Gene regulatory networks underlying the compartmentalization of the Ciona central nervous system

285RESEARCH ARTICLE

INTRODUCTIONThe expression profiles of neural genes in both protostomes anddeuterostomes suggest that the tripartite structure of the CNS mightbe evolutionarily ancient (Wada et al., 1998; Lowe et al., 2003;Reichert, 2005). The vertebrate CNS has elaborated on this basictripartite structure to produce complexity and novelty. The anteriorneural tube develops into the brain, whereas posterior regions formthe spinal cord. The brain becomes regionalized into forebrain,midbrain and hindbrain, which are further subdivided at later stages.The MHB organizer is one of the mechanisms used for theelaboration of the complex vertebrate brain (Liu and Joyner, 2001;Wurst and Bally-Cuif, 2001; Rhinn et al., 2006). A number oftranscription factors and signaling molecules are expressed in theMHB region, and these molecules constitute a complex geneticnetwork that establishes and maintains features of the organizer.Fgf8 and Wnt1 are the key factors in this network controlling thedevelopment of the midbrain and hindbrain.

The ascidian tadpole represents the closest living relative of thevertebrates and possesses a simplified CNS derived from the dorsalneural tube. The ascidian CNS consists of a centralized sensoryvesicle (SV), visceral ganglion (VG) and nerve cord composed of~260 cells (Meinertzhagen et al., 2004). There is a morphologicallydistinguishable domain called the ‘neck’ between the SV and VG(Nicol and Meinertzhagen, 1988). Although the SV corresponds tothe forebrain, it is unclear whether the neck and VG correspond tothe hindbrain and/or spinal ganglia of vertebrates (Wada et al., 1998;Dufour et al., 2006). It is conceivable that a distinct midbraincounterpart is absent in the ascidian larva (Takahashi and Holland,2004). Therefore, the MHB organizer has been regarded as a novel

property of the vertebrate CNS. The vertebrate MHB organizersecretes Wnt1 and Fgf8, which are important for the MHBorganizing activity. Although the Ciona genome does not containWnt1 (Hino et al., 2003), an ortholog of Fgf8 exhibits localizedexpression during the development of the Ciona CNS at the lategastrula stage (Imai et al., 2004; Hudson and Yasuo, 2005; Imai etal., 2006) and at the middle tailbud stage (Imai et al., 2002), therebyraising the possibility that an MHB organizer-like structure operatesin ascidians.

To determine the extent to which the compartmentalization of theascidian and vertebrate CNS are controlled by conserved and non-conserved gene circuits, we have examined the expression andfunction of a comprehensive set of neural regulatory genes. Thesestudies permitted the reconstitution of a provisional gene regulatorynetwork for the development of the ascidian CNS based onsystematic gene disruption assays. The resultant network can explainthe causalities of the gene expression profiles seen in the CNS andprovide insights into the evolutionary origin of the vertebrate CNS.

MATERIALS AND METHODSAscidian embryosC. intestinalis adults were obtained from the Maizuru Fisheries ResearchStation of Kyoto University (Kyoto, Japan) and from Half Moon Bay harbor(CA, USA). We used late gastrula embryos just after the neural plateformation and late gastrula embryos just prior to the next division in thepresent study, and early and mid-tailbud embryos, which correspond to E55and E75-E80 (Cole and Meinertzhagen, 2004).

Expression profiles of regulatory genesMost cDNA clones were obtained from our EST collection (Satou et al.,2002). The detailed procedure for whole-mount in situ hybridization hasbeen described (Satou and Satoh, 1997). Cell identities were determined byDAPI staining of nuclei.

Gene knockdownsMO oligonucleotides were purchased from Gene Tools: BMP2/4,AAGTCCAATCCGTAAGCGCCACCAT; Cdx, TTGTGCGTTTCTCAT-CAATGGTTGC; Delta-like, GAAGTAATATAAGCTTGATGCTCAT;

Gene regulatory networks underlying thecompartmentalization of the Ciona central nervous systemKaoru S. Imai1,2, Alberto Stolfi2, Michael Levine2 and Yutaka Satou1,*

The tripartite organization of the central nervous system (CNS) may be an ancient character of the bilaterians. However, theelaboration of the more complex vertebrate brain depends on the midbrain-hindbrain boundary (MHB) organizer, which is absentin invertebrates such as Drosophila. The Fgf8 signaling molecule expressed in the MHB organizer plays a key role in delineatingseparate mesencephalon and metencephalon compartments in the vertebrate CNS. Here, we present evidence that an Fgf8ortholog establishes sequential patterns of regulatory gene expression in the developing posterior sensory vesicle, and theinterleaved ‘neck’ region located between the sensory vesicle and visceral ganglion of the simple chordate Ciona intestinalis. Thedetailed characterization of gene networks in the developing CNS led to new insights into the mechanisms by which Fgf8/17/18patterns the chordate brain. The precise positioning of this Fgf signaling activity depends on an unusual AND/OR network motifthat regulates Snail, which encodes a threshold repressor of Fgf8 expression. Nodal is sufficient to activate low levels of the Snailrepressor within the neural plate, while the combination of Nodal and Neurogenin produces high levels of Snail in neighboringdomains of the CNS. The loss of Fgf8 patterning activity results in the transformation of hindbrain structures into an expandedmesencephalon in both ascidians and vertebrates, suggesting that the primitive MHB-like activity predates the vertebrate CNS.

KEY WORDS: Ciona intestinalis, Gene regulatory network, Fgf8

Development 136, 285-293 (2009) doi:10.1242/dev.026419

1Department of Zoology, Graduate School of Science, Kyoto University, Sakyo-ku,Kyoto, 606-8502, Japan. 2Department of Molecular and Cellular Biology, Division ofGenetics and Development, University of California, Berkeley, CA 94720, USA.

