Murine Cerberus Homologue mCer-1: A Candidate Anterior ... · Mouse Cerberus Homologue mCer-1 137...

DEVELOPMENTAL BIOLOGY 194, 135–151 (1998)ARTICLE NO. DB978812

Murine Cerberus Homologue mCer-1: A CandidateAnterior Patterning Molecule

C. Biben,*,1 E. Stanley,*,1 L. Fabri,† S. Kotecha,‡ M. Rhinn,§C. Drinkwater,† M. Lah,† C-C. Wang,* A. Nash,† D. Hilton,*,†S-L. Ang,§ T. Mohun,‡ and R. P. Harvey**The Walter and Eliza Hall Institute of Medical Research and Cooperative Research Centre forCellular Growth Factors, Post Office, Royal Melbourne Hospital, Victoria 3050, Australia;†Amrad Operations Pty Ltd, 576 Swan Street, Richmond 3121, Australia; ‡National Institutefor Medical Research, The Ridgeway, Mill Hill, London NW7 1AA, United Kingdom; and§Institut de Genetique et de Biologie Moleculaire et Cellulaire, CNRS/INSERM/UniversiteLouis Pasteur/College de France, BP163, 67404 Illkirch Cedex, C.U. de Strasbourg, France

Xenopus cerberus (Xcer) is a cytokine expressed in anterior mesendoderm overlapping and surrounding Spemann’sgastrula organiser. When misexpressed in blastomeres, Xcer can induce ectopic heads with well-defined brain, cementgland, olfactory placodes, cyclopic eye, and occasionally liver and heart. We report here the identification of mCer-1,a murine gene related to cerberus. Both mCer-1 and Xcer appear to belong to the cystine knot superfamily, whichincludes TGFbs and BMPs. In Xenopus animal cap assays, mCer-1 and Xcer induced cement glands and markers ofanterior neural tissue and endoderm, characteristic of BMP inhibition. Furthermore, both antagonised the ventrolateralmesoderm-inducing activity of coexpressed BMP4. In mouse embryos, mCer-1 was expressed at early gastrulation ina stripe of primitive endoderm along the future anterior side of the egg cylinder, a region essential for anterior patterning.A second phase of expression was detected in anterior embryonic mesendoderm, and by late-streak stages most of theanterior half of the embryo was positive, except for the node and cardiac progenitors. Expression was later seen in thecranial portion of the two most-recently formed somites and in two stripes within presomitic mesoderm. In embryoslacking Otx2, a homeogene with a demonstrated role in anterior patterning, mCer-1 was still expressed in an anteriorzone, although often abnormally. The data suggest that mCer-1 shares structural, functional, and expression character-istics with Xcer and may participate in patterning the anterior of the embryo and nascent somite region, in part,through a BMP-inhibitory mechanism. q 1998 Academic Press

INTRODUCTION 1994). The organiser itself is induced at an earlier time bya signalling centre within dorsal endoderm (the Nieuwkoop

Studies on vertebrate axis formation have been guided by center), established through activation of b-catenin andthe pioneering work of Spemann and Mangold, who demon- other downstream components of the Wnt signalling cas-strated that the dorsal blastoporal lip from a salamander cade, and expression of the homeodomain protein Siamoisgastrula could induce a complete secondary body axis when (Carnac et al., 1996; Wylie et al., 1996; Brannon and Kimel-grafted to the opposite ventral side (Spemann and Mangold, man, 1996; Fagotto et al., 1997). Functional equivalents of1924). This inducing region, Spemann’s organiser, contri- Spemann’s organiser have been recognised as the node inbutes cells to the notochord, prechordal mesoderm, floor- mice, Hensen’s node in chickens, and the shield in zebrafishplate, and foregut and has three principle roles: it initiates (Waddington, 1933; Ho, 1992; Beddington, 1994).and executes the convergent extension movements that The first hint of the existence of a distinct head organiserdrive elongation of the body axis, dorsalises mesoderm, and came from Spemann’s own observations that early organiserinduces neural tissue (Gerhart et al., 1991; Sulik et al.,

grafts induce a complete secondary axis, including heads,while grafts from older embryos induce only trunk (Spem-ann, 1931). Similar findings have been made in other verte-1 These authors contributed equally to this work.

135

0012-1606/98 $25.00Copyright q 1998 by Academic PressAll rights of reproduction in any form reserved.

AID DB 8812 / 6x37$$$141 02-09-98 14:51:23 dba

136 Biben et al.

brates. In chickens, early node grafts form a complete axis, son et al., 1997; Dosch et al., 1997). Second, dorsal fates areestablished autonomously after inhibition of ventral factorsbut lose head organising activity abruptly with age (Kintner

and Dodd, 1991; Storey et al., 1992). In mouse and zebrafish, (Hemmati-Brivanlou and Melton, 1997; Sasai and De Rob-ertis, 1997). This has led to the remarkable conclusion thatnode and shield grafts also induce secondary axes, but these

lack markers of the most anterior reaches of the central induction of neural tissue, one of the key functions of theorganiser, is mediated by the activity of negative rather thannervous system (CNS) (Beddington, 1994; Shih and Fraser,

1996; P. Tam, pers. comm.), suggesting that the head or- positive factors and that neural is both the dorsal and thedefault state of ectoderm (Sasai and De Robertis, 1997;ganising centre in these species is physically separate. Indi-

rect support for distinct head and trunk organising centres Weinstein and Hemmati-Brivanlou, 1997; but see Streit etal., 1997). Third, as originally proposed by Nieuwkoop, neu-has also come from molecular analysis of the frog organiser.

A growing number of signalling molecules and transcription ral tissue thus induced has anterior (head) character, unlessposteriorised by additional factors such as retinoic acid, ba-factors are known to be expressed in subdomains of the

organiser, and some have axis-inducing potential (Lemaire sic fibroblast growth factor, or Wnts (Sasai and De Robertis,1997).and Kodjabachian, 1996). Those that do, however, generally

induce secondary axes without heads. The only molecules The ventral factors BMP4 and Wnt-8, when misexpressedin embryos, interfere with head formation (Dale et al., 1992;that can alone induce a full secondary axis when injected

into ventral blastomeres are those that recreate a Nieuw- Jones et al., 1992; Christian and Moon, 1993), but neitherBMP nor Wnt inhibition alone induces ectopic heads (Hop-koop centre—Wnts and their downstream signalling com-

ponents, as well as Siamois (Brannon and Kimelman, 1996; pler et al., 1996; Lemaire and Kodjabachian, 1996; Leyns etal., 1997). In a recent report, the combined effects of BMPCarnac et al., 1996; Fagotto et al., 1997).

The location of the head inducer in mammalian embryos and Wnt inhibition were found to induce a full secondarybody axis including head, without inducing an additionalis unknown, although recent reports suggest that it lies

within primitive endoderm, the physical and genetic integ- Nieuwkoop centre (Glinka et al., 1997). The head organisermay therefore function to create a BMP- and Wnt-free zonerity of which is required for anterior patterning (Popperl et

al., 1995; Thomas and Beddington, 1996; Varlet et al., 1997). throughout the prospective head region.A recent addition to the list of genes known to be ex-Analysis of the transcription factor genes Lim1 and Otx2,

both expressed in primitive endoderm, supports such a role. pressed in the Xenopus organiser is cerberus (Bouwmeesteret al., 1996), encoding a novel secreted protein. cerberusMouse embryos carrying targeted mutations in Lim1 and

Otx2 lack anterior mesoderm and brain structures cranial (Xcer) expression occurs in the deep mesendodermal layerof the organiser. When misexpressed throughout embryos,of rhombomere 3, and rare Otx2//0 and Lim10/0 embryos

that survive to birth lack heads (Matsuo et al., 1995; Xcer induces enlarged heads at the expense of trunk. Re-markably, if expression is targeted to a specific blastomere,Shawlot and Behringer, 1995; Acampora et al., 1995; Ang

et al., 1996). ectopic heads that contain well-defined brain, placodes,cyclopic eye, cement gland (an anterior secretory organ),Molecular data have hinted that head induction, at least

in frogs, can be understood in the context of dorsal/ventral and foregut tissues, associated occasionally with a heart andliver, can be induced (Bouwmeester et al., 1996). In animal(D/V) axis specification and how the organiser participates

in this process. An emerging concept is that D/V values are cap assays, Xcer induces a range of anterior tissues such asforebrain and midbrain, cement gland, and foregut, in theestablished through mutual antagonism between signalling

molecules expressed at one axial pole and inhibitory mole- absence of mesoderm induction. These latter properties areshared with noggin, chordin, and follistatin, as well as othercules expressed at the other (Sasai et al., 1994; Thomas and

Beddington, 1996; Smith et al., 1997; Leyns et al., 1997; BMP signalling inhibitors (Hemmati-Brivanlou and Melton,1997; Tsuneizumi et al., 1997; Sasai and De Robertis, 1997).Sasai and De Robertis, 1997; Wang et al., 1997). Bone mor-

phogenetic protein 4 (BMP4) and Wnt-8, for example, are Xcer has also been shown to have potent anti-Wnt activity(Glinka et al., 1997). Thus, Xcer, with characteristics ofexpressed on the ventral side of the embryo, while the anti-

BMP factors noggin, chordin, and follistatin, and anti-Wnt both a BMP and a Wnt inhibitor, may participate directlyor indirectly in freeing the prospective head region of thefactor Frzb-1, are expressed on the dorsal side as products

of the organiser. These inhibitory molecules sequester their influence of these ventralising factors.To advance our understanding of anterior patterning intarget factors in the extracellular space and prevent them

activating cognate receptors (Zimmerman et al., 1996; Pic- mammals, we searched for murine Xcer homologues by da-tabase screening. We report here the identification ofcolo et al., 1996; Wang et al., 1997; Fainsod et al., 1997;

Leyns et al., 1997). There are several important implications mCer-1, initially detected as an EST derived from an embry-onic day E7.5 (E7.5) cDNA library (Harrison et al., 1995).from these findings. First, antagonism between ventral fac-

tors and their inhibitors is thought to establish a graded mCer-1 is related to Xcer principally over a cysteine-richregion similar to the cystine knot motif. Even though theconcentration of signalling molecules across the D/V axis,

eliciting a concentration-dependent, stepwise series of tis- homology between them is only modest, the expressionpatterns of mCer-1 and Xcer are remarkably similar. Fur-sue responses. This morphogen-like activity has been dem-

onstrated for BMP4 in both ectoderm and mesoderm (Wil- thermore, mCer-1 could mimic the anti-BMP activity of

Copyright q 1998 by Academic Press. All rights of reproduction in any form reserved.

