabin,C.J.,IzpisuaBelmonte,J.C., 2000,
Mechanisms of left- right determination in vertebrates. Cell 1 O1 , 9-2 1 .
Corbo, J. C., Levine, M. and Zeller, R. W., 1997, Characterization ofa notochord- specific enhancer from the Brachyury promoter region ofthe ascidian, Ciona intestinalis. Development 124, 589-602.
Da"no, H., Michiue, T., Hitachi, K., Yukita, A., Ishiura, S., Asashima, M., 2007, Molecular links among the causative genes for ocular malformation: Otx2 and Sox2 coregulate Rax expression. PNAS 105, 5408-5413.
D'Aniello, S., D'Aniello, E., Locascio, A.,Memoli, A., Corrado, M., Russo, M.T., Aniello, F., Fucci, L., Brown, E.R., Branno, M., 2006, The ascidian homolog ofthe vertebrate homeobox gene Rx is essential for ocellus development and function.
Differentiation 74 222-234.
D'Aniello, E., Pezzotti, M.R., Locascio, A., Branno, M., 201 1, Onecut is a direct neural-specific transcriptional activator of Rx in Ciona intestinalis. Dev. Biol. 355 358-371 .
Duboc, V., Lepage, T., 2008, conserved role for the nodal signaling pathway in the establisiment ofdorso-ventral and left-right axes in deuterostomes. J. Exp. Zool. B Mol. Dev. Evol. 310, 41-53.
Fire, A., Harrison, S.W., Dixon, D., 1990, A modular set oflacZ fusion vectors for studying gene expression in Caenorhabditis elegans. Gene 93 189-198.
Grande, C., Patel, N.H., 2009, Nodal signalling is involved in left-right asymmetry in snails. Nature 457, 1O07-1O11.
Haeussler, M., Jaszczyszyn, Y., Christiaen, L., Joly, J.S., 201O, A cis-regulatory signature for chordate anterior neuroectodermal genes. Plos Genet.6(4).
Hamada, H., Meno, C., Watanabe, D., Saijoh, Y., 2002, Establisiment ofvertebrate left- right asymmetry. Nat. Rev. Genet, 3, 103-113.
Harafuji, N., Keys, D.N., and Levine, M., 2002, Genome-wide identification oftissue- specific enhancers in the Ciona tadpole. Proc. Natl. Acad. Sci. 99, 6802-6805.
Hino K., Satou, Y., Yagi, K., Satoh,N., 2003, A genomewide survey of developmentally relevant genes in Ciona intestinalis. VI. Genes for Wnt, TGFP, Hedgehog and JAK/STAT signaling pathways. Dev. Genes. Evol.213, 264-272 Ikuta, T., Saiga, H., 2007, Dynamic change in the expression of developmental genes in the ascidian central nervous system: revisit to the tripartite model and the origin of the midbrain-hindbrain boundary region. Dev. Biol. 3 12 , 63 1-643.
Ik-ta, T., Yoshida, N., Satoh, N., Saiga, H., 2004, Ciona intestinalis Hox gene cluster:
its dispersed structure and residual colinear expression in development. Proc. Natl Acad.
Sci. U.S.A. 1O1 15118-15123.
'
lnman, G.J., Nicoias, F.J., Callahan, J.F., Harling, J.D., Gaster, L.M., Reith, A.D., Laping, N.J., Hiil, C.S., 2002, SB--431542 is a potent and specific inhibitor of
transforming growth factor-beta superfamily type I activin receptor-like kinase (ALK) receptors ALK4, ALK5 , and AL K7 . Mol. Pharmacol. 62, 65-74.
Levin, M., 2005, Left-right asymmetry in embryonic development: a comprehensive review. Mech. Dev. 122, 3-25.
Massague, J. and Wotton, D., 2000, Transcriptional control by the TGF-/Smad signaling system. EMBO J. 19, 1745-1754.
Massague, J. and Wotton, D., 2005, Smad transcription factors. Genes Dev. 19, 2783-281O.
Mathers, P.H., Grinberg, A., Mahon, K.A., Jamrich, M., 1997, The Rx homeobox gene is essential for vertebrate eye development. Nature 387, 603-607.