*Author for correspondence (e-mail: [email protected])

Accepted 7 November 2008 DEVELO

PMENT

286

Emc, CAACTTTAACCATTTTGCTGATTCT; en, TGGGCAACCTTG-TATTTCGCTTCAT; ephrinA-b, ACAAAGGCATGGTGATATACG-CATT; Fgf8/17/18, TACTCGCAATGCATTAAATCCGAAT; FoxB,AGTCTCGTCCTGGTCGTGGCATTTT; Gli, CGCGTTCTCCATG-TAAAATCTACGA; Hedgehog2, ACTGTCCCGCTTATACGT-TACTCGC; Hnf6, CAGACTGACCGAGCGAAACTGGCAT; Irx-C,TAAGACCGGGCAGCTCCGACTGCAT; Lhx3, AGAATTAACTGTAA-CAAATTGATCG; Lmx, TTTCGTCGTTAGAAGAACGCAGCAT; Mnx,CTGACTTTGAAGTACTTAGCATCAT; Msxb, ATTCGTTTACTGT-CATTTTTAATTT; Neurogenin, AAATCCAACATTTTGTAGCAA-GAGC; Nk6, GAACCAGATTCTTCCATGGACATCA; Otx, CATGT-TAGGAATTGAACCCGTGGTA; Pax2/5/8-A, CAGTTCATATTC -AAACTTACTAACA; Pax3/7, AATTAGACCCTGGATGCATCATGTT;Pax6, GCCTACAAGAATCGTACGTCGCCAT; Snail, GTCATGATG-TAATCACAGTAATATA; SoxB1, AACATGAAGTCGTTCTGAGATG-GCT; Tbx2/3, GAGGTCCACACCAACACTTTAACAT; and Raldh2,GTACTGCTGATACGACTGAAGACAT. Microinjections wereperformed as described (Imai et al., 2000). Blast searches of the MOsequences could find no possible mis-targeting sites in the genomesequence (Dehal et al., 2002), and we found no relations that cannot beexplained from the expression profiles. Different effective MOs produceddifferent phenotypes (see Fig. S1 in the supplementary material), and wecould not find any effects in embryos injected with an MO against lacZ(data not shown), indicating that there were no generalized disruptions. Forfurther confirmation of specificities of the MOs, we injected second MOsagainst Fgf8/17/18 (CATTTTCGTATGTAATCCAAGAGAA) and Snail(TATTTCACAGTGAGAATTTTAATAT), and MOs against Otx (AGT-

GTGGAGATTTCAAGTATGACAT) and Neurogenin (ATCGGTTTGCA-GAATAATCCAACAT) of Ciona savignyi, a closely related species, intoCiona savignyi eggs. We carried out in situ hybridization of Otx andPax2/5/8-A for Fgf8/17/18 morphant embryos, of Fgf8/17/18 for Snailmorphant embryos, of Snail for Neurogenin morphant C. savignyi embryosand of Cyp26, FoxB and Hox1 for Otx morphant C. savignyi embryos (seeFig. S1 in the supplementary material). They recapitulated the originalphenotypes. All of these observations support specificities of the MOs usedin the present study. Whole-mount in situ hybridization was used fordetermining genes expressed in the downstream of genes that wereknocked down.

Electroporation4.8 kb and 4.2 kb of flanking 5� sequences of Fgf8/17/18 and FoxBrespectively were PCR-amplified from genomic DNA. AchTP promoter wasobtained as described previously (Yoshida et al., 2004). All were cloned intopCESA vector upstream of unc-76-tagged GFP or mCherry reporter gene(Dynes and Ngai, 1998). The detailed procedure for electroporation has beendescribed previously (Corbo et al., 1997).

RESULTSThe ascidian brain has distinct domains withdifferent propertiesAccording to previous descriptions (Nicol and Meinertzhagen,1988), the posterior region of the sensory vesicle (PSV) and VG arederived from the posterior neural plate (Fig. 1A). The short region

RESEARCH ARTICLE Development 136 (2)

Fig. 1. Development of the A- and b-line central nervous system in the Ciona embryo. (A)Schematic representations of A- and b-line neuronalcell lineages of the bilaterally symmetrical embryos at the late gastrula stage. Cell names are shown in the lower panel. (B,C)A schematic representationof the CNS cells at the tailbud stage from the (B) dorsal and (C) lateral views, and their lineages. Arrows indicate cell lineages and the cells with the samecolors are derived from the single cells at the late gastrula stage (enclosed by pink lines). Cells enclosed by thick black lines are post-mitotic cells destinedto become motoneurons. At the early tailbud stage, three pairs of presumptive motoneurons are post-mitotic. Until the mid-tailbud stage, two pairs ofpost-mitotic presumptive motoneurons are differentiated from the remaining two pairs of the visceral ganglion (VG) cells after cell divisions, as shown inthe upper part of B. (D)The central nervous system of a tailbud embryo developed from an egg electroporated with Fgf8/17/18>RFP/AchTP>GFP. (E)Alate tailbud embryo developed from an egg electroporated with Fgf8/17/18>RFP and AchTP>GFP. Arrowheads indicate motoneurons. (F)The centralnervous system of a tailbud embryo developed from an egg electroporated with Fgf8/17/18>RFP / FoxB>GFP. RFP marks the A9.30-descendants and GFPmarks neurons. LNC, the lateral rows of the nerve cord; DNC, the dorsal row of the nerve cord; VNC, the ventral row of the nerve cord.

DEVELO

PMENT

between the PSV and VG is called the ‘neck’. Both the PSV andneck are derived from a single pair of neuronal progenitor cells ingastrulating embryos, A9.16, whereas the VG is derived from twoneighboring blastomeres, A9.30 and A9.29 (Fig. 1B). The latter cellalso contributes to the caudal nerve cord, which consists of non-neuronal ependymal cells. The ventral and dorsal rows of the neuraltube are derived from A9.13/A9.14/A9.15 and b9.37/b9.38,respectively (Fig. 1C).

