Ciliation and gene expression distinguish between node and posterior notochord in the mammalian embryo

r 2006, Copyright the Authors Differentiation (2007) 75:133–146 DOI: 10.1111/j.1432-0436.2006.00124.x Journal compilation r 2006, International Society of Differentiation

O RI G INA L AR T I C L E

Martin Blum . Philipp Andre . Kerstin Muders . Axel Schweickert . Anja Fischer . Eva Bitzer . Susanne Bogusch . Tina Beyer . Henny W. M. van Straaten . Christoph Viebahn

Ciliation and gene expression distinguish between node and posterior notochord in the mammalian embryo

Received June 1, 2006; accepted in revised form August 14, 2006

Abstract The mammalian node, the functional equivalent of the frog dorsal blastoporal lip (Spemann’s organizer), was originally described by Viktor Hensen in 1876 in the rabbit embryo as a mass of cells at the anterior end of the primitive streak. Today, the term ‘‘node’’ is commonly used to describe a bilaminar epithelial groove presenting itself as an indentation or ‘‘pit’’ at the distal tip of the mouse egg cylinder, and cilia on its ventral side are held responsible for molecular laterality (left–right) determination. We find that Hensen’s node in the rabbit is devoid of cilia, and that ciliated cells are restricted to the notochordal plate, which emerges from the node rostrally. In a comparative approach, we use the organizer marker gene Goosecoid (Gsc) to show that a region of densely packed epithelium-like cells at the anterior end of the primitive streak represents the node in mouse and rabbit and is

. )
Philipp Andre
Kerstin Muders
Axel Martin Blum (* Schweickert
Anja Fischer
Eva Bitzer
Susanne Bogusch
Tina Beyer Institute of Zoology University of Hohenheim D-70593 Stuttgart, Germany Tel: 149 7 1145922255 Fax: 149 7 1145923450 E-mail: [email protected] Henny W. M. van Straaten Department of Anatomy and Embryology University of Maastricht 6200 MD Maastricht, The Netherlands Christoph Viebahn Department of Anatomy and Embryology Centre of Anatomy, University of Go¨ttingen D-37075 Go¨ttingen, Germany U.S. Copyright Clearance Center Code Statement:

covered ventrally by a hypoblast (termed ‘‘visceral endoderm’’ in the mouse). Expression of Nodal, a gene intricately involved in the determination of vertebrate laterality, delineates the wide plate-like posterior segment of the notochord in the rabbit and mouse, which in the latter is represented by the indentation frequently termed ‘‘the node.’’ Similarly characteristic ciliation and nodal expression exists in Xenopus neurula embryos in the gastrocoel roof plate (GRP), i.e., at the posterior end of the notochord anterior to the blastoporal lip. Our data suggest that (1) a posterior segment of the notochord, here termed PNC (for posterior notochord), is characterized by features known to be involved in laterality determination, (2) the GRP in Xenopus is equivalent to the mammalian PNC, and (3) the mammalian node as defined by organizer gene expression is devoid of cilia and most likely not directly involved in laterality determination. Key words Hensen’s node
node
organizer
cilia
notochord
notochordal plate
PNC
left–right asymmetry
Nodal
Gsc
mouse
rabbit
Xenopus

Introduction In 1876, Viktor Hensen coined the term ‘‘node’’ to describe a marked thickening at the anterior end of the primitive streak of the 7-day rabbit embryo, which revealed itself by its dark appearance upon fixation of appropriately staged specimens (Hensen, 1876; Viebahn, 2001). According to his original account, the ‘‘node’’ was characterized by a mass of cells in which the germ layers were indistinguishable (Hensen, 1876). It was only after Spemann and Mangold’s (1924)