Mita, K., Koyanagi, R., Azumi, K., Sabau, S.V., Fujiwara, S., 201O, Identification of Genes Downstream ofNodal in the Ciona intestinalis Embryo. Zoological Science, 27(2), 69-75.
Morokuma, J., Ueno, M., Kawanishi, H., Saiga, H., Nishida, H., 2002, HrNodal, the ascidian nodal-related gene, is expressed in the left side ofthe epidermis, and lies upstream ofHrPitx. Dev. Genes Evol. 212, 439446.
Putnam, N.H., Butts, T., Ferrier, D.E., Furlong, R.E, Hellsten, U., Kawashima, T., Robinson- Rechavi, M., Shoguchi, E., Terry, A., Yu, J.K., 2008, The amphioxus genome and the evolution ofthe chordate karyotype. Nature 453, 1064-1071.
Shi, Y., VVang, Y.F., Jayaraman, L., Yang, H., rViassague, J., and Pavletich, N.P., 1998, Crystal structure ofa Smad MHI domain bound to DNA: Insights on DNA bindinginTGF6-signaling. Cell94,585-594.
Shiratori, H., Sakuma, R., VVatanabe, M., Hashiguchi, H., Mochida, K., Sakai, Y., Nishino, J., Saijoh, Y., Whitman, M., Hamada, H., 2001, Two-step regulation of left- right asymmetric expression ofPitx2: initiation by nodal signaling and maintenance by Nkx2. Mol. Cell 7, 137-149.
M. Tada, M.-A. J. O'Reilly and J. C. Smith, 1997, Analysis ofcompetence and of Brachyury autoinduction by use ofhormoneinducible Xbra. Development 124, 2225-234.
Yoshida, K., Saiga, H., 2008, Left-right asymmetric expression ofPitx is regulated by the asymmetric Nodai signaling through an intronic enhancer in Ciona intestinalis.
Dev.Genes Evol. 218, 353-360.
Yoshida, K., Saiga, H., 2011, Repressoin ofRx gene on the left side ofthe sensory vesicle by Nodal signaling is crucial for light-sided formation ofthe ocellus
photoreceptor in the development ofCiona intestinalis. Dev. Biol. 354, 144-150.
Yu, J.K., Holland, L.Z., Holland, N.D., 2002, An amphioxus nodal gene
(AmphiNodal) with early symmetrical expression in the organizer and mesoderm and later asymmetrical expression associated with left-right axis formation. Evol Dev 4,
418--425.
Zawel, L., Dai, J.L., Buckhaults, P., Zhou, S., Kinzler, K.W., Vogelstein, B., and Kem, S.E., 1998, Human Smad3 and Smad4 are sequence-specific transcription activators. Mol.Cell 1 611---617.
'
Zernicka-Goetz, M., Pines, J., Ryan, K., Siemering, K.R., Haseloff, J., Evans, M.J., Gurdon, J.B., 1996, An indelible lineage marker for Xenopus using a mutated green fluorescent protein. Development 122, 3719-3724
Fig. I Expression patterns of Ci-Rx and Ci-Nodal in the dechorionated and control normal embryos.
(A-D) Rx and Nodal expression at the early (A and C) and mid (B and D) tailbud stage embryos that have developed in the chorion. (E-J) Examples of Rx and Nodal
expression pattern in the dechorionated embryo at the mid tailbud stage. Dorsal views are shown with anterior to the left. etb, early tailbud stage; mtb, mid tailbud stage.
Av---t
o
Åí
E
Rx Noda/
A
ixtz,,i,siiill
B t t t tt /t. .
- " nf'f " '
"i''
/,evptuwo't/' +/.t ''''
i' ''
'.r
'' ]" . .V. 'b-
', a tfftt ' .e
D .'"" hb,ts '' ,,.
t tt;
'
",.deff.esx" x - .,sk l :" ,$
-ee s
' "W
t tt ee -pt
i-ts,,
Rx
E
Noda/
v"li3Åë
=
qc(
MN.-m,-o
=N
ca
vo
co"o
>Åë
a:
•* ,.lli '
•..,1/,,•, 1.l.ig.•. .