It is thought that the PSV and VG produce the core synapticactivity for the swimming behavior of the tadpole, because the VGincludes five pairs of motoneurons that innervate the tail musclesand are modulated by cholinergic neurons located in the PSV (Coleand Meinertzhagen, 2004). To confirm this idea, we traced the A9.30lineage with an RFP fusion gene containing the Fgf8/17/18 enhancer(Fig. 1D-F) throughout development, as Fgf8/17/18 expression isrestricted to A9.30 at the late gastrula stage (Imai et al., 2004). Co-electroporation of this fusion gene with a GFP fusion genecontaining the enhancer of the acetylcholine transporter gene(AchTP) identifies two pairs of early post-mitotic cholinergicmotoneurons derived from A9.30 (yellow cells, Fig. 1D). Theanterior A9.30 lineage forms non-neuronal ependymal cells (redcells, Fig. 1E), as well as two additional pairs of motoneurons in the

VG of older embryos (anterior two yellow cells, Fig. 1E). Theposterior-most pair of motoneurons (A10.57) arises from the A9.29lineage (posterior green cells, Fig. 1D,E). Embryos electroporatedwith a GFP fusion gene containing the FoxB enhancer permittedvisualization of the anterior lineage arising from A9.16, which formsthe PSV. As seen for endogenous FoxB transcripts, GFP is stronglyexpressed in the anterior A9.16 lineage that forms the PSV, but notin the posterior A9.16 lineage that forms the neck region (Fig. 1F).The neck consists of quiescent undifferentiated precursors that formthe neurons homologous to cranial motoneurons of vertebrates aftermetamorphosis (Dufour et al., 2006). These molecular dataconfirmed the idea that the PSV, neck and VG have distinctiveproperties (Cole and Meinertzhagen, 2004).

The regulatory states of cells in the developingCNSAn earlier study identified a comprehensive list of regulatory genesexpressed in the developing nervous system (Imai et al., 2004). Therecent advances in imaging technology and the knowledge of thecell lineages of the CNS (Cole and Meinertzhagen, 2004) allowedus to elaborate on this description at single cell resolution by in situhybridization (summarized in Fig. 2A-C; see also in situ

287RESEARCH ARTICLECompartmentalization of the Ciona CNS

B

D

ArixCOEEmc

enets/pointed2

FoxA-aFoxB

FoxD-a/bGli

HNF6Hox1Hox5Irx-Cislet

Lhx3Lmx

macho-1Mnx

MYTFNeurogenin

NK6Otx

Pax2/5/8-APax3/7

Pax6SoxB1Tbx2/3Unc4-A

Delta-likeEphrinA-d

FGF9/16/20Hedgehog2

Wnt7

A11.

64A1

1.63

A11.

62A1

1.61

A11.

120

A11.

119

A11.

116

A11.

115

A11.

126-

8A1

1.11

8A1

1.11

7A1

0.57

b10.

75b1

0.76

b10.

74b1

0.73

A10.

28A1

0.27

A10.

26A1

0.30

A10.

25A1

0.29

PSV neckanterior VG

LNCposterior VG

DNC VNC

A9.1

3A9

.14

A9.1

5A9

.16

A9.2

9A9

.30

A9.3

2b9

.37

b9.3

8

CdxCOE

E(spl)/hairy-bEmc

ets/pointed2FoxA-a

FoxBLmxMnx

msxbMYTF

NeurogeninPax3/7

Pax6Snail

SoxB1ZF (C2H2)-33

ZicL

BMP2/4Delta-like

EphrinA-bEphrinA-d

FGF8/17/18FGF9/16/20

A12.

240

A12.

238

A13.

474

A13.

473

A12.

239

A11.

118

A11.

117

A10.

57

Delta-likeEphrinA-bEphrinA-d

FGF8/17/18

Chox10COEEmc

enFoxBHNF6Hox1Hox3Irx-Cislet

Lhx1Lhx3Mnx

MYTFNeurogenin

NK6Pax6

POU4SoxB1

A

C

Fig. 2. Expression profiles of the regulatorygenes. (A) A- and b-line neural cells at the lategastrula stage, (B) posterior sensory vesicle (PSV),neck, VG and nerve cord at the early tailbud stage,and (C) VG at the middle tailbud stage. Eachexpression is shown by a red rectangle. (D) Ahierarchical clustering of the neural cells based onthe expression profiles of transcription factor genesshown in B.

DEVELO

PMENT

288

hybridizations of control embryos in Fig. S1 in the supplementarymaterial). The resultant diagrams of the expression profiles ofindividual cells define the regulatory states of the cells from the lategastrula to the mid-tailbud stage.

The hierarchical clustering of the expression profiles confirmedand refined the morphological differences of the brain regions froma molecular viewpoint (Fig. 2D). Two pairs of the PSV cells (A11.63and A11.64 in the early tailbud embryo) are in the same regulatorystate, based on selective expression of Otx, FoxB and En. The neckcells (A11.61 and A11.62) selectively express Pax2/5/8-A, Hox1 andGli. The VG and caudal nerve cord cells express differentcombinations of regulatory genes. Anterior and posteriormotoneurons in the VG express Hox1/En/Neurogenin andLhx3/Hnf6/Neurogenin, respectively. Posterior portions of the neuraltube express Cdx, Snail and Pax6. The dorsal and ventral rows of theneural tube also express different sets of regulatory genes.

Provisional gene regulatory networks in thedeveloping CNSOn the basis of the expression profile data, we systematicallyperturbed the functions of regulatory genes expressed in the CNSwith specific morpholino oligonucleotides (MOs) in order to

determine the molecular basis for the compartmentalization of theCiona brain. We succeeded in disrupting the activities of a total of25 regulatory genes (Tables 1 and 2). In situ hybridization assayswere used to monitor the effects of the different MO-inducedmutants on the expression of all of the regulatory genes expressed atthe stages examined (see Fig. S1 and Table S1 in the supplementarymaterial). This information, along with earlier results (Imai et al.,2002; Lemaire et al., 2002; Hudson and Yasuo, 2005; Moret et al.,2005; Imai et al., 2006; Hudson et al., 2007; Ikuta and Saiga, 2007),was used to create a provisional circuit diagram showing theinterconnections among the regulatory genes controlling theregionalization of the CNS (Fig. 3A). The simplicity of the systempermits the elucidation of gene networks at single cell resolution(Fig. 3B-E; see Figs S2-S4 in the supplementary material).