AID DB 8812 / 6x37$$$142 02-09-98 14:51:23 dba

137Mouse Cerberus Homologue mCer-1

Xcer in Xenopus animal caps (Bouwmeester et al., 1996), as 3 of 4 additional cysteines present in NDP and mucins(Fig. 1B). Molecular modelling of NDP using coordinatesalthough with reduced potency. In the early gastrula, mCer-

1 expression was restricted to anterior primitive endoderm, from the TGFb2 crystal structure suggests that these addi-tional cysteines form two disulphide bridges which couldthe tissue thought to initiate head induction in mammals

(Thomas and Beddington, 1996; Popperl et al., 1997; Varlet further stabilise the knot (Meitinger et al., 1993). NeithermCer-1 nor Xcer contain the cysteine which, in TGFbs, iset al., 1997). Expression was also detected in anterior embry-

onic mesendoderm, required for regionalisation of the cra- utilised for dimerisation (Fig. 1B).nial CNS (Ang et al., 1994), and in the rostral domain ofnascent somites. These features make mCer-1 an excellent

mCer-1 Is a Secreted Dimercandidate patterning molecule for anterior development inthe mouse. A PCR-generated cDNA fragment spanning the mCer-1

coding region was cloned into the expression vector pEFBOSI FLAG for production of recombinant protein carrying a C-terminal FLAG epitope. CFLAG–mCer-1 could be recov-RESULTSered from culture supernatants of both Chinese hamsterovary (CHO) cells, after stable integration of vector, andXcer-Related cDNA Clones293T cells after transient transfection (Fig. 2). CFLAG–

Database searches detected homology between Xcer and mCer-1 secreted from both cell types migrated principallyan open reading frame within GenBank EST AA120122, at 38 kDa on reducing and denaturing gels. The size of thea mouse E7.5 cDNA (Harrison et al., 1995) deposited by CHO cell product was reduced to 32 kDa after treatmentR. S. P. Beddington (NIMR, London). Primers based on the with N-glycosidase F (data not shown), demonstrating thatEST were used in combination with vector primers to ex- mCer-1, as expected of a secreted protein, is N-glycosylated.tend the EST sequence in 5* and 3* directions. The complete Examination of CHO cell preparations on denaturing andcoding region was then isolated on a single fragment using nonreducing gels indicated that most unaggregated proteinprimers flanking the putative initiation and termination migrated as monomer, with little dimer detected (data notcodons. Multiple isolates were sequenced and a consensus shown). Since all known cystine knot cytokines are secretedwas determined. Two clones containing the consensus only as homo- or heterodimers (Isaacs, 1995), we examinedwere used for expression and functional studies (see Materi- CFLAG–mCer-1 secreted from 293T cells and, in this case,als and Methods and below). found migration at a size consistent with homodimerisa-

tion, with little monomer detected (Fig. 2). The data demon-strate that mCer-1 is a secreted glycoprotein that can formmCer-1 Proteinpredominantly dimers when expressed from certain celltypes. Whether these are in fact homodimers will requireThe PCR-generated murine cDNA (mCer-1) has an open

reading frame of 273 amino acids, 2 amino acids longer than further analysis.that of Xcer (Fig. 1A). Overall, the mouse and frog proteinsshare 26% identity and 42% similarity, much lower than

Anti-BMP-like Activity of mCer-1 in Xenopusfor frog and mouse homologues of other developmentallyAnimal Capsimportant cytokines such as BMPs and activins and their

inhibitors noggin, chordin, and follistatin (see Materials and Xcer has been shown to have an anti-BMP-like activitywhen expressed in Xenopus animal cap assays, inducingMethods for GenBank accession numbers). A higher level

of identity (57%) was seen over a C-terminal cysteine-rich Otx2, a marker of anterior embryonic structures, as wellas markers of neural tissue, cement gland, and endodermregion within which the spacing of cysteines conformed to

that expected for a motif termed the cystine knot (McDon- (Bouwmeester et al., 1996). To examine whether mCer-1shared this property, we compared the activities ofald and Hendrickson, 1993; Isaacs, 1995), the structural

backbone of a number of cytokines including the trans- mCer-1 and Xcer in animal cap assays after injection ofsynthetic mRNAs at the fertilised egg stage. First, formationforming growth factor b superfamily (TGFbs), BMPs, nerve

growth factor, platelet-derived growth factor, and Norrie of cement glands in individual animal caps was scored atstage 35, since these are easily recognisable as superficialdisease protein (NDP). The same motif is seen in hormones

such as human chorionic gonadotrophin, in secreted struc- darkly pigmented patches secreting a sticky exudate. Bothnative mCer-1 and CFLAG–mCer-1 were able to inducetural proteins such as mucins and von Willebrand Factor,

and in a variety of inhibitors and toxins (Isaacs, 1995). pigmented cement glands (Fig. 3A) with equal frequency(Fig. 3B), although at only one-quarter the frequency of Xcer.mCer-1 and Xcer contain the 6 conserved cysteines thought

to configure the cystine knot: a peptide ring formed from Next, RT-PCR analysis was used to assess expression ofvarious markers in pools of 25–30 animal caps derived fromtwo closely spaced disulphide bridges, with a third disul-

phide bridge passing through its centre in a knot-like topol- injected and uninjected embryos (Fig. 3C). Both mCer-1 andXcer mRNAs showed near identical activities. As pre-ogy (Figs. 1B and 1C). Both cer proteins contain the sequence

C-X-G-X-C conserved in all cystine knot cytokines, as well viously demonstrated for Xcer (Bouwmeester et al., 1996),


AID DB 8812 / 6x37$$$142 02-09-98 14:51:23 dba

138 Biben et al.

FIG. 1. (A) Alignment of the predicted amino acid sequences of mCer-1 and Xenopus Xcer. Amino acid identities are shaded andsimilarities (see Materials and Methods) shown within open boxes. The putative cystine knot domains of mCer-1 and Xcer are outlined.The potential hydrophobic signal sequence of mCer-1 is overlined. The arrowhead indicates the N-terminus of secreted CFLAG–mCer-1 produced from CHO and 293T cells, as determined by N-terminal amino acid sequence analysis. (B) Alignment of the putativecystine knot regions of mCer-1, Xcer, mouse Norrie disease protein (NDP), and rat mucin with the equivalent domain of human TGFb2,known to form a cystine knot from crystallographic analysis. Cysteines are numbered according to Meitinger et al. (1993), with 1–6 (red)forming the conserved core of the cystine knot (see C), whilst a–d (yellow) are proposed to form additional stabilising disulphide bridgesin some members. Note that mCer-1 and Xcer have additional cysteines a–c, but not d. Another cysteine (blue), used for dimerisation inTGFbs, is absent in mCer-1 and Xcer. The star indicates the glycine conserved in all cytokine members of the cystine knot superfamily.(C) Schematic representation of a cystine knot protein. The disulphide bonds that configure the cystine knot are indicated by red verticallines, while the proposed additional disulphide bridges in NDP (Meitinger et al., 1993) are represented by blue lines. The GenbankAccession No. for mCer-1 is AF035579.

mCer-1 induced the panneural marker N-CAM (Kintner and Mohun, 1993), even when caps were cultured beyond theequivalent of stage 40 (data not shown). Thus, the homeo-Melton, 1987), although rather weakly; the cement gland

marker CG13 (Jamrich and Sato, 1989; Sive et al., 1989); box markers may be expressed in endoderm, in which casethey would indicate foregut character (see Sasai et al., 1996).and the endoderm-enriched marker Edd (Sasai et al., 1996).

Induced tissues apparently had anterior character, since However, it is also possible that early stages of cardiogenesisare activated, without realisation of the whole program.Otx2, a marker of anterior embryonic structures such as

forebrain, midbrain, and cement gland (Pannese et al., 1995; The inductive activities of mCer-1 and Xcer in animalcaps are characteristic of a partial inhibition of BMP signal-Gammill and Sive, 1997) was strongly induced, while

Krox20, a hindbrain marker (Bradley et al., 1993), was not. ling (Sasai et al., 1996). In animal caps, BMPs can act asboth mesoderm-inducing and ventralising agents, whilstThe related homeobox genes, XNkx-2.5 and -2.3, normally

expressed in cardiogenic mesoderm and anterior (pharyn- strongly inhibiting formation of neural tissue (Jones et al.,1992; Dale et al., 1992). Consistent with these reports, wegeal) endoderm (Tonissen et al., 1994; Evans et al., 1995),

were strongly induced. There was no induction of the car- found that BMP4 induced expression of T4 globin (Banvilleand Williams, 1985) and XeHAND (D. B. Sparrow, S. Ko-diac-specific differentiation marker, XMLC2a (Logan and


AID DB 8812 / 6x37$$$142 02-09-98 14:51:23 dba


FIG. 1—Continued

techa, N. Towers, and T. J. Mohun, submitted for publica- Xcer and BMP4 act in a mutually antagonistic way throughthe same pathway. Although only semiquantitative, thetion), markers of ventral and lateral mesoderm, respectively