0301–4681/2007/7502–133 $ 15.00/0

134

ground-breaking work on the primary embryonic organizer in amphibian embryos, and Waddington’s demonstration that the node of chick and rabbit was the equivalent of the amphibian dorsal blastopore lip (Waddington, 1933; Waddington and Schmidt, 1933) that the term node changed its character from a descriptive morphological entity into a functional concept (Joubin and Stern, 1999; Davidson and Tam, 2000; Viebahn, 2001). The extensive molecular and embryological characterization of the node or organizer in recent years, primarily in Xenopus, chick, mouse, and zebrafish embryos, has demonstrated that the organizer function changes over time, confirming the distinction between the early head organizer and the late tail organizer originally noted by Spemann and confirmed by Mangold in Einsteck experiments on amphibian embryos (Spemann, 1931; Mangold, 1933; Schier and Talbot, 1998; Joubin and Stern, 1999; De Robertis et al., 2000; Boettger et al., 2001; Kinder et al., 2001; Niehrs, 2004; De Robertis, 2006). According to the current concept (Joubin and Stern, 1999; Kinder et al., 2001; Niehrs, 2004), the organizer activity resides in an ever-changing complement of cells that provides the cellular source for the notochord and the overlying floor plate (of the neuroectoderm) extending anteriorly from the organizer region and, which, therefore, constantly changes the expression of genes such as gsc, chordin, noggin, and follistatin (Joubin and Stern, 1999; De Robertis et al., 2000; Niehrs, 2004) commonly held responsible for organizer activity. The term ‘‘node’’ is used for mammalian and avian embryos almost exclusively, while—due to their respective modes of gastrulation—the corresponding tissues localize to the dorsal lip of the blastopore in amphibians and to the embryonic shield in teleost fish larvae (Oppenheimer, 1936; Ho, 1992; Viebahn, 2001; Niehrs, 2004). Although the criteria for organizer function have been rather well defined, particularly in model organisms amenable to experimental manipulations such as frog, chick, and rabbit (Joubin and Stern, 1999; Knoetgen et al., 2000; Boettger et al., 2001; Niehrs, 2004), the mouse as the bona fide mammalian model organism has suffered from varying definitions of the organizer, both in morphological and in molecular terms. While initially the entire primitive streak was supposed to correspond to the amphibian dorsal lip, molecular characterizations as well as heterotypic and homotypic organizer transplants have shifted the focus toward the anterior end of the primitive streak (Blum et al., 1992; Beddington, 1994) or to different parts of the streak at different stages (e.g., early-gastrula organizer [EGO] defined by Tam and Steiner, 1999). However, the term ‘‘node’’ in the mouse is now most commonly used to pinpoint an indentation at the distal tip of the egg cylinder, which first appears around 7.5 days post fertilization, and that persists for about 24 hr (Beddington and Robertson,

1999). Originally, this indentation was referred to as ‘‘archenteron’’ (Theiler, 1972), and its continuity with the notochord anteriorly was well documented. However, the term ‘‘archenteron’’ was considered a misnomer and renamed ‘‘node’’ in an attempt to find a corresponding structure to the amphibian dorsal lip and Hensen’s node in chick (Beddington et al., 1992). Subsequently, the homeobox gene goosecoid (Gsc) was described as the first organizer-specific gene in the mouse, and mouse organizer function was demonstrated by heterologous grafts into Xenopus recipient embryos (Blum et al., 1992); later, this was confirmed by homologous grafts at E7.5, and, most convincingly, through the detailed embryological analysis by Beddington, Tam, and colleagues (Blum et al., 1992; Beddington, 1994; Kinder et al., 2001). The term ‘‘node’’ has received additional attention recently with the discovery of a vectorial extracellular fluid flow generated by motile monocilia, which protrude from the ventral epithelial cells at the site of the distal indentation of the 7.75-day mouse egg cylinder (Nonaka et al., 1998; Hamada et al., 2002; Hirokawa et al., 2006; Raya and Belmonte, 2006). This activity, shown to be connected with the generation of laterality in mice, was coined ‘‘Nodal Flow,’’ referring to the ‘‘node’’ as the underlying ciliated structure (Nonaka et al., 1998; Hamada et al., 2002; Nonaka et al., 2002). In an attempt to study this fluid flow in a prototype mammalian embryo, the rabbit, which, contrary to mice, develops via a flat blastodisc stage (Hensen, 1876; Idkowiak et al., 2004), we found no evidence of cilia on cells of Hensen’s node. This finding prompted us to present the following study in which monocilia were visualized by scanning electron microscopy (SEM) and by immunohistochemistry (IHC) on a subset of notochordal plate cells, which— like the depression at the distal tip of the mouse egg cylinder—are organized in a bilaminar epithelium and delineated by bilateral expression of nodal, a gene closely associated with mouse gastrulation and the node, initially (Zhou et al., 1993; Conlon et al., 1994), and later found to be intricately involved in the determination of left–right (LR) asymmetry (Collignon et al., 1996; Lowe et al., 1996). Comparative histological and electron microscopical analysis of mouse and rabbit embryos using Gsc and Nodal as marker genes further demonstrated that the ciliated depression in the mouse commonly referred to as the ‘‘node’’ represents the posterior portion of the notochordal plate, while the mouse node is a distinct entity posterior to the notochordal plate. These results raise the questions whether the mammalian organizer, represented by the node and devoid of freely moving cilia, is involved in the determination of laterality and whether the term ‘‘Nodal Flow’’ is still appropriate to describe ciliadriven fluid flow on the ventral surface of the mammalian gastrula.