For example, Fgf9/16/20 expression in the PSV and neckprogenitors is controlled by Emc and Pax3/7 (Fig. 3B). In the neckregion, Pax2/5/8-A also contributes to the expression of Fgf9/16/20(Fig. 3C). Neurogenin was found to be a crucial determinant of themotoneurons in the VG (Fig. 3D). Cdx is required for the developmentof ependymal cells in posterior regions of the neural tube (Fig. 2E),whereas Lmx contributes to specification of the dorsal nerve cord cellsby activating Pax3/7 (see Fig. S3 in the supplementary material).


A11.64

A11.62

PSV neck VG LNC

Delta-like

en FGF8/17/18

FoxB

MYTF Otx

E(spl)/hairy-b Emc

Pax6 SoxB1

Arix

FGF9/16/20

Gli Hox1

Pax2/5/8-A Pax3/7

A11.118

Hox1

E(spl)/hairy-b

FGF8/17/18

Pax6

Snail SoxB1

COE Delta-like

Emc

EphrinA-d

HNF6

Irx-C Lhx3

Mnx Neurogenin

NK6

A11.116

COE

Delta-like

en

FoxA-a

Lhx3 Mnx

Neurogenin

Cdx

E(spl)/hairy-b

EphrinA-b

EphrinA-d

MYTF

Snail

Hox1 Hox5

Pax6

SoxB1

B

A

C D E

Arix BMP2/4 Cdx Chox10 COE

Delta-like E(spl)/hairy-b Emc en EphrinA-b

EphrinA-d ets/pointed2 FGF8/17/18 FGF9/16/20 FoxA-a

FoxB FoxD-a/b Gli Hedgehog2 HNF6

Hox1 Hox3 Hox5 Irx-C islet

Lhx1 Lhx3 Lmx macho-1 Mnx

msxb MYTF Neurogenin NK6 Otx

Pax2/5/8-A Pax3/7 Pax6 POU4 Snail

SoxB1 Tbx2/3 Unc4-A Wnt7

Delta-like

Hox1 MYTF

E(spl)/hairy-b

Emc

Pax6 SoxB1

en

FGF9/16/20 FoxB

Otx Pax3/7

Fig. 3. Gene regulatory networksin the ascidian CNS. (A) Summary ofthese networks. (B-E) Diagrams ofgene circuits in (B) A11.64 (PSV), (C) A11.62 (neck), (D) A11.118 (VG)and (E) A11.116 (caudal nerve cord) atthe early tailbud stage. The color codefor cells is the same as in Fig. 1.Transcription factor genes andsignaling ligand genes are indicated byrectangles and ovals, respectively.Genes expressed in the ancestors ofthe cell but not expressed at this stageare enclosed by broken lines. Genesthat are not expressed in either thespecified cell or its ancestors areshown in gray. Arrows indicatetranscriptionally regulatoryinteractions. The lines ending in a barindicate repression. See the text fordetails.

DEVELO

PMENT

Localized Fgf8/17/18 delineates PSV and neckregions of the Ciona CNSThe reconstituted networks reveal a central role of Fgf8/17/18 ingenerating regional patterns of gene expression. Fgf8/17/18 isexpressed at the late gastrula stage in A9.30 (see Fig. 5C), whichforms most of the VG (A11.117 through A11.120) at later stages.This Fgf signal acts on the neighboring A9.16 cell to define the neckregion of the definitive tadpole CNS, which forms between the PSVand VG. Otx and FoxB expression is normally restricted to theanterior-most regions of the CNS, including the presumptive PSV(Fig. 4A; see Fig. S1Q in the supplementary material; see also Fig.2B). There is a posterior expansion of both expression patterns inmorphant embryos injected with an Fgf8/17/18 MO (Fig. 4B; seeFig. S1Q in the supplementary material). Thus, Fgf8/17/18 inhibitsOtx and FoxB expression in the neck, thereby restricting theiractivities to the PSV. En normally displays periodic expression inthe PSV and VG (Fig. 4C), but expression extends into the neck ofmorphant embryos (Fig. 4D). There is also a loss of Pax2/5/8-Aexpression (Fig. 4F), which is normally restricted to a tight stripe ofexpressing cells in the neck (Fig. 4E). Gli, Fgf9/16/20 and Arix,which are normally expressed in the neck, are lost in morphantembryos, because these genes are under the control of Pax2/5/8-A(see Fig. S1A,O,S in the supplementary material). Finally, Hox1expression is normally restricted to the neck and VG (Fig. 4G), butexpression is lost in the neck of morphant embryos (Fig. 4H).Altogether, the network analysis suggests that the neck is not formedin Fgf8/17/18 morphants, but is transformed into an expanded PSV(Fig. 4I,J).

Fgf8/17/18 is first expressed at the 64-cell stage in A7.6, whichabuts A7.8 (a progenitor of A9.29-A9.32, the VG and caudal nervecord lineage cells), but this expression disappears before the lategastrula stage. Although this early expression is also suppressed in

the Fgf8/17/18 morphant, this early Fgf8/17/18 expression is notlikely to be required for the CNS regionalization, because Pax2/5/8-A is not expressed in embryos treated with a MEK inhibitor, U0126,from the late gastrula (Fig. 5). The same Fgf gene is again expressedin the VG at the middle tailbud stage (Imai et al., 2002), but thisexpression is later than Pax2/5/8-A expression. Therefore, theexpression in A9.30 at the late gastrula stage is most likely to beresponsible for the regionalization of the CNS.