(Fig. 3C). To examine the relationship between cerberus and RT-PCR results suggest that BMP4 more readily overridesthe inductive effects of mCer-1 versus those of Xcer (Fig.BMP4 signalling and to further compare the activities of the

mouse and frog proteins, mCer-1 and Xcer were coexpressed 3C), consistent with the mouse gene being less potent inthe animal cap assay.with BMP4 in animal caps by coinjection of equimolar

amounts of their mRNAs into eggs. Both mCer-1 and Xcerantagonised BMP4 responses, whilst BMP4 antagonised

mCer-1 Expression in Gastrulating EmbryosmCer-1 and Xcer responses (Fig. 3C). Thus, markers inducedspecifically by mCer-1 and Xcer but not BMP4 (XNkx-2.3, Expression of mCer-1 was examined during postimplanta-

tion development by in situ hybridisation using a whole-XNkx-2.5, N-CAM, Otx2, CG13, and Edd) were reduced, aswere those specifically induced by BMP4 but not mCer-1 mount protocol and digoxygenin-labelled RNA probes.

mCer-1 transcripts were first evident at or just before theor Xcer (T4 globin and XeHAND). In agreement with Bouw-meester et al. (1996), we found that coinjected BMP4 mRNA onset of gastrulation in a stripe, several cell diameters in

width, along one side of the egg cylinder. This stripe ex-extinguished Xcer-induced N-CAM and CG13. However,other markers, Nkx-2.3, Nkx-2.5, and Edd (Xcer-induced) tended proximally from the embryonic/extraembryonic

junction to just short of the distal tip (Fig. 4A) and was onand T4 globin and XeHAND (BMP4-induced), were dimin-ished but not extinguished, leaving open the possibility that the opposite side of the egg cylinder to the primitive streak,


AID DB 8812 / 6x37$$$143 02-09-98 14:51:23 dba

140 Biben et al.

both mesodermal and endodermal layers (Figs. 4H and 4I).At no stage was expression seen in epiblast or ectoderm.

At ÇE7.75, expression was seen in a wedge of mesodermand endoderm at the anterior tip of the embryo (Fig. 4J).The positive zone clearly excluded cardiogeneic mesodermand its underlying endoderm (Fig. 4K). Of further interestwas whether mCer-1 is expressed in precursors of the noto-chord and prechordal mesoderm, since Xcer becomes ex-cluded from these tissues in the frog. The notochordal andprechordal plates are derivatives of the node which migrateanteriorly as the head process. In mouse, these cells arecontiguous with the endodermal layer, but have a distinct

FIG. 2. Western blot analysis of CFLAG–mCer-1 protein secreted morphology with a smaller ventral surface area (Sulik et al.,from 293T cells with (/) and without (0) reduction with 100 mM 1994). At E7.75, they extend far anteriorly, as judged bydithiothreitol (DTT). mCer-1 protein was detected with the M2 Shh expression (Echelard et al., 1993). In serial transverseanti-FLAG antibody. The mobility of molecular weight standards sections at this stage, notochordal plate was clearly positive(size shown in kilodaltons) is indicated on the left. for mCer-1 (Fig. 5). At more anterior levels, a broader zone

of distinctive midline cells was evident, consistent with thewedge-shaped pattern of prechordal mesoderm seen in chickand mouse (Sulik et al., 1994; Dale et al., 1997). These cells

abutting the future anterior region of epiblast fated to form were also positive for mCer-1.anterior neurectoderm (Quinlan et al., 1995). This locationwas confirmed by hybridising early streak embryos with

mCer-1 Expression in Nascent Somitesprobes for both mCer-1 and brachyury, which marks pro-spective mesodermal cells immediately after their passage mCer-1 expression faded completely from the anteriorthrough the streak (Herrmann, 1991) (Fig. 4B). Histological region during headfold stages. When expression was reducedsections of early streak embryos showed mCer-1 expression to a trace at the anterior ventral midline (Fig. 6A), twoin endoderm only (data not shown). Anterior endoderm at stripes became apparent in anterior paraxial mesodermearly streak stages is part of the primitive (or visceral) endo- (Figs. 6A and 6B). In progressively older embryos, expressiondermal lineage, which is eventually displaced into the extra- was detected in three or four stripes at more and more poste-embryonic region by definitive endoderm emanating from rior positions along the axis (Fig. 6C), with no other regionsthe streak (Lawson and Pedersen, 1987; Thomas and Bed- expressing the gene, even as late as E12.5 (Fig. 6F). Histologi-dington, 1996). Thus, cells in the mCer-1-positive stripe cal sections of an E9.5 embryo revealed that the paraxialmost likely belong to that primitive lineage. stripes spanned the zone of new somite formation (Fig. 6E).

In midstreak embryos, a second site of expression was Two stripes were restricted to the rostral half of the twodetected (Fig. 4A). Positive cells could be seen in a limited most-recently formed somites, while the additional stripeszone on either side of the extended streak. In slightly older were positioned within proximal presomitic mesoderm.embryos (Fig. 4A), this domain expanded and extended more Transverse sections showed that within the cranial aspectinto the anterior half of the embryo. The impression from of a single somitic stripe, expression occurred in all cellsexamining several embryos of slightly different stages was (Fig. 6D).that bilateral populations of mCer-1-positive cells migratefrom the streak and take a broadening lateral path into the

mCer-1 Expression in Adult Tissuesanterior region of the embryo. A lateral migratory path isconsistent with the known route of cells derived from the mCer-1 expression was analysed in adult tissues byanterior streak caudal to the node (Tam et al., 1997). These RNase protection, using an mCer-1-specific probe and acells, most likely head mesendodermal precursors, appeared cyclophilin probe (Hasel and Sutcliff, 1990) to control forto progressively occupy the anterior half of the embryo, RNA recovery and integrity. All adult tissues examineduntil most of this region was positive by late-streak stages were negative, including brain, skeletal muscle, salivary(Fig. 4C). However, expression now no longer extended to gland, tongue, thymus, heart, lung, stomach, spleen, liver,the embryonic/extraembryonic junction: it was set back pancreas, intestine, kidney, bladder, uterus, testes, andmedially and fell away laterally with a relatively sharp ovary (data not shown).boundary (Fig. 4E). The negative region formed the shape ofa crescent reminiscent of the zone that expresses BMP2 (Fig.

mCer-1 Expression in Otx20/0 Embryos4F) and, later, the first cardiac marker Nkx2-5 (Fig. 4G).Expression was also absent from a region at the distal tip Anterior primitive endoderm is known to be essentialcorresponding to the position of the node (Fig. 4D). Histolog- for patterning the anterior epiblast and neural plate

(Acampora et al., 1995; Thomas and Beddington, 1996;ical sections of late-streak embryos revealed expression in


AID DB 8812 / 6x37$$$143 02-09-98 14:51:23 dba


FIG. 3. Comparison of mCer-1 and Xcer activities in Xenopus animal cap assays. (A) Formation of cement glands in individualisedanimal caps (stage 35) after mock injection (a) or injection of mRNAs encoding mCer-1 (b), CFLAG–mCer-1 (c), or Xcer (d). A singledarkly pigmented cement gland is induced by injected cerberus mRNAs in some cases (see B), but not by mock injection. Note in b thesecretion of sticky exudate. (B) Histogram depicting frequency of cement gland induction in animal caps after injection of mRNAs encodingXcer, mCer-1, or CFLAG–mCer-1, compared to uninjected controls. Data were compiled for each mRNA after examination of 100–120injected caps. (C) RT-PCR analysis of markers induced in animal caps at stage 28 after injection of mRNAs encoding Xcer (XCer; lane 1),mCer-1 (MCer; lane 2), and BMP4 (lane 5) or after coinjection of 1:1 Xcer:BMP4 (lane 3) and 1:1 mCer-1:BMP4 (lane 4), compared touninjected control caps (lane 7) and whole stage 25 embryos (lane 6). Nkx2.3 and Nkx2.5 are expressed in cardiac progenitors and anteriorpharyngeal endoderm (Tonissen et al., 1994; Evans et al., 1995). T4 globin is specific to blood, a ventral mesodermal derivative (Banvilleand Williams, 1985). XeHAND marks cardiac and vascular smooth muscle progenitors in lateral mesoderm (D. B. Sparrow, S. Kotecha,N. Towers, and T. J. Mohun, submitted for publication). N-CAM is a panneural marker (Kintner and Melton, 1987). Otx2 is a marker ofanterior tissues, expressed in midbrain, forebrain, placodes, cement gland, and anterior mesoderm (Pannese et al., 1995; Gammill andSive, 1997). Krox20 is expressed in rhombomeres 3 and 5 in the hindbrain (Bradley et al., 1993). CG13 is specific to cement gland (Jamrichand Sato, 1989; Sive et al., 1989). Edd expression is ubiquitous at low levels (data not shown), but enriched in endoderm (Sasai et al.,1996). Equal mRNA input was assessed by expression of EF-1 a (EF1a) (Krieg et al., 1989).