135

Methods

SEM

Animals

SEM analysis was performed following published protocols (Sulik et al., 1994). In brief, embryos were dissected and immediately fixed in 2.5% glutaraldehyde in Soerensen’s buffer. Specimens were postfixed in 1% OsO4, critical point dried, sputter coated, and examined using a Zeiss DSM 940A SEM (Oberkochen, Germany).

Rabbit embryos: Rabbits (New Zealand White) were purchased as pregnant females from a commercial breeder (Lammers, Germany) or mated in the animal facility at Hohenheim University. Embryos from timed matings were dissected from uteri in sterile PBS at room temperature (RT) and fixed for SEM, transmission electron microscopy (TEM), or whole-mount in situ hybridization. For staging of the embryos, the system previously generated (Viebahn et al., 2002) for early-gastrula stages (stages 1–3) on the basis of the chick stages by (Hamburger and Hamilton, 1992) was extended to early somite stages as follows (Fig. 1): a Hensen’s node distinct in density from the anterior end of the primitive streak defines stage 4; a notochordal process, defined by a reduced density compared with Hensen’s node, with the same length as the primitive streak defines stage 5; and a notochordal process longer than the primitive streak and a neural plate distinct from surrounding surface ectoderm defines stage 6. Somite stages are defined by the number of somites formed. Embryonic age ranged between 6.5 days post conception (d.p.c.) for stage 3 embryos and 8.0 d.p.c. for early somite stages. Mouse embryos: Mouse embryos (C57Bl/6j) were recovered from timed matings following standard procedures, staged according to Downs and Davies (1993) and fixed for further use.

TEM Embryos were dissected and fixed in a mixture of 4% paraformaldehyde–2% glutaraldehyde in Soerensen’s buffer for 1 hr at RT or at 41C overnight. Postfixing was performed in 1% osmium tetroxide in Soerensen’s buffer for 1 hr at 41C. Following dehydration, embryos were embedded in araldite CY212 (Plano, Wetzlar, Germany), and polymerized for up to 48 hr at 601C. Ultrathin sections (70 nm) of NP tissue were cut directly from these specimens or after re-embedding of serial semithin (1 mm) sections (Viebahn et al., 1995) using a Leica Ultracut S, mounted on copper grids, poststained with lead citrate, and viewed under a LEO 912 AB TEM (Carl Zeiss) at 80 kV.

Results Hensen’s node in the rabbit embryo is devoid of cilia

Cloning of goosecoid in rabbit A 792 bp fragment of the rabbit Gsc gene (accession number AM236594) was cloned by polymerase chain reaction from cDNA synthesized by reverse transcription of total RNA of E12 rabbit embryos, using primers targeted to conserved sequences between mouse and human. The primers used were forward 5 0 -ATGCCCGCCAGCATGTTCAG-3 0 , reverse 5 0 -CCGCGGCCGT CAGCTGTCCGAGTCC-3 0 , at an annealing temperature of 621C using 6% DMSO, 40 cycles. Whole-mount in situ hybridization, IHC, and histology Non-radioactive whole-mount in situ hybridization and vibratome sections of stained specimens were prepared using a Leica VT 1000 S (Bensheim, Germany) as described (Fischer et al., 2002). IHC was performed following standard protocols using a mouse antiacetylated tubulin monoclonal antibody (Sigma, Schnelldorf, Germany). For antibody detection, either Cy2- (Jackson Immunoresearch, Newmarket, UK) or Cy3- (Sigma) conjugated rabbit antimouse IgG antibodies were used. Propidium iodide (Fluka, Schnelldorf, Germany) as a nuclear stain was added before mounting (1 mg/ml). Embryos were observed with a confocal laser microscope (LSM 5 Pascal; Carl Zeiss, Oberkochen, Germany).

Fig. 1 Staging system of rabbit embryos. Between 7.0 and 8.0 d.p.c., the rabbit embryo develops from stage (st.) 4 (characterized by the presence of Hensen’s node, n), via st. 5 (np, notochordal plate; pp, prechordal plate) and st. 6 (elongated np and headfold) to somitogenesis. ps, primitive streak.