This phenotype of Fgf8/17/18 morphants is evocative of thetransformation of the metencephalon into an expandedmesencephalon seen in ace (Fgf8a) mutants in the zebrafish CNSwhere there is expanded expression of Otx2 (mesencephalon) anda loss of Pax8 (metencephalon) (Jaszai et al., 2003). Similartransformations are also seen in conditional knockout mice ofFgf8 (Chi et al., 2003). Thus, in the vertebrate CNS, Fgf8 isexpressed in the MHB region and required for properspecification of the midbrain and anterior hindbrain. We mighttherefore be able to regard the PSV and neck as counterparts ofthe vertebrate midbrain and anterior hindbrain, although this hasbeen a debatable issue (Wada et al., 1998; Takahashi and Holland,2004; Canestro et al., 2005; Dufour et al., 2006). Regardless oftheir exact evolutionary counterparts and the timing of thesignaling, we propose that the recruitment of Fgf8 signaling mighthave been a crucial event for the compartmentalization of thechordate brain. It is less likely that the same signaling system wasindependently acquired for similar uses by tunicates andvertebrates.

Despite the apparent similarities in the patterning of the ascidianPSV and neck with the compartmentalization of the vertebratemidbrain and anterior hindbrain, we note a number of differences inthese processes. First, Ciona Fgf8/17/18 acts much earlier – duringlate gastrulation – than it does in vertebrate embryos. Fgf8 might act


Table 1. Summary of knockdown experiments with morpholino oligonucleotides for genes expressed at the late gastrula stageNumber of genes

A Genes analyzed in detail

Genes whose downstream genes were foundCdx, Bmp2/4, Delta-like, Emc, Fgf8/17/18, FoxB, Lmx, Mnx, Msxb, Neurogenin, Pax3/7, Pax6, Snail, SoxB1 14

B Genes whose functions in the later development were hardly examined, because of their expression in the early embryo

ets/pointed2, EphrinA-d, FGF9/16/20, FoxA-a, ZicL 5

C Genes for which no good morpholino oligonucleotides were obtained

COE, E(spl)/hairy-b, Mytf, ZF-C2H2-33 4

Table 2. Summary of knockdown experiments with morpholino oligonucleotides for genes expressed in the brain, boundary andvisceral ganglion at the early tailbud stage

Number of genes

A Genes analyzed in detail

Genes whose downstream genes were foundEmc, FoxB, Delta-like, EphrinA-b, Gli, Hnf6, Lmx, Mnx, Neurogenin, Otx, Pax2/5/8-A, Pax3/7, Pax6, SoxB1 14

Genes whose downstream genes were not foundEn, Hedgehog2, Irx-C, Lhx3, Nk6, Tbx2/3 6

B Genes whose functions in the later development were hardly examined, because of their expression in the early embryo

EphrinA-d, ets/pointed2, Fgf9/16/20, FoxA-a, macho-1 5

C Genes not analyzed

Genes for which no good morpholino oligonucleotides were obtainedCOE, Hox1, Mytf, Unc4-A, Wnt7 5

Genes staring to be expressed late in the early tailbud stageArix, FoxD-a/b, Hox5, islet 4 D

EVELO

PMENT

290

early in the ancestral chordates, as seen in the Ciona embryos, or thetiming of Fgf8 signaling might have shifted to earlier stages duringthe retrograde evolution of tunicates. Second, Wnt1 is absent in the

Ciona genome (Hino et al., 2003), and, hence, it is not involved inthe regionalization of the Ciona CNS. It is possible that the ancestralchordate had a regionalization mechanism directed by Fgf8 andWnt1, but that Wnt1 was subsequently lost by ascidians. However,in amphioxus, Wnt1 is not expressed around the boundary of theputative midbrain and hindbrain, and therefore the ancestralchordate might have relied solely on an Fgf8-dependent mechanismfor regionalization of the CNS.

Localized expression of Fgf8/17/18 depends on theSnail repressorIn vertebrate embryos, localized expression of Fgf8 in the MHB isdirected by the interaction of Otx and Gbx. However, restrictedexpression of Fgf8/17/18 in the A9.30 lineage of ascidian embryosdoes not depend on Otx and Gbx. MO-induced suppression of Otxgene activity did not affect Fgf8/17/18 expression (see Fig. S1N inthe supplementary material). Moreover, Gbx is not present in theCiona genome (Dehal et al., 2002; Wada et al., 2003). In contrast tovertebrates, Fgf8/17/18 is positively regulated by Nodal signalsemanating from the b-line neural cells (blue cells in Fig. 1) (Imai etal., 2006). The key component of Fgf8/17/18 regulation is thedifferential expression of the Snail repressor in the A9.30 and A9.32neural progenitors. Snail expression is explicitly stronger in A9.32than A9.30 (Fig. 6A). This augmented expression depends onNeurogenin, which is expressed only in A9.32. MO-inducedsuppression of Neurogenin resulted in reduced levels of Snail inA9.32, similar to the levels normally seen in A9.30 (whitearrowhead, Fig. 6A�). This reduction in Snail causes ectopicactivation of Fgf8/17/18 in the A9.32 lineage (white arrowhead, Fig.6B�). There is a similar de-repression of Fgf8/17/18 expression uponMO-mediated disruption of Snail activity (Fig. 6C).

Neurogenin is under the control of Nodal signaling, andsuppression of Nodal via MO injection led to the completeelimination of Snail expression in both A9.30 and A9.32 (Imai etal., 2006). Thus, Nodal is sufficient to induce low levels of Snail


Fig. 4. FgF8/17/18 delineates PSV and neck regions.(A-H) Expression of (A,B) Otx, (C,D) En, (E,F) Pax2/5/8-A and (G,H) Hox1in A,C,E,G control embryos and (B,D,F,H) experimental embryosdeveloped from eggs injected with Fgf8/17/18 MO. Cell identities weredetermined by DAPI staining of nuclei (inserts in A-H). Cells from theposterior end of the SV (A11.63) to the middle part of the VG (A11.118or its descendants) are enclosed by red and light-blue lines, which showthe expression or lack of expression of the indicated genes, respectively.Broken white lines indicate the boundaries of the PSV/neck and theneck/VG. Note that A11.119 is about to divide, or has recently divided,into two daughter cells in A,B,E,F. For simplicity, they are enclosed bysingle lines. The embryos shown in G and H are at a slightly later stage,when Hox1 gene expression is more prominent. (I,J) Schematicrepresentations of the brain regionalization mechanism by Fgf8/17/18in (I) the normal embryo and (J) Fgf8/17/18-morphant embryo. Notethat Fgf8/17/18 is not expressed in the VG at the tailbud stage but isexpressed in the progenitor cells (A9.30) in the neural plate.