Varlet et al., 1997), as is anterior embryonic mesendod- Acampora et al., 1995; Ang et al., 1996). A possible rolefor Otx2 in the patterning function of primitive endodermerm at later stages (Ang et al., 1994; Blitz and Cho, 1995;

Pannese et al., 1995). Since mCer-1 is expressed in both was suggested by examination of null embryos carryinga LacZ reporter gene, which showed that while Otx2of these zones, it could participate in the patterning func-

tion, either as an inducer or as an inhibitor. Another gene could be expressed in primitive endoderm in the mutantcontext, it could not be induced and/or maintained inthat may be involved is Otx2, which encodes a homeodo-

main protein related to Drosophila orthodenticle and anterior neural plate (Acampora et al., 1995).We examined mCer-1 expression in Otx20/0 embryos byempty spiracles, both involved in head development in

the fly (Simeone et al., 1992). Otx2 is initially expressed whole-mount in situ hybridisation (Fig. 7) and found that itwas still expressed in an anterior region, although to varyingthroughout the primitive endoderm and epiblast before

gastrulation, then becomes restricted during gastrulation extents. In more severely affected embryos (see Ang et al.,1996), expression was restricted to the distal tip and wasto anterior tissues including chordamesoderm, forebrain,

and midbrain (Ang et al., 1994; Simeone et al., 1995; absent at the embryonic/extraembryonic junction, markedby a prominent constriction. As noted above, mCer-1 wouldAcampora et al., 1995). Targeted Otx20/0 mutants show

severe abnormalities at gastrulation, including defective normally extend up to this point. In the less affected Otx20/0

mutant examined (Fig. 7), mCer-1 expression tended moreformation of chordamesoderm and loss of all brain com-partments anterior to rhombomere 3 (Matsuo et al., 1995; towards the normal pattern, although, since anterior pat-


AID DB 8812 / 6x37$$$143 02-09-98 14:51:23 dba

142 Biben et al.

FIG. 4. Expression of mCer-1 during gastrulation. (A–C, H, and I) Lateral views of embryos in which anterior is to the left, posterior tothe right; the arrowhead represents the embryonic/extraembryonic junction. (A) Three early primitive streak-stage embryos showingmCer-1 expression in a midline stripe on the anterior side (left embryo), then progressively (middle and right embryos) in migratingmesendodermal wings arising in the anterior region of the streak. (B) Early streak embryo hybridised with probes for both mCer-1 andbrachyury. Nascent mesodermal wings positive for brachyury are seen on the posterior side (Herrmann, 1991), while the mCer-1 stripeis seen anteriorly. At that stage, most or all of the anterior endoderm is of the primitive lineage (see text). (C) Late primitive streak-stageembryo hybridised with an mCer-1 probe. Expression is confined to anterior mesendoderm. (D) Distal view of the same embryo as in C,showing lack of mCer-1 expression in the region of the node. (E) Anterior view of the same embryo as in C, showing mCer-1 expressionset back from the embryonic/extraembryonic junction and absent from the cardiac progenitor region (bracket). (F) Early neurula embryo(anterior view) hybridised with a BMP2A probe. Expression occurs in the domain of the cardiac progenitors (bracket), but also somewhatinto the extraembryonic domain. (G) Neurula embryo (anterior view) hybridised with Nkx2-5, specific to the cardiac progenitors at thatstage (bracket). (H) Sagittal section through a late-streak embryo highlighting expression of mCer-1 in anterior mesendoderm. The arrow-head indicates the position of the node, where expression is absent. (I) Enlargement of a section adjacent to the one shown in H, depictingmCer-1 expression in both endoderm and associated mesoderm. Note that anterior mesendoderm lacks expression. (J) Headfold-stageembryo (anterior view) showing mCer-1 expression in a wedge of anterior mesendoderm. (K) Transverse section through the embryo in Jshowing mCer-1 expression in anterior mesendoderm underlying the neural plate and excluded from the more lateral cardiogenic region(bracket). Abbreviations: en, endoderm; ep, epiblast; me, mesoderm; np, neural plate.


AID DB 8812 / 6x37$$8812 02-09-98 14:51:23 dba


ter et al., 1996). The degree of identity seen between thetwo proteins (26%) is modest when compared to that seenbetween species cognates of other developmentally im-portant cytokines such as BMPs and activins and indeedtheir inhibitors noggin, chordin, and follistatin. On se-quence comparisons alone, then, mCer-1 and Xcer wouldappear to be family members, rather than cognates. How-ever, two lines of evidence suggest that they are moreclosely related. First, both show anti-BMP-like activitywhen expressed in Xenopus animal caps. mCer-1 is some-what less potent than Xcer, perhaps because of species dif-ferences. There may be a family of proteins related tomCer-1 in mammals, since we have identified another puta-tive cystine knot factor from mouse which, although moredistantly related to Xcer and expressed in a different pattern,has anti-BMP-like activity in animal caps (unpublisheddata). More compelling, therefore, is the observation thatmCer-1 and Xcer are expressed in remarkably similar pat-terns during embryogenesis (see below). While this does notpreclude the existence of more closely related Xcer homo-logues in mammals, it strongly suggests that mCer-1 andXcer perform similar functions at similar times and placesin development, as discussed in the following sections.

A Possible Role for Xcer and mCer-1 in NeuralInduction

The activity of mCer-1 and Xcer in animal caps, manifestas induction of anterior CNS, cement gland, and foregut, isidentical to that elicited by a range of treatments which inhibitBMP signalling (Hemmati-Brivanlou and Melton, 1997; Tsu-neizumi et al., 1997; Sasai and De Robertis, 1997). BMPs actas morphogens in D/V axis specification and are strong ven-tralising agents in ectoderm and mesoderm (Dosch et al., 1997;Wilson et al., 1997). Conversely, anti-BMPs are dorsalising fac-FIG. 5. mCer-1 expression in anterior axial mesoderm. Transversetors. The emerging view is that the most-dorsal state in eachhistological sections of an ÇE7.75 embryo hybridised in whole

mount with an mCer-1 probe at the level of (A) prechordal meso- competent germ layer (anterior CNS for ectoderm, notochordderm, (B) notochordal precursors, and (C) node. Expression is seen for mesoderm) is established by default once BMP signallingin prechordal and notochordal plates, but is absent from the node. is inhibited (Hemmati-Brivanlou and Melton, 1997; Sasai andArrows delimit the apparent extent of these structures. The node De Robertis, 1997). Xcer has characteristics that suggest a roleis clearly recognised by the recessed nature of its innermost cells in realising the default state in ectoderm and perhaps also(the ‘‘pit’’) (Bellomo et al., 1996). Notochordal and prechordal plate mesoderm. The Xcer expression domain in mesendoderm ex-cells have a morphology distinct from surrounding endoderm, con- tends anteriorly and laterally from the organiser (Bouwmeestersistent with their having a smaller surface area ventrally when

et al., 1996), underlying the most anterior reaches of the neuralviewed by scanning electron microscopy (Sulik et al., 1994). Pre-plate (Keller, 1975). Furthermore, the panneural marker N-chordal plate is wider than notochordal precursors, forming a wedgeCAM and the anterior marker Otx2, expressed in forebrain,shape (Sulik et al., 1994; Dale et al., 1997). Note mCer-1 expressionmidbrain, and cement gland and required for anterior CNSin head mesoderm.patterning (Ang et al., 1994; Acampora et al., 1995; Matsuo etal., 1995; Blitz and Cho, 1995; Pannese et al., 1995; Ang etal., 1996), are induced by Xcer (and mCer-1) in animal caps(Bouwmeester et al., 1996; this paper). Extrapolating theseterning is still disturbed in these embryos, it is difficult tofindings to the whole embryo, Xcer may signal inductively tosay to what extent.neurectoderm from underlying mesendoderm, activating theneural pathway and Otx2 (Ruiz i Altaba, 1993).

DISCUSSION Blocking BMP and Wnt Signaling Leads to HeadInduction

We report the identification of murine mCer-1, encoding One problem with the idea that Xcer can induce ante-rior CNS and head development as an anti-BMP is thata secreted protein related to Xenopus cerberus (Bouwmees-


AID DB 8812 / 6x37$$$143 02-09-98 14:51:23 dba

144 Biben et al.

FIG. 6. Expression of mCer-1 in paraxial mesoderm. (A) Early headfold stage embryo showing the mCer-1 anterior expression domain(left; see Fig. 4), now faded almost completely, and the first appearance of two stripes within paraxial mesoderm. (B) Ventroanterior viewof a late headfold stage embryo showing absence of mCer-1 expression in anterior mesendoderm, but stronger expression in paraxialmesoderm. (C) Dorsal view of the tail region of a E9.5 embryo (anterior to the left) showing mCer-1 expression in four stripes in paraxialmesoderm. (D) Transverse section through an E8.5 embryo showing that most or all cells with a paraxial stripe express mCer-1. (E) Paraxialsection through the tail of an E9.5 embryo (anterior to the left) showing all four mCer-1 stripes in paraxial mesoderm. The weakeranteriormost stripes mark the rostral region of the most recently formed somites, while the stronger posterior stripes are within thepresomitic mesoderm. Definitive somite boundaries are indicated by solid arrows. The open arrow indicates the poorly condensed boundarybetween the forming somite and adjacent presomitic mesoderm. (F) E12.5 embryo showing mCer-1 expression only within nascent andnewly formed somites within the tail. Abbreviations: nt, neural tube; s, somites.

other anti-BMPs generally induce a truncated axis when Xwnt-8 is absent (Christian et al., 1991). Thus, the anti-BMP and anti-Wnt activities of Xcer may be necessary andexpressed in whole embryos (Lemaire and Kodjabachian,

1996; Hemmati-Brivanlou and Melton, 1997; Sasai and sufficient for head and anterior CNS induction, althoughXcer presumably acts in concert with the other anti-BMPsDe Robertis, 1997). However, simultaneously blocking

the BMP and Wnt signalling pathways in Xenopus em- or anti-Wnts expressed in the anterior region (Smith andHarland, 1992; Hemmati-Brivanlou et al., 1994; Sasai etbryos leads to induction of a secondary body axis includ-

ing cyclopic heads (Glinka et al., 1997). Furthermore, Xcer al., 1994; Leyns et al., 1997; Wang et al., 1997). SinceWnts are required for trunk muscle formation (Hopplerwas shown to be a potent Wnt inhibitor. This supports

previous observations that Wnts antagonise organiser and et al., 1996; Leyns et al., 1997), the anti-Wnt activity ofXcer may explain why trunk mesoderm is inhibited whenhead development (Christian and Moon, 1993; Popperl et

al., 1995) and that Frzb, an anti-Wnt, is expressed in the Xcer is misexpressed, a property not shown by anti-BMPs(Bouwmeester et al., 1996).organiser (Wang et al., 1997; Leyns et al., 1997), while


AID DB 8812 / 6x37$$$143 02-09-98 14:51:23 dba


FIG. 7. mCer-1 expression in Otx20/0 embryos. Three Otx20/0 embryos harvested at E6.7 (embryo on the left) and E7.5 are shown. Inthe two most-affected (and younger) embryos at the left and centre (see Ang et al., 1996), mCer-1 expression (arrows) is seen on theanterior side, but only distally, while in the less-affected embryo (right), it extends more towards the embryonic/extraembryonic junction,as in normal embryos (see Fig. 4). Abbreviations: A, anterior; em, embryonic region; ex, extraembryonic region; P, posterior.