,•-
i•
G .vt l tt' /rvastss-' t
ttigttt.
t-t ift/t
e.
t tditt
ttl t
V{.feil.iilii,. ,X-•:':lp,si
"..
s"tr
[tw.ewA,es, ,i. ,.i•,.
t ,/,-,,Ftst-'",, t ,
•h•• •••"Q.k••• •' "pt
H ., ' .,,, I'is'thkraj,,• k'-.,,..r.,,,.1/X'geig-tw,1."'
.v''1'ge, ,i' '' , '' , , 'as•.t. "'"' '".. ..,• '{'
" ,fi)/
'
g",'.vak,,.Sij!lpt. ii.ew,"' ss.wwi,x .xtu, ,g.ss ,.i"e.•f..,x.. h,. "•t ,.
"te,.x •ss"g$istK•ts ge asligee, li,si'.g"., '
"t-
'x.'"-'" • ..
"tt
.-
'
J
x pt t, v,tY
.vi tt";1•
.""ttty;t,L
L•
es,as'.x.tt:'sitystrt'.}rw..,•"
tk,". ilil/ili
Fig. 2 Designing the method of reporter assay to identify negative transcriptional regulatory region of Ci-Rx in response to the Nodal signaling.
(A) Schematic drawing of the method for the reporter assay. Fertilized eggs were introduced with the three plasmid DNAs, culture up to the neurula stage (7 hrs after fertilization) and the resulting embryos were divided into two groups. Embryos of one
group were continuously cultured (OE embryos) and those of the other group were cultured in the presence of 5pM SB431542 (SI embryos). (B and C) Whole mount in situ hybridization for Ci-Rx transcripts ofOE and SI embryos, respectively.
A Fertilizedeggs
z,2,1;l"rv-'Mma:-"-":>,>x
11Ke,,'iN.... ././t}x
%k-...:..rl!Tl1/1 :rv/
ldechorionation
electroporation, cultured aM 80CfzaxX
lx f"ig'%a,.tc. /
k""'"-.e..ycbM•
7hpf
Nodal over-expressed (OE)
embryo
Plasmid DNAs
e Rx upstream-GFP
• Cralbp-Nodal e Pitx Dl-GFP
SB431542
treatment
Nodal signaling- inhibited (Sl) embryo
qX
6
OE Sl
c
f
' .ga. ts'
,,x'
//f e'"' ., •s,
k, "
sdeng.4tZhe -
esT wwhn
st:s"•.f..."tbetwpaig
oig 'f
,.•Ak
'k.,r.
9t+ Xva
x,.,..fx•"[il '';'%
"""deK2
sly+
'tij''t pt 'tlt
s e wa.t'es
.WSGij1\$6/1ut6'rk
Fig. 3 Reporter assay to identify negative transcriptional regulatory region of Ci- Rx.
(A) Schematic drawing of Rx upstream-GFP reporter constructs. Solid lines indicate Ci-
Rx upstream region, blue and red boxes indicate 29bp untranslated region (UTR) and 13bp coding region, respectively.
(B) Whole mount in situ hybridization for GFP transcripts of the embryos introduced with reporter and internal control constructs. The expression due to the reporter construct and the internal control construct is indicated by red and white arrowheads,
respectively. Dorsal views are shown with anterior to left. Expression levels of reporter
GFP were classified into three; embryos below the red, orange and yellow boxes indicate specimens with the reporter GFP expression level stronger than, equal to and weaker than GFP expression level of the internal control, respectively.
(C-E) Graphs showing the difference between OE and SI embryos in the ratio of the three expression levels (as defined in B) exhibited by a given reporter construct. Color
codes are the same in (B). Reporter constructs assayed are indicated at the top of each
graph. The numbers ofembryos examined are indicated to the right.
A
-390o
us 3k-GFP:
-2000
us 2k-GFP:
-IOOO -700 ATG
,+.1 +l5
iill•l•lll•llli l GFP
li• GFP
ORF
-- - :: :
-- .-I
uslk-GFP: : '
::
te
i •1
-t t- l- -e e- ::
::
--
;
usO.7k-GFP-,
• -.l
l, t-
l -473 i
tt -- ts
I Smad i
- •t
us O.7k-muSmad-GFP-
••l
GFP GFP GFP
i.c.<Rxtr ieCs=RX
<?
.sbeWpt"ptSltutw'wwli"F.