Fig. 5. The Fgf signaling between the late gastrula stage and theneurula stage is required for Pax2/5/8-A expression.(A-D) Expression of Fgf8/17/18 from the 44-cell stage to the neurulastage. Expression in A7.6 is shown by black arrowheads and expressionin the A9.30 is shown by white arrowheads. (E-H) Expression ofPax2/5/8-A following U0126 treatment at different time points.Numbers indicate the total number of embryos analyzed. Thedevelopmental time point when embryos were placed in sea watercontaining U0126 is shown in the top. Embryos treated with U0126from the late gastrula stage when Fgf8/17/18 expression in A9.30begins do not express Pax2/5/8-A, whereas embryos treated withU0126 from the neurula stage express Pax2/5/8-A.

DEVELO

PMENT

expression in A9.30. These low levels fail to block Fgf8/17/18expression, but might restrict the levels of expression. Theenhanced expression of Snail seen in A9.32 arises from a feed-forward loop: Nodal induces Neurogenin and the two regulatorswork together to activate Snail (Fig. 6D). These augmented levelsof Snail result in the complete repression of Fgf8/17/18 in theA9.32 lineage of the CNS.

The localized expression of Neurogenin is achieved by twosignaling molecules. Nodal is expressed in the b-line cells (bluecells in Fig. 1 and Fig. 6D), which are juxtaposed to A9.32 but notA9.30. Similarly, Delta-like is expressed in the b-line cells underthe control of Nodal signaling (Hudson and Yasuo, 2005; Imai etal., 2006), and the Delta-Notch signaling contributes to activationof Neurogenin (see Fig. S1Ad in the supplementary material).Thus, these two inputs induce localized expression of Neurogeninin A9.32 cooperatively or by a simple cascade mechanism.

It has been proposed that the ancestral chordates had theOtx/Gbx system but lacked an apparent MHB organizer, as theOtx and Gbx expression patterns abut in amphioxus CNS,although other components of the MHB organizer such as En,Pax2/5/8 and Wnt1 are not expressed at this boundary (Castro etal., 2006). If so, the recruitment of Fgf8 signaling as an MHBorganizer might have evolved after the divergence of thecephalochordates and tunicate/vertebrate lineages. There are twopossible scenarios for the advent of Fgf8 in thecompartmentalization of the vertebrate CNS. First, the Otx/Gbxgene circuit might have been used to regulate Fgf8 in the ancestral

chordate, and the switch in Fgf8 regulation to theNodal/Neurogenin/Snail gene circuit might reflect the streamlinedcell lineages in the early Ciona embryo and the need for theprecise regulation of Fgf8 at single-cell resolution. Alternatively,Fgf8 regulation by the Nodal/Neurogenin/Snail gene circuit mightbe ancient, and in vertebrates the upstream regulatory mechanismwas integrated into a pre-existing Otx/Gbx gene circuit. In thisregard, we note that orthologs of Fgf8 are regulated by Snailduring gastrulation of the Drosophila embryo (e.g. Stathopouloset al., 2004).

Hox1 expression is controlled by Fgf8/17/18through retinoic acid signalingAs discussed earlier, Fgf8/17/18 signaling inhibits Otx expressionin the posterior A9.16 lineage, thereby restricting its expressionto the anterior lineage, the future PSV. Otx either directly orindirectly activates the forkhead regulatory gene FoxB, whichrepresses Hox1 expression in the PSV. It is well known that Hoxgenes are regulated by retinoic acid (RA) signaling in chordates(Holland and Holland, 1996; Maden, 2002). It has previouslybeen shown that RA enhances Hox1 expression in the Cionaembryo (Nagatomo and Fujiwara, 2003). RA is synthesized byRaldh2 (retinaldehyde dehydrogenase 2), which is expressed inthe most anterior muscle cells at the tailbud stage (Fig. 7A)(Nagatomo and Fujiwara, 2003). Indeed, knockdown of Raldh2eliminates Hox1 expression (Fig. 7B,C). Endogenouslysynthesized RA is therefore responsible for Hox1 expression.Interestingly, Cyp26 is expressed in the presumptive PSV regionof the Ciona CNS, and this expression is lost in morphantembryos injected with either Otx or FoxB MOs (see Fig. S1F inthe supplementary material). Cyp26 encodes an enzymeresponsible for degrading RA. These observations raise the


Fig. 6. Snail directs localized expression of Fgf8/17/18. (A-C) Expression of (A,A�) Snail and (B,B�,C) Fgf8/17/18 was examinedin (A,B) control unperturbed embryos, (A�,B�) Neurogenin-knockdownembryos and (C) Snail-knockdown embryos. Red and white arrowheadsindicate A9.30 and A9.32, respectively. (Insets in A and A�) High-magnification images of the left side of the embryos shown in A andA�. A9.30 and A9.32 are enclosed by red and white lines, respectively.Note that the embryos shown in B,B� and C are slightly older thanthose in A and A�. (D) A schematic diagram of regulation of Fgf8/17/18.See the text for details.

Fig. 7. Retinoic acid is required for Hox1 expression. (A) Radlh2 isexpressed in the most anterior muscle cells at the tailbud stage (upperpanel, a lateral view; lower panel, a dorsal view). (B) Hox1 is expressedin the neck region, the anterior part of the VG and the anterior part ofthe caudal nerve cord within the CNS of control unperturbed embryos.Red and yellow arrowheads indicate boundaries between the neck andthe visceral ganglion and between the visceral ganglion and the caudalnerve cord, respectively. (C) Hox1 expression is not observed in RaldH1-knockdown embryos. (D) A schematic diagram of regulation of Hox1.FoxB represses Hox1, and this repression may be through Cyp26, whichis expressed under the control of FoxB. See the text for details.