mCer-1 Is Expressed in the Putative Mouse Head Both Otx2 and Lim-1 are expressed in primitive endodermand have been implicated in specification and/or patterningOrganizerof anterior CNS (Taira et al., 1994; Ang et al., 1994; Acamp-

The expression pattern of mCer-1 is consistent with it ora et al., 1995; Blitz and Cho, 1995; Pannese et al., 1995;having a role similar to that of Xcer in anterior CNS and Varlet et al., 1997). The majority of Otx20/0 and Lim-10/0

head induction. mCer-1 was first expressed in an anterior mutant embryos are abnormal at gastrulation, lacking ante-stripe of primitive endoderm in the early gastrula, on the rior structures, but some Otx2//0 and Lim-10/0 embryosside opposite to the primitive streak. A/P polarity in the survive to birth and remarkably have a normal body but noembryo is determined soon after implantation, as shown head (Matsuo et al., 1995; Shawlot and Behringer, 1995).by expression of the VE-1 epitope in an anterior stripe of Recent experiments show that Otx2 is essential in primi-primitive endoderm (Gardner et al., 1992; Rosenquist and tive endoderm for its anterior-inducing function (AcamporaMartin, 1995). Subsequently, the homeobox genes Hesx1 et al., 1995; S-L. Ang, unpublished results). Together withand Hex (Bedford et al., 1993; Thomas and Beddington, the results of mouse node grafts (see Introduction), these1996; R. Beddington, pers. comm.), and mCer-1 itself, be- experiments suggest that head organiser activity in mousecome expressed in a similar region at or before streak forma- initially lies within the anterior region of primitive endo-tion. The Hesx1 pattern is more restricted than that of derm, geographically separate from the node (Thomas andVE-1 and mCer-1 (Thomas and Beddington, 1996), showing Beddington, 1996).that anterior primitive endoderm has regional character at The fact that Xcer can induce Otx2 expression in Xenopusthat time. animal caps suggests a hierarchical relationship between

Recent experiments suggest that the physical and genetic the two genes. However, a simple linear pathway is notintegrity of anterior primitive endoderm is required for ante- consistent with the complex and dynamic patterns ofrior embryonic patterning (Thomas and Beddington, 1996; mCer-1 and Otx2 in the mouse. In light of this, we exam-Varlet et al., 1997), and the common phenotype of knock- ined Otx2 mutants for mCer-1 expression and found thatouts in two transcription factor genes, Otx2 and Lim-1, it could still be expressed, although the extent of the patternfurther supports this idea (Matsuo et al., 1995; Shawlot and was variable and appeared to depend upon the severity of

the mutant phenotype (Matsuo et al., 1995; Acampora etBehringer, 1995; Acampora et al., 1995; Ang et al., 1996).


AID DB 8812 / 6x37$$$143 02-09-98 14:51:23 dba

146 Biben et al.

al., 1995; Ang et al., 1996). It is currently not possible to mCer-1 expression seen at the junction between head andcardiac mesoderm suggests that mCer-1 ultimately sets thesay whether there is a direct dependence for mCer-1 expres-

sion on Otx2, moderated by variable phenotypic expressiv- medial limit of cardiac induction in the mouse.ity, or whether Otx2 has some other function within primi-tive endoderm which disrupts regionalised gene expression.

mCer-1 in Nascent SomitesFurther experiments will be needed to address this issueand indeed whether mCer-1 itself mediates some of the The last site of mCer-1 expression was in nascent so-patterning properties of endoderm. mites, although this site was not reported for Xcer in Xeno-

pus (Bouwmeester et al., 1996). mCer-1 transcripts wereseen in the anterior half of the two most-recently formedAnterior Embryonic Mesendoderm and Neuralsomites, as well as in two stripes within presomitic meso-Inductionderm. A number of genes have been described that show A/

A second site of mCer-1 expression during gastrulation P-regionalised patterns within this region (Fig. 8), includingwas anterior embryonic mesendoderm, a tissue shown to some thought to act in the mammalian Delta/Notch path-induce or maintain Otx2 expression in explant experiments way and another belonging to the Eph family of receptors(Ang et al., 1994). Our impression is that the mCer-1-posi- (see legend to Fig. 8). While the patterns are varied andtive mesendodermal cells arise in the anterior part of the complex, they support the previous view that A/P patternprimitive streak and migrate into the anterior part of the is established before somite epithelialisation (Aoyama andembryo. The node itself was mCer-1 negative. Unlike Xcer Asamoto, 1988). Furthermore, they suggest that A/P differ-in Xenopus (Bouwmeester et al., 1996), mCer-1 was also ences are established within presomitic mesoderm, proba-expressed in notochord and prechordal mesoderm progeni- bly in a domain covering two somite equivalents. Targetedtors at E7.75. The discrepancy may relate to differences in mutations in presenilin 1 and delta-homologue DII1 displaythe timing of axial development in the two species. Through severe disturbances in somite A/P patterning and epitheli-its two modes, mCer-1 expression eventually comes to un- alisation, providing compelling evidence that the observedderlie most of the anterior neural plate, consistent with a patterns are causally related to this process (Wong et al.,role in neural induction and/or patterning and activation of 1997; Hrabe de Angelis et al., 1997). Segmental expressionOtx2 (Ruiz i Altaba, 1993; Ang et al., 1994). Whether primi- of mCer-1 in nascent and newly formed somites could betive endoderm and anterior embryonic mesendoderm have involved in their intrinsic patterning or, alternatively, indistinct or synergistic roles in anterior development is an providing positional cues to the spinal chord or neural crestinteresting question. for guiding their morphogenesis.

In summary, mCer-1 is expressed in the anterior regionof the gastrula and somites. The similarities in structure,mCer-1 and Cardiac Developmentinductive activity, and expression pattern between mCer-1

mCer-1 expression is absent in cardiac precursors at late and Xcer suggest a common role in development. Inactiva-streak stages. This would seem necessary, since cardiac in- tion of mCer-1 by gene targeting will help to delineate itsduction, at least in the chick, is dependent upon BMP signal- function in head formation in the mouse and the relativeling (Schultheiss et al., 1997). Cardiac development in verte- roles of primitive endoderm and anterior mesendoderm inbrates may also utilise Wnt signalling, since induction of this process.Drosophila heart progenitors, thought to occur throughanalogous genetic pathways (Harvey, 1996), requires theWnt-related gene wingless (Park et al., 1996). Cardiac devel- MATERIALS AND METHODSopment presents an interesting paradox with respect to or-ganiser function. Heart progenitors map to a position just

Cloning of mCer-1 cDNA Fragmentslateral to the organiser and depend on its function (Saterand Jacobson, 1990), yet have ventral character in the sense To isolate extension fragments of GenBank EST AA120122, PCRthat they are induced by BMPs (Schultheiss et al., 1997). was performed on an E7.5 embryonic region cDNA library in pSport

vector (Harrison et al., 1995) using nested primers specific for theThe paradox is seen clearly in embryos treated with theEST (R1, 5*cttggaagattctggaagaaacctg3*; R2, 5*caggactgtgcccttcaa-dorsalising agent LiCl, in which large hearts form in theccagac3*; R3, 5*gccccttctccgggaaaacgaatg3*; and R4, 5*cggtctcag-centre of radially symmetrical head-like structures (Gerhartgtttcttccagaatc3*) and vector (V1, 5*ggaaacagctatgaccatgattac3*;et al., 1991). The organiser must therefore retain the capac-V2, 5*ctgcaggtaccggtccggaattcc3*; V3, 5*cgccagggttttcccagtcacgac3*;ity to sequester a zone in its immediate vicinity permissiveand V4, 5*gacgtcgcatgcacgcgtacgtaag3*) in the following combina-for BMP signalling and heart induction. In addition to BMPs,tions: 5* extension: R3 1 V1, followed by R4 1 V2; 3* extension:

cardiac induction in chicken also requires an unknown fac- R1 1 V3, followed by R2 1 V4. After subcloning and sequencingtor expressed only in anterior endoderm (Schultheiss et al., (ABI automated sequencer), additional PCR was performed on a1997). While it is not clear how these regionalisations occur, different E7.5 cDNA library (gift from Dr. Anne Flenniken, NIMR,one would predict that the cardiogenic region must be free London) to amplify fragments for insertion into vectors Bluescript

SK(/) and pEFBOS I FLAG, using primers encompassing the pre-of anti-BMPs and anti-Wnts. Indeed, the sharp boundary of


AID DB 8812 / 6x37$$$144 02-09-98 14:51:23 dba


used. Relevant GenBank accession numbers (Figs. 1A and 1B andtext) are Xcer, U64831; rat intestinal mucin, M81920; mouse NDP,X92397; human TGFb2, M19154; Xenopus BMP2, X55031; mouseBMP2a, L25602; Xenopus chordin, L35764; rat chordin, D86581;Xenopus follistatin, S69801; mouse follistatin, Z29532; Xenopusnoggin, M98807; mouse noggin, U79163; Xenopus activin bB;mouse inhibin bB.

mCer-1 Protein Expression and Purification

pEFBOS I FLAG vector carrying the mCer-1 insert was cotrans-fected with a puromycin resistance plasmid into CHO cells usingLipofectamine (Gibco BRL) according to the manufacturer’s in-structions. Following puromycin selection (Sigma; 25 mg/ml), re-sistant colonies were picked and expanded, and culture supernatantwas assayed for CFLAG-mCer-1 by analysis of binding to immobi-lised M2 (anti-FLAG) antibody (Kodak Eastman) on a Biosensor2000 (Pharmacia). CFLAG–mCer-1 was purified from conditionedmedium of clone CL47 using M2 affinity resin (Kodak Eastman),