•tsi.su••Y"
tt tttttt tsi
..l.l/'''' te''}'f
I•
-- l''IEi,••.
,/t
tt'fl""'"
//•thpm••ge- tlt...ttttt
'' '' t'
.tt tt
..".FSlki'' -de
..,
,..
af,•S'4/• .'
tt
,,,., .,y',1,
tttt
nny•tanv
!ills
'i'x''.--•wt•-. ""'.,,' ,s,•}"
ttttttt/.Jwipt"'t'tt"tttttt
A Rxus3k-GFP
ieCs> RX
"'p 1tv ]f' '"' " :t'` " :" ys'nt
.f vvas" " '10- ,}
. •e
, •<i•IA/riiii'i:iii'.'lll""' •i'li••1"i''"; i"'11iil••i',i'li"' ''••
'' .. x
A internaicontroi
c
D
l i'
OE
i'
-l•
l
Sl
yeq
i'
l i
l---•----• t -
O 25
us3k-GFP
lii
l
i l
l l l
l
l
E
OE
Sl
Il i•
OE
l•
il l•
Sl
l }
-- e --•
50 75
usO.7k-GFP
ii
i
l
l
i
t+"tttt bt"t-tt""lit--"t
s l
i
l
I
g..---.--.-...
o
hfi tt- -t-"-t-tt-i-t - ' "-t-t -ny"-tt-lt-ny-Fr-ni/ npn-nrr--t---tnyn+pny"n-"-Tnn"rr-jtnn-rrt-nn
25 50 75
usO.7k-muSmad-GFP
li il
' '
111 lil :1
t k.,
i' •'• .--.-- -. .k, im -- --lle--,V- -HinH.
50 75
I i
l'n=46 i'
l
I•
li 11/• n=48
i i
...--..-....- s
1ooO/o
ii il
ln=59
l i i
ll
i li' n=54
l'
i
--4--i/
loo O/o
o 25
i' l'
n=47
i'
i n=19
1ooO/o
- i.c.<Rx i.c.=Rx i.c.>Rx
Fig. 4 Model for the right-sided Ci-Rx expression and formation of the ocellus by the left-sided Nodal signaling.
Transcription of Ci-Rx, the master gene of the ocellus formation, is negatively regulated
by Nodal via Smad and factor X TFBSs and positively regulated by Onecut and Clox, leading to the ocellus formation in the right dorsal trunk. If the Nodal signaliRg is
inhibited, Ci-Rx transcription would take place also on the left side, leading to the extra- ocellus formation on the left side (Yoshida and Saiga, 2011). Thus, the left-sided Nodal
signaling is involved in the right-sided ocellus formation and pigment cell movement to
the right side by repressing Ci-Rx transcription on the left side.
L
NodaI
FS
apatgdrx' X..
[:iiill
ssOnecut clpx
#: ix
"i g l :
R
:
: OceI1us
Supplemental Fig.1 Whole mount in situ hybridization for GFP transcripts of the mid tailbud embryos introduced with us3k-GFP, SI embryo. Dashed lines indicate the midline. Dorsal views are shown with anterior to Ieft. Red arrowhead indicates us 3k expression, and white arrowhead indicate i.c. expression.
igX Rxus3k-GFp A internai controi
qK e
us3k--GFP + Crabp-Nodal + SB431 542
pmf '
''1/ "
,/..I.gf...•. . Tll•• ' f'::';1 ,/...pt/• ••..1
f. .t.VMI
/t•tlt tza,t
tttr
'
.1
.,i., •s
il•,),.i,..,l,ii.ili'i 1•,• •
Midline
. fijlpt{
.{,av '' ,' 'ty 'f ., !1 ' ais;1'
,. •i,i 1'11I/1.. ..11;'. .i' ;' //+'i ' ""f' 'tttt'' """' ''
ts'' •' "lr.t .,.s,.,...,.fi //i'i"'r/p,'". " aAt` ,.jtt.2 .•
I'f''"1;'/k2' ' `11'w"/'Ii')'ttt'
ttt
2O/22
sS} .x.t
Bi1ateral
•'g .pt, "'"
'th '1 a
sc
.be,.
X, .3
r•
/g,
. ,w:.
;t ... -
.'lv.""
2/22