DEVELO

PMENT

292

possibility that FoxB indirectly represses Hox1 expression in thepresumptive PSV by activating Cyp26, which in turn inhibits RAsignaling (Fig. 7D). Previous studies suggested a possibleconnection between Fgf signaling and Hox expression (Irving andMason, 2000; Shimizu et al., 2006; Skromne et al., 2007) andbetween Cyp26 and Hox expression in the vertebrate CNS(Hernandez et al., 2007). We propose that this connection mightreflect a direct regulatory connection between the MHB organizerand RA signaling.

DISCUSSIONThe comprehensive analysis of CNS regulatory genes led to theelucidation of a provisional gene network in the early Ciona tadpole.These networks provide a number of key insights into thecompartmentalization of the chordate CNS. First, a localized Fgf8signaling center was probably used by the last shared ancestor ofascidians and vertebrates to delineate two regions of the chordatebrain (mesencephalon and metencephalon). Second, Fgf8 signalingin Ciona leads to restricted expression of Otx and FoxB in the PSV,as well as restricted expression of Pax2/5/8-A in the neck. Otx andFoxB might inhibit Hox1 expression in the forebrain via Cyp26,whereas Pax2/5/8-A might coordinate the expression of theregulatory genes required for the differentiation of metencephalonmotoneurons, such as Phox2a/Arix (e.g. Engle, 2006). Finally,although the regulatory genes responsible for thecompartmentalization of the vertebrate CNS (e.g. Otx, Pax2,Neurogenin, etc.) exhibit comparable patterns of expression in theCiona CNS, there are both conserved and distinctive features of theunderlying mechanism. Localized Fgf8 signaling is used to deploythese expression patterns in both systems, even though differentregulatory mechanisms are used to restrict Fgf8.

This research was mainly supported by Grants-in-Aid from MEXT (17687022)to Y.S. and the NSF (IOB 0445470) to M.L. and partly by a grant-in-aid of thegCOE program from MEXT to K.S.I. We thank Prof. Nori Satoh for thegenerous use of his laboratory facilities.

Supplementary materialSupplementary material for this article is available athttp://dev.biologists.org/cgi/content/full/136/2/285/DC1

ReferencesCanestro, C., Bassham, S. and Postlethwait, J. (2005). Development of the

central nervous system in the larvacean Oikopleura dioica and the evolution ofthe chordate brain. Dev. Biol. 285, 298-315.

Castro, L. F., Rasmussen, S. L., Holland, P. W., Holland, N. D. and Holland,L. Z. (2006). A Gbx homeobox gene in amphioxus: insights into ancestry ofthe ANTP class and evolution of the midbrain/hindbrain boundary. Dev. Biol.295, 40-51.

Chi, C. L., Martinez, S., Wurst, W. and Martin, G. R. (2003). The isthmicorganizer signal FGF8 is required for cell survival in the prospective midbrainand cerebellum. Development 130, 2633-2644.

Cole, A. G. and Meinertzhagen, I. A. (2004). The central nervous system ofthe ascidian larva: mitotic history of cells forming the neural tube in lateembryonic Ciona intestinalis. Dev. Biol. 271, 239-262.

Corbo, J. C., Levine, M. and Zeller, R. W. (1997). Characterization of anotochord-specific enhancer from the Brachyury promoter region of theascidian, Ciona intestinalis. Development 124, 589-602.

Dehal, P., Satou, Y., Campbell, R. K., Chapman, J., Degnan, B., De Tomaso,A., Davidson, B., Di Gregorio, A., Gelpke, M., Goodstein, D. M. et al.(2002). The draft genome of Ciona intestinalis: insights into chordate andvertebrate origins. Science 298, 2157-2167.

Dufour, H. D., Chettouh, Z., Deyts, C., de Rosa, R., Goridis, C., Joly, J. S.and Brunet, J. F. (2006). Precraniate origin of cranial motoneurons. Proc.Natl. Acad. Sci. USA 103, 8727-8732.

Dynes, J. L. and Ngai, J. (1998). Pathfinding of olfactory neuron axons tostereotyped glomerular targets revealed by dynamic imaging in living zebrafishembryos. Neuron 20, 1081-1091.

Engle, E. C. (2006). The genetic basis of complex strabismus. Pediatr. Res. 59,343-348.

Hernandez, R. E., Putzke, A. P., Myers, J. P., Margaretha, L. and Moens, C.B. (2007). Cyp26 enzymes generate the retinoic acid response patternnecessary for hindbrain development. Development 134, 177-187.

Hino, K., Satou, Y., Yagi, K. and Satoh, N. (2003). A genomewide survey ofdevelopmentally relevant genes in Ciona intestinalis. VI. Genes for Wnt,TGFbeta, Hedgehog and JAK/STAT signaling pathways. Dev. Genes Evol. 213,264-272.

Holland, L. Z. and Holland, N. D. (1996). Expression of AmphiHox-1 andAmphiPax-1 in amphioxus embryos treated with retinoic acid: insights intoevolution and patterning of the chordate nerve cord and pharynx.Development 122, 1829-1838.

Hudson, C. and Yasuo, H. (2005). Patterning across the ascidian neural plate bylateral Nodal signalling sources. Development 132, 1199-1210.

Hudson, C., Lotito, S. and Yasuo, H. (2007). Sequential and combinatorialinputs from Nodal, Delta2/Notch and FGF/MEK/ERK signalling pathwaysestablish a grid-like organisation of distinct cell identities in the ascidian neuralplate. Development 134, 3527-3537.

Ikuta, T. and Saiga, H. (2007). Dynamic change in the expression ofdevelopmental genes in the ascidian central nervous system: revisit to thetripartite model and the origin of the midbrain-hindbrain boundary region.Dev. Biol. 312, 631-643.