Hensen’s node appears as a dark-stained area at the tip of the primitive streak in whole-mount views of osmium-fixed embryos from stage 4 to early somite stages (Figs. 2A,3A). Histologically, the node area consists of a multilayered, epithelial-like cell density in which the dorsalmost cells are frequently bottle shaped as they appear to leave the epiblast layer of the embryo (Figs. 2B,4C,4D). Sagittal sections reveal the anterior limit of the node area to be the posterior extension of the basement membrane under the (presumptive or definitive) floor plate, the position of which is indicated by the alignment of the basal cell surfaces of the floor plate cells (Fig. 2B). A narrow extracellular space, sometimes highlighted by a wide cleft artificially arising during the histological processing of later stages (asterisks in Figs. 4D,4F,4G), is thus created between the floor plate cells dorsally and the prechordal mesoderm cells ventrally. Prechordal mesoderm cells (Fig. 2B) are the first cells leaving the node area toward the anterior of the embryo, and continue to migrate rostral to the notochordal cells emanating from the node a little later. Anterior to a fully grown notochordal plate, they will form the prechordal plate (cf. Figs. 6C,6C 0 ). Some of these early prechordal mesoderm cells that have just left the node rostrally carry nascent monocilia projecting into the narrow extracellular space between the prechordal cells and the underlying continuous hypoblast (Fig. 2C). Soon after bona fide notochordal cells have left the node area in early stage 5 embryos, the continuity of the hypoblast is lost in the midline of the embryo anterior to the node; consequently, the ventral surface of the notochordal plate cells is exposed to the fluid in the yolk sac cavity (Fig. 3) while the node area remains covered by hypoblast until early somite stages (Fig. 4). The

136

Fig. 2 Ultrastructure of Hensen’s node in the rabbit gastrula embryo. (A) Whole-mount view of osmium fixed embryo at stage (st.) 4. Arrow points to the anterior limit of Hensen’s node (n), which, due to its high cell density, appears darker than most of the embryonic disc. Black lines indicate the plane of section shown in B. (B) Semithin sagittal section of embryo in (A), anterior to the left. Cells in the area of the node and of the prechordal mesoderm (pm) show a dense, epithelium-like arrangement with little extracellular space. At the level of the pm, the basal cell membranes of the (presumptive) floor plate epithelium (fp) are aligned to form a straight line (arrowheads) dorsal to the pm while at the level of the

node (n), the pseudostratified epiblast epithelium lying dorsally is continuous with the dense cell mass of the node ventrally. Node, pm, and presumptive fp epithelium anterior to the node are covered ventrally by a continuous layer of hypoblast (hy). The arrow indicates anteriormost pm cell shown at high magnification in C. (C) Electron microscopy of pm cell (arrow) shown in B, carrying nascent cilium (s. inset) on its anteroventral surface, and intimately associated with the underlying hypoblast. Arrowheads point to desmosome-like junctions between hypoblast cells. Scale bars: 200 mm in A; 20 mm in B; 1.5 mm in C, and 0.3 mm in inset.

posteriormost segment of the early notochord, the notochordal process, is still multilayered (Fig. 4B) in a fashion similar to the prechordal mesoderm cells at this stage (Fig. 2B), and almost completely covered by hypoblast (Fig. 4B). Further anteriorly, the notochordal cells form a simple cuboidal epithelium, provide the notochordal process with a plate-like appearance, and are free of hypoblast and, therefore, in direct contact with the fluid of the yolk sac cavity (Fig. 4A). Single monocilia extend among small stud-like microvilli from every notochordal cell over the entire length of the exposed notochordal process and stand in clear contrast to the dense microvillous coat on neighboring hypoblast or endoderm cells (Fig. 3). Similar cilia are found over the entire length of the notochordal process exposed to the yolk sac cavity also at stage 6 (Figs. 3C–3E) and during early somitogenesis (not shown). IHC for acetylated tubulin confirmed the presence of cilia on the ventral surface of the notochordal process free of hypoblast (Figs. 5C,5C 0 ), while the ventral surface of the node did not show any signs of

immunohistochemically stained monocilia (Figs. 5C, 5C00 ; cf. Z-stack in Supplementary Movie 1). Taken together, this analysis demonstrates that Hensen’s node in the rabbit is devoid of monociliated cells at all stages, and that cilia are confined to the notochordal plate. Goosecoid expression defines the mammalian node In the rabbit embryo, strong Gsc signals were found in the anterior two-thirds of the primitive streak at stage 3 (not shown), and throughout Hensen’s node at stage 4 (Figs. 6A,6B). In late stage 4 embryos, Gsc was additionally found in the first prechordal mesodermal cells leaving the node anteriorly (arrow in Fig. 6B). Transcription of Gsc declined markedly toward the beginning of stage 5, when Gsc was not present in Hensen’s node but expressed in prechordal cells instead (not shown). Signals progressively intensified in the prechordal mesoderm as the notochordal plate formed and lengthened during stage 5 (not shown) and stage 6 (Fig. 6C). A mid-sagittal section at stage 6 demon-

137

Fig. 3 Notochordal plate of the rabbit carries cilia on its ventral side as determined by scanning electron microscopy (SEM). Comparison of osmium fixed (A) and SEM (B) analyzed embryos reveals a longitudinal gap in the hypoblast (B) over most