FIG. 8. Summary of expression patterns of mouse genes known eluting with FLAG peptide (60 mg/ml; Kodak Eastman) in Tris-to be segmentally restricted in newly formed somites and/or pre- buffered saline. Fractions were monitored by Western blottingsomitic mesoderm. Anterior is to the left. Numbered circles repre- probed with M2 antibody. Pooled fractions were further purifiedsent fully segmented somites (I–III). The budding somite (not de- by reversed phase–HPLC using a Brownlee C8 column (100 1 2.1marcated) is within the proximal presomitic mesoderm, shown as mm i.d.; Perkin–Elmer, U.S.A.) on a SMART system (Pharmacia)a block to the right. Expression patterns, as determined by in situ using a flow rate of 100 ml/min and a linear gradient between 0.15%hybridisation, were interpreted from the following sources: (v/v) trifluoroacetic acid (TFA) and 0.13% (v/v) TFA containingmCer-1 (this study), Notch2 (Swiatek et al., 1994; Saga et al., 1997), 60% (v/v) n-propanol. Fractions were monitored on the BiosensorSek-1 (Angela Nieto et al., 1992), DII1 (Dunwoodie et al., 1997), and by SDS–PAGE. For N-glycosylation analysis, purified proteinFGFR-1 (Yamaguchi et al., 1992), Notch1 (Reaume et al., 1992; (1 mg) was treated with N-glycosidase F (Boehringer), as per theSwiatek et al., 1994; Conlon et al., 1995), Paraxis (Burgess et al., manufacturer’s instructions, before Western blotting.1995), presenilin (Wong et al., 1997), DII3 (Dunwoodie et al., 1997),and MesP2 (Saga et al., 1997). The exact position of the presomitic/somitic boundary and extent of expression within presomitic meso- Xenopus laevis Embryo Proceduresderm was in some cases difficult to judge from published accounts.

Embryo production, staging, microinjection, and manipulationInterpretations different from the ones shown have been given forwere as previously described (Chambers et al., 1992). SyntheticDII1 (Bettenhausen et al., 1995; Saga et al., 1997) as well as Notch1,mRNA encoding the mouse or Xenopus cerberus genes (0.5–1.0Notch2, and FGFR-1 (Saga et al., 1997).ng) was injected into the animal pole of fertilised eggs. mCer-1 andCFLAG–mCer-1 templates were generated by subcloning codingfragments (see above) into the transcription vector pT7TS (Cleaveret al., 1996). Xcer template was a PCR-generated coding fragmentdicted initiation and termination codons (R6, 5*ccatctgtgaatctaacc-cloned into pCS2 (constructed and provided by Mike Jones, NIMR,tcagtctc3*; R7, 5*agctggcgcgcctgcatctcctcttagttcagctg3*; R8, 5*aac-London). Animal caps were cut at stage 8–9 and cultured untiltcacataacatttccagattg3*; and R9, 5*agctacgcgttgggtttgtttttgaagctg3*)siblings reached stage 28–35. RNA harvested from 25–30 capsin the following combinations: for Bluescript SK(/), R6 1 R8, and(Chambers et al., 1992) was treated with DNase I and one-tenthfor pEFBOS I FLAG, R7 1 R9. pEFBOS I FLAG is a derivative ofwas reverse-transcribed using MMLV reverse transcriptase at 427CpEFBOS (Mizushima and Nagata, 1990) containing DNA sequencesfor 30 min. PCR cycle times using Taq polymerase (Perkin–Elmer;encoding a FLAG epitope (DYKDDDDDK) inserted 3* of the multi-1.25 u) and primers at 0.7 mM were as follows: 937C for 3 min, thenple cloning site. Sixteen clones were sequenced and a consensus25–27 cycles of 937C for 30 s, 657C for 1 min, and 727C for 30 s.was established. One clone from each group which contained theTwenty-seven cycles were used for all primer pairs except EF1 aconsensus only was used for subsequent experiments. PCRs wereat 25 cycles. Reactions performed under these conditions yieldedperformed with Taq polymerase (Gibco BRL) using 2.5 mM MgCl2product in approximately linear relationship to template when var-and 200 mM dNTPs in the buffer supplied. Cycle conditions wereied across a 10–20 fold range. Control reactions which lacked re-301 947C for 10 s, 607C for 20 s, and 727C for 60 s.verse transcriptase yielded no product. 32P-labelled PCR productswere resolved on native 5% polyacrylamide gels and exposed for 1h to overnight. Primer pairs were as follows: T4 globin, 5*agctgccaa-Amino Acid Sequence Comparisons andgcacatcgatgac3* and 5*gtgagctgtccttgctgatacc3*; XNkx2.5, 5*gagcta-Alignmentscagttgggtgtgtgtggt3* and 5*gtgaagcgactcggtatgtgttca3*; XNkx2.3, 5*aaa-cgagccactgaagtgag3* and 5*cacctttgctgcatgtatcc3*; XeHand, 5*acccct-Sequence comparisons and alignments were performed using the

Clustal method contained within the DNA* Megalign program gattgttctatttgcc3* and 5*gaatacaattctggctcattgc3*; N-CAM, 5*cacagt-tccaccaaatgc3* and 5*ggaatcaagcggtacaga3*; Otx2, 5*ggatggatttgttgc-using default conditions. Where alignments recorded similarities

as well as identities, the default setting (chemical similarities) was accagt3* and 5*cactctccgagctcacttctc3*; Krox20, 5*aaccgccccagtaag-


AID DB 8812 / 6x37$$$144 02-09-98 14:51:23 dba

148 Biben et al.

acc3* and 5*gtgtcagcctgtcctgttag3*; CG13, 5*agttgttgattttgtgaaaacc3* (1993). HEX: A novel homeobox gene expressed during haemato-poiesis and conserved between mouse and human. Nucleic Acidsand 5*cttcttctaaatcaacaacagg3*; EF1 a, 5*cagattggtgctggatatgc3* and

5*actgccttgatgactcctag3*; and Edd, 5*tattctgactcctgaaggtg3* and 5*gag- Res. 21, 1245–1249.aactgcccatgtgcctc3*. Bellomo, D., Lander, A., Harragan, I., and Brown, N. A. (1996). Cell

proliferation in mammalian gastrulation: The ventral node andnotochord are relatively quiescent. Dev. Dyn. 205, 471–485.

In Situ Hybridisation Bettenhausen, B., Hrabe de Angelis, M., Simon, D., Guenet, J.-L.,and Gossler, A. (1995). Transient and restricted expression during

Embryos were dissected from timed matings of mixed C57Bl/6mouse embryogenesis of DII1, a murine gene closely related to1 C57Bl/10 background. Midday of the day of plugging was takenDrosophila Delta. Development 121, 2407–2418.

as E0.5. Whole-mount in situ hybridisation was performed as pre-Biben, C., and Harvey, R. P. (1997). Homeodomain factor Nkx2-5

viously described (Biben and Harvey, 1997). Probe template forcontrols left–right asymmetric expression of bHLH eHand dur-

mCer-1 was generated by PCR using primers R1 1 R3 (see above)ing murine heart development. Genes Dev. 11, 1357–1369.

subcloned into Bluescript SK(/). Templates for the brachyury (giftBlitz, I. C., and Cho, K. W. Y. (1995). Anterior neurectoderm is pro-from Bernard Herrmann, Tubingen) and Nkx2-5 probes were as

gressively induced during gastrulation: The role of the Xenopusdescribed (Herrmann, 1991; Lints et al., 1993). BMP2a templatehomeobox gene orthodenticle. Development 121, 993–1004.was the 1-kb XmaIII/EcoRI fragment of the mouse cDNA (gift from

Bouwmeester, T., Kim, S.-H., Sasai, Y., Lu, B., and De Robertis,Rosa Beddington, London) subcloned into the EcoRI/NarI sites ofE. M. (1996). Cerberus is a head-inducing secreted factor ex-Bluescript SK(/).pressed in the anterior endoderm of Spemann’s organizer. Nature382, 596–601.

Bradley, L. C., Snape, A., Bhatt, S., and Wilkinson, D. G. (1993). TheACKNOWLEDGMENTS structure and expression of the Xenopus Krox-20 gene: Conserved

and divergent patterns of expression in rhombomeres and neuralcrest. Mech. Dev. 40, 73–84.We thank Rosa Beddington, Bernard Herrmann, and Mike Jones

for supplying plasmids; Steven Mihajlovic for histology; Tanya Hat- Brannon, M., and Kimelman, D. (1996). Activation of Siamois byzistavrou for technical assistance; and Zdenka Bolcevic for admin- the Wnt pathway. Dev. Biol. 180, 344–347.istration. This work was supported by the National Health and Burgess, R., Cserjesi, P., Ligon, K. L., and Olson, E. N. (1995). Par-Medical Research Council of Australia (NH&MRC), the National axis: A basic helix–loop–helix protein expressed in paraxialHeart Foundation of Australia, and the Human Frontiers Science mesoderm and developing somites. Dev. Biol. 168, 296–306.Program. E.S. holds an NH&MRC C. J. Martin Fellowship. C-C. Carnac, G., Kodjabachian, L., Gurdon, J. B., and Lemaire, P. (1996).Wang holds an Australian Development Cooperation Scholarship. The homeobox gene Siamois is a target of the Wnt dorsalisation

pathway and triggers organiser activity in the absence of meso-derm. Development 122, 3055–3065.