Imai, K., Takada, N., Satoh, N. and Satou, Y. (2000). β-catenin mediates thespecification of endoderm cells in ascidian embryos. Development 127, 3009-3020.

Imai, K. S., Satoh, N. and Satou, Y. (2002). Region specific gene expressions inthe central nervous system of the ascidian embryo. Mech. Dev. 119 Suppl. 1,S275-S277.

Imai, K. S., Hino, K., Yagi, K., Satoh, N. and Satou, Y. (2004). Geneexpression profiles of transcription factors and signaling molecules in theascidian embryo: towards a comprehensive understanding of gene networks.Development 131, 4047-4058.

Imai, K. S., Levine, M., Satoh, N. and Satou, Y. (2006). Regulatory blueprintfor a chordate embryo. Science 312, 1183-1187.

Irving, C. and Mason, I. (2000). Signalling by FGF8 from the isthmus patternsanterior hindbrain and establishes the anterior limit of Hox gene expression.Development 127, 177-186.

Jaszai, J., Reifers, F., Picker, A., Langenberg, T. and Brand, M. (2003).Isthmus-to-midbrain transformation in the absence of midbrain-hindbrainorganizer activity. Development 130, 6611-6623.

Lemaire, P., Bertrand, V. and Hudson, C. (2002). Early steps in the formationof neural tissue in ascidian embryos. Dev. Biol. 252, 151-169.

Liu, A. and Joyner, A. L. (2001). Early anterior/posterior patterning of themidbrain and cerebellum. Annu. Rev. Neurosci. 24, 869-896.

Lowe, C. J., Wu, M., Salic, A., Evans, L., Lander, E., Stange-Thomann, N.,Gruber, C. E., Gerhart, J. and Kirschner, M. (2003). Anteroposteriorpatterning in hemichordates and the origins of the chordate nervous system.Cell 113, 853-865.

Maden, M. (2002). Retinoid signalling in the development of the centralnervous system. Nat. Rev. Neurosci. 3, 843-853.

Meinertzhagen, I. A., Lemaire, P. and Okamura, Y. (2004). The neurobiologyof the ascidian tadpole larva: recent developments in an ancient chordate.Annu. Rev. Neurosci. 27, 453-485.

Moret, F., Christiaen, L., Deyts, C., Blin, M., Vernier, P. and Joly, J. S. (2005).Regulatory gene expressions in the ascidian ventral sensory vesicle:evolutionary relationships with the vertebrate hypothalamus. Dev. Biol. 277,567-579.

Nagatomo, K. and Fujiwara, S. (2003). Expression of Raldh2, Cyp26 and Hox-1 in normal and retinoic acid-treated Ciona intestinalis embryos. Gene Expr.Patterns 3, 273-277.

Nicol, D. and Meinertzhagen, I. A. (1988). Development of the centralnervous system of the larva of the ascidian, Ciona intestinalis L. II. Neural platemorphogenesis and cell lineages during neurulation. Dev. Biol. 130, 737-766.

Reichert, H. (2005). A tripartite organization of the urbilaterian brain:developmental genetic evidence from Drosophila. Brain Res. Bull. 66, 491-494.

Rhinn, M., Picker, A. and Brand, M. (2006). Global and local mechanisms offorebrain and midbrain patterning. Curr. Opin. Neurobiol. 16, 5-12.

Satou, Y. and Satoh, N. (1997). Posterior end mark 2 (pem-2), pem-4, pem-5,and pem-6: maternal genes with localized mRNA in the ascidian embryo. Dev.Biol. 192, 467-481.

Satou, Y., Yamada, L., Mochizuki, Y., Takatori, N., Kawashima, T., Sasaki,A., Hamaguchi, M., Awazu, S., Yagi, K., Sasakura, Y. et al. (2002). AcDNA resource from the basal chordate Ciona intestinalis. Genesis 33, 153-154.

Shimizu, T., Bae, Y.-K. and Hibi, M. (2006). Cdx-Hox code controlscompetence for responding to Fgfs and retinoic acid in zebrafish neural tissue.Development 133, 4709-4719.

Skromne, I., Thorsen, D., Hale, M., Prince, V. E. and Ho, R. K. (2007).Repression of the hindbrain developmental program by Cdx factors is requiredfor the specification of the vertebrate spinal cord. Development 134, 2147-2158.


DEVELO

PMENT

Stathopoulos, A., Tam, B., Ronshaugen, M., Frasch, M. and Levine, M.(2004). pyramus and thisbe: FGF genes that pattern the mesoderm ofDrosophila embryos. Genes Dev. 18, 687-699.

Takahashi, T. and Holland, P. W. (2004). Amphioxus and ascidian Dmbxhomeobox genes give clues to the vertebrate origins of midbraindevelopment. Development 131, 3285-3294.

Wada, H., Saiga, H., Satoh, N. and Holland, P. W. (1998). Tripartiteorganization of the ancestral chordate brain and the antiquity of placodes:insights from ascidian Pax-2/5/8, Hox and Otx genes. Development 125, 1113-1122.

Wada, S., Tokuoka, M., Shoguchi, E., Kobayashi, K., Di Gregorio, A.,Spagnuolo, A., Branno, M., Kohara, Y., Rokhsar, D., Levine, M. et al.(2003). A genomewide survey of developmentally relevant genes in Cionaintestinalis. II. Genes for homeobox transcription factors. Dev. Genes Evol.213, 222-234.

Wurst, W. and Bally-Cuif, L. (2001). Neural plate patterning: upstream anddownstream of the isthmic organizer. Nat. Rev. Neurosci. 2, 99-108.

Yoshida, R., Sakurai, D., Horie, T., Kawakami, I., Tsuda, M. and Kusakabe,T. (2004). Identification of neuron-specific promoters in Ciona intestinalis.Genesis 39, 130-140.


DEVELO

PMENT

Documents

Gene regulatory networks underlying the compartmentalization of the Ciona central nervous system