Chambers, A. E., Kotecha, S., Towers, N., and Mohun, T. J. (1992).REFERENCESMuscle-specific expression of SRF-related genes in the early em-bryos of Xenopus laevis. Development 118, 1095–1106.Acampora, D., Mazan, S., Lallemant, Y., Avantaggio, V., Maury,

Christian, J. L., McMahon, G. A., McMahon, A. P., and Moon, R. T.M., Simeone, A., and Brulet, P. (1995). Forebrain and midbrain(1991). Xwnt-8, a Xenopus Wnt-1/int-1-related gene responsiveregions are deleted in Otx20/0 mutants due to a defective ante-to mesoderm-inducing growth factors, may play a role in ventralrior neurectoderm specification during gastrulation. Develop-mesodermal patterning during embryogenesis. Developmentment 121, 3279–3290.111, 1045–1055.Ang, S.-L., Conlon, R. A., Jin, O., and Rossant, J. (1994). Positive

Christian, J. L., and Moon, R. T. (1993). Interactions between Xwnt-and negative signals from mesoderm regulate the expression of8 and Spemann organizer signaling pathways generate dorsoven-mouse Otx2 in ectoderm explants. Development 120, 2979–tral pattern in the embryonic mesoderm of Xenopus. Genes Dev.2989.7, 13–28.Ang, S. L., Jin, O., Rhinn, M., Daigle, N., Stevenson, L., and Ros-

Cleaver, O. B., Patterson, K. D., and Krieg, P. A. (1996). Overex-sant, J. (1996). A targeted mouse Otx2 mutation leads to severepression of the tinman-related genes XNkx-2.5 and XNkx-2.3 indefects in gastrulation and formation of axial mesoderm and toXenopus embryos results in myocardial hyperplasia. Develop-deletion of rostral brain. Development 122, 243–252.ment 122, 3549–3556.Angela Nieto, M., Gilardi-Hebenstreit, P., Charnay, P., and Wilkin-

Conlon, R. A., Reaume, A. G., and Rossant, J. (1995). Notch1 isson, D. G. (1992). A receptor protein tyrosine kinase implicatedrequired for the coordinate segmentation of somites. Develop-in the segmental patterning of the hindbrain and mesoderm. De-ment 121, 1533–1545.velopment 116, 1137–1150.

Dale, J. K., Vesque, C., Lints, T. J., Kuber Sampath, T., Furley, A.,Aoyama, H., and Asamoto, K. (1988). Determination of somiteDodd, J., and Placzek, M. (1997). Cooperation of BMP7 and SHHcells: Independence of cell differentiation and morphogenesis.in the induction of forebrain ventral midline cells by prechordalDevelopment 104, 15–28.mesoderm. Cell 90, 257–269.Banville, D., and Williams, J. G. (1985). The pattern of expression

Dale, L., Howes, G., Price, B. M., and Smith, J. C. (1992). Boneof the Xenopus laevis tadpole alpha-globin genes and the aminomorphogenetic protein 4: A ventralizing factor in early Xenopusacid sequence of the three major tadpole alpha-globin polypep-development. Development 115, 573–585.tides. Nucleic Acids Res. 13, 5407–5421.

Dosch, R., Gawantka, V., Delius, H., Blumenstock, C., and Niehrs,Beddington, R. (1994). Induction of a second neural axis by theC. (1997). Bmp-4 acts as a morphogen in dorsoventral mesodermmouse node. Development 120, 613–620.

Bedford, F. K., Ashworth, A., Enver, T., and Wiedemann, L. M. patterning in Xenopus. Development 124, 2325–2334.


AID DB 8812 / 6x37$$$144 02-09-98 14:51:23 dba


Dunwoodie, S. L., Henrique, D., Harrison, S. M., and Beddington, Isaacs, N. W. (1995). Cystine knots. Curr. Opinion Struct. Biol. 5,R. S. P. (1997). Mouse DII3: A novel divergent Delta gene which 391–395.may complement the function of other Delta homologues during Jamrich, M., and Sato, S. (1989). Differential gene expression in theearly pattern formation in the mouse embryo. Development 124, anterior neural plate during differentiation. Development 105,3065–3076. 779–786.

Echelard, Y., Epstein, D. J., St-Jacques, B., Shen, L., Mohler, J., Jones, C. M., Lyons, K. M., Lapan, P. M., Wright, C. V., and Hogan,McMahon, J. A., and McMahon, A. P. (1993). Sonic hedgehog, a B. L. (1992). DVR-4 (bone morphogenetic protein-4) as a poste-member of a family of putative signaling molecules, is implicated rior–ventralizing factor in Xenopus mesoderm development. De-in the regulation of CNS polarity. Cell 75, 1417–1430. velopment 115, 639–647.

Evans, S., Yan, W., Murillo, M. P., Ponce, J., and Papalopulu, N. Keller, R. (1975). Vital dye mapping of the gastrula and neurula of(1995). tinman, a Drosophila homeobox gene required for heart the superficial layer. Dev. Biol. 42, 222–241.and visceral mesoderm specification, may be represented by a Kintner, C. N., and Dodd, J. (1991). Hensen’s node induces neuralfamily of genes in vertebrates: XNkx-2.3, a second homologue of tissue in Xenopus ectoderm: Implications for the action of thetinman. Development 121, 3889–3899. organizer in neural induction. Development 113, 1495–1505.

Fagotto, F., Guger, K., and Gumbiner, B. M. (1997). Induction of Kintner, C. R., and Melton, D. A. (1987). Expression of Xenopus N-the primary dorsalizing centre in Xenopus by the Wnt/Gsk/beta CAM RNA in ectoderm is an early response to neural induction.catenin signaling pathway, but not by Vg1, Activin or Noggin. Development 99, 311–325.Development 124, 453–460. Krieg, P. A., Varnum, S. M., Wormington, W. M., and Melton, D. A.

Fainsod, A., Deissler, K., Yelin, R., Marom, K., Epstein, M., Pil- (1989). The mRNA encoding elongation factor 1-alpha (EF-1lemer, G., Steinbeisser, H., and Blum, M. (1997). The dorsalizing alpha) is a major transcript at the mid-blastula transition in Xeno-and neural inducing gene follistatin is an antagonist of BMP-4. pus. Dev. Biol. 133, 93–100.Mech. Dev. 63, 39–50.

Lawson, K. A., and Pedersen, R. A. (1987). Cell fate, morphogeneticGammill, L. S., and Sive, H. (1997). Identification of otx2 target

movement and population kinetics of embryonic endoderm atgenes and restrictions in ectodermal competence during Xenopus

the time of germ layer formation in the mouse. Developmentcement gland formation. Development 124, 471–481.

101, 627–652.Gardner, R. L., Meredith, M. R., and Altman, D. G. (1992). Is the

Lemaire, P., and Kodjabachian, L. (1996). The vertebrate organizer:anterior–posterior axis of the fetus specified before implantationStructure and molecules. Trends Genet. 12, 525–531.in the mouse? J. Exp. Zool. 264, 437–443.

Leyns, L., Bouwmeester, T., Kim, S.-H., Piccolo, S., and De Rob-Gerhart, J., Doniach, T., and Stewart, R. (1991). Organizing theertis, E. M. (1997). Frzb-1 is a secreted antagonist of Wnt signalingXenopus organizer. In ‘‘Gastrulation’’ (R. Keller, Ed.), pp. 57–77.expressed in the Spemann organizer. Cell 88, 747–756.Plenum, New York.

Lints, T. J., Parsons, L. M., Hartley, L., Lyons, I., and Harvey, R. P.Glinka, A., Wu, W., Onichtchouk, D., Blumenstock, C., and Niehrs,(1993). Nkx-2.5: A novel murine homeobox gene expressed inC. (1997). Head induction by simultaneous repression of Bmpearly heart progenitor cells and their myogenic descendants. De-and Wnt signalling in Xenopus. Nature 389, 517–519.velopment 119, 419–431.Harrison, S. M., Dunwoodie, S. L., Arkell, R. M., Lehrach, H., and

Logan, M., and Mohun, T. (1993). Induction of cardiac muscle dif-Beddington, R. S. (1995). Isolation of novel tissue-specific genesferentiation in isolated animal pole explants of Xenopus laevisfrom cDNA libraries representing the individual tissue constit-embryos. Development 118, 865–875.uents of the gastrulating mouse embryo. Development 121,

Matsuo, I., Kuratani, S., Kimura, C., Takeda, N., and Aizawa, S.2479–2489.(1995). Mouse Otx-2 functions in the formation and patterningHarvey, R. P. (1996). NK-2 homeobox genes and heart development.of rostral head. Genes Dev. 9, 2646–2658.Dev. Biol. 178, 203–216.

McDonald, N. Q., and Hendrickson, W. A. (1993). A structural su-Hasel, K. W., and Sutcliff, J. G. (1990). Nucleotide sequence of aperfamily of growth factors containing a cystine knot motif. CellcDNA coding for mouse cyclophilin. Nucleic Acids Res. 18,73, 421–424.4019.

Meitinger, T., Meindl, A., Bork, P., Rost, B., Sander, C., Haasemann,Hemmati-Brivanlou, A., Kelly, O. G., and Melton, D. A. (1994). Fol-M., and Murken, J. (1993). Molecular modelling of the Norrielistatin, an antagonist of activin, is expressed in the Spemanndisease protein predicts a cystine knot growth factor tertiaryorganizer and displays direct neuralizing activity. Cell 77, 283–structure. Nature Genet. 5, 376–380.295.

Mizushima, S., and Nagata, S. (1990). pEF-BOS, a powerful mamma-Hemmati-Brivanlou, A., and Melton, D. (1997). Vertebrate embry-lian expression vector. Nucleic Acids Res. 18, 5322.onic cells will become nerve cells unless told otherwise. Cell 88,

Pannese, M., Polo, C., Andreazzoli, M., Vignali, R., Kablar, B., Bar-13–17.sacchi, G., and Boncinelli, E. (1995). The Xenopus homologue ofHerrmann, B. (1991). Expression pattern of the Brachyury gene inOtx2 is a maternal homeobox gene that demarcates and specifieswhole-mount TWis/TWis mutant embryos. Development 113,anterior body regions. Development 121, 707–720.913–917.

Park, M., Wu, X., Golden, K., Axelrod, J. D., and Bodmer, R. (1996).Ho, R. (1992). Axis formation in the embryo of the zebrafish, Bra-The wingless pathway is directly involved in Drosophila heartchydanio rerio. Semin. Dev. Biol. 3, 53–64.development. Dev. Biol. 177, 104–116.Hoppler, S., Brown, J. D., and Moon, R. T. (1996). Expression of a

Piccolo, S., Sasai, Y., Lu, B., and De Robertis, E. M. (1996). Dorso-dominant-negative Wnt blocks induction of MyoD in Xenopusventral patterning in Xenopus: Inhibition of ventral signals byembryos. Genes Dev. 10, 2805–2817.direct binding of chordin to BMP-4. Cell 86, 589–598.Hrabe de Angelis, M., McIntyre, J., II, and Gossler, A. (1997). Main-

Popperl, H., Bienz, M., Studer, M., Chan, S.-K., Aparicio, S., Bren-tenance of somite borders in mice requires the Delta homologueDIII. Nature 386, 717–722. ner, S., Mann, R., and Krumlauf, R. (1995). Segmental expression


AID DB 8812 / 6x37$$$145 02-09-98 14:51:23 dba

150 Biben et al.

of Hoxb-1 is controlled by a highly conserved autoregulatory loop an der orientierung und beschaffenheit der induzierten embryo-nalanlage. Wilhelm Roux Arch. Entwicklungsmech. Org. 123,dependent upon exd/pbx. Cell 81, 1031–1042.389–517.Popperl, H., Schmidt, C., Wilson, V., Hume, C. R., Dodd, J., Krum-

lauf, R., and Beddington, R. S. P. (1997). Misexpression of Spemann, H., and Mangold, H. (1924). Uber induction von embryo-nalanlagen durch implantation artfremder organisatoren. Wil-Cwnt8C in the mouse induces an ectopic embryonic axis and

causes truncation of the anterior neurectoderm. Development helm Roux Arch. Entwicklungsmech. Org. 100, 599–638.124, 2997–3005. Storey, K. G., Crossley, J. M., De Robertis, E. M., Norris, W. E., and

Stern, C. (1992). Neural induction and regionalization in theQuinlan, G. A., Williams, E. A., Tan, S.-S., and Tam, P. P. L. (1995).Neuroectodermal fate of epiblast cells in the distal region of the chick embryo. Development 114, 729–741.mouse egg cylinder: Implications for body plan organisation dur- Streit, A., Sockanathan, S., Perez, L., Rex, M., Scotting, P. J., Sharpe,ing early embryogenesis. Development 121, 87–98. P. T., Lovell-Badge, R., and Stern, C. D. (1997). Preventing the

loss of competence for neural induction—HGF/SF, L5 and SOX-Reaume, A. G., Conlon, R. A., Zirngibl, R., Yamaguchi, T. P., andRossant, J. (1992). Expression analysis of a Notch homologue in 2. Development 124, 1191–1202.the mouse embryo. Dev. Biol. 154, 377–387. Sulik, K., Dehart, D. B., Inagaki, T., Carson, J. L., Vrablic, T.,

Gesteland, K., and Schoenwolf, G. C. (1994). Morphogenesis ofRosenquist, T. A., and Martin, G. R. (1995). Visceral endoderm-1(VE-1): An antigen marker that distinguishes anterior from poste- the murine node and notochordal plate. Anat. Rec. 239, 103–

112.rior embryonic visceral endoderm in the early post-implantationmouse embryo. Mech. Dev. 49, 117–121. Swiatek, P. J., Lindsell, C. E., Franco del Amo, F., Weinmaster, G.,

and Gridley, T. (1994). Notch1 is essential for postimplantationRuiz i Altaba, A. (1993). Induction and axial patterning of the neuralplate: Planar and vertical signals. J. Neurobiol. 24, 1276–1304. development in mice. Genes Dev. 8, 707–719.

Taira, M., Otani, H., Saint-Jeannet, J. P., and Dawid, I. B. (1994).Saga, Y., Hata, N., Koseki, H., and Taketo, M. M. (1997). MesP2:A novel mouse gene expressed in the presegmented mesoderm Role of the LIM class homeodomain protein Xlim-1 in neural

and muscle induction by the Spemann organizer in Xenopus.and essential for segmentation initiation. Genes Dev. 11, 1827–1839. Nature 372, 677–679.

Tam, P. P. L., Parameswaran, M., Kinder, S. J., and Weinberger,Sasai, Y., and De Robertis, E. M. (1997). Ectodermal patterning invertebrate development. Dev. Biol. 182, 5–20. R. P. (1997). The allocation of epiblast cells to the embryonic

heart and other mesodermal lineages: The role of ingression andSasai, Y., Lu, B., Piccolo, S., and De Robertis, E. M. (1996). Endo-derm induction by the organizer-secreted factors chordin and tissue movement during gastrulation. Development 124, 1631–noggin in Xenopus animal caps. EMBO J. 15, 4547–4555. 1642.

Sasai, Y., Lu, B., Steinbeisser, H., Geissert, D., Gont, L. K., and Thomas, P., and Beddington, R. S. P. (1996). Anterior primitive en-De Robertis, E. M. (1994). Xenopus chordin: A novel dorsalizing doderm may be responsible for patterning the anterior neuralfactor activated by organizer-specific homeobox genes. Cell 79, plate in the mouse embryo. Curr. Biol. 6, 1487–1496.779–790. Tonissen, K. F., Drysdale, T. A., Lints, T. J., Harvey, R. P., and

Sater, A. K., and Jacobson, A. G. (1990). The role of the dorsal lip Krieg, P. A. (1994). XNkx2.5, a Xenopus gene related to Nkx-2.5in induction of heart mesoderm in Xenopus laevis. Development and tinman: Evidence for a conserved role in cardiac develop-108, 461–470. ment. Dev. Biol. 162, 325–328.

Schultheiss, T. M., Burch, J. B. E., and Lassar, A. B. (1997). A role Tsuneizumi, K., Nakayama, T., Kamoshida, Y., Kornberg, T. B.,for bone morphogenetic proteins in the induction of cardiac myo- Christian, J. L., and Tabata, T. (1997). Daughters against dppgenesis. Genes Dev. 11, 451–462. modulates dpp organizing activity in Drosophila wing develop-

ment. Nature 389, 627–631.Shawlot, W., and Behringer, R. R. (1995). Requirement for Lim1 inhead-organizer function. Nature 374, 425–430. Varlet, I., Collignon, J., and Robertson, E. J. (1997). Nodal expres-

sion in the primitive endoderm is required for specification ofShih, J., and Fraser, S. E. (1996). Characterizing the zebrafish orga-nizer: Microsurgical analysis at the early-shield stage. Develop- the anterior axis during mouse gastrulation. Development 124,

1033–1044.ment 122, 1313–1322.Simeone, A., Acampora, D., Gulisano, M., Stornaiuolo, A., and Bon- Waddington, C. H. (1933). Induction by the primitive streak and

its derivatives in the chick. J. Exp. Biol. 10, 38–46.cinelli, E. (1992). Nested expression domains of four homeoboxgenes in developing brain. Nature 358, 687–690. Wang, S., Krinks, M., Lin, K., Luyten, F. P., and Moos, J. M. (1997).

Frzb, a secreted protein expressed in the Spemann organizer,Simeone, A., Avantaggiato, V., Moroni, M. C., Mavillo, F., Arra,C., Cotelli, R., Nigro, V., and Acampora, D. (1995). Retinoic acid binds and inhibits Wnt-8. Cell 86, 757–766.induces stage-specific antero-posterior transformation of rostral Weinstein, D. C., and Hemmati-Brivanlou, A. (1997). Neural induc-central nervous system. Mech. Dev. 51, 83–98. tion in Xenopus laevis: Evidence for the default model. Curr.

Opinion Neurobiol. 7, 7–12.Sive, H., Hattori, K., and Weintraub, H. (1989). Progressive determi-nation during formation of the anteroposterior axis in Xenopus Wilson, P. A., Lagna, G., Suzuki, A., and Hemmati-Brivanlou, A.laevis. Cell 58, 171–180. (1997). Concentration-dependent patterning of the Xenopus ecto-

derm by BMP4 and its signal transducer Smad1. DevelopmentSmith, S. M., Dickman, E. D., Thompson, R. P., Sinning, A. R.,Wunsch, A. M., and Markwald, R. R. (1997). Retinoic acid directs 124, 3177–3184.cardiac laterality and the expression of early markers of precar- Wong, P. C., Zheng, H., Chen, H., Becher, M. W., Sirinathsinghji,diac asymmetry. Dev. Biol. 182, 162–171. D. J. S., Trumbauer, M. E., Chen, H. Y., Price, D. L., Van der

Ploeg, L. H. T., and Sisodia, S. S. (1997). Presenilin 1 is requiredSmith, W. C., and Harland, R. M. (1992). Expression cloning of nog-gin, a new dorsalizing factor localized to the Spemann organizer for Notch1 and DII1 expression in the paraxial mesoderm. Nature

387, 288–292.in Xenopus embryos. Cell 70, 829–840.Spemann, H. (1931). Uber den anteil von implantat und wirtskeim Wylie, C., Kofron, M., Payne, C., Anderson, R., Hosobuchi, M.,


AID DB 8812 / 6x37$$$145 02-09-98 14:51:23 dba


Joseph, E., and Heasman, J. (1996). Maternal b-catenin establishes Zimmerman, L. B., De Jesus-Escobar, J. M., and Harland, R. M.a ‘‘dorsal signal’’ in early Xenopus embryos. Development 122, (1996). The Spemann organizer signal noggin binds and inacti-2987–2996. vates Bone morphogenetic protein 4. Cell 86, 599–606.

Yamaguchi, T. P., Conlon, R. A., and Rossant, J. (1992). Expressionof the fibroblast growth factor receptor FGFR-1/flg during gastru-

Received for publication October 24, 1997lation and segmentation in the mouse embryo. Dev. Biol. 152,Accepted November 17, 199775–88.


AID DB 8812 / 6x37$$$145 02-09-98 14:51:23 dba

Murine Cerberus Homologue mCer-1: A Candidate Anterior ... · Mouse Cerberus Homologue mCer-1 137...

Documents

Transcript of Murine Cerberus Homologue mCer-1: A Candidate Anterior ... · Mouse Cerberus Homologue mCer-1 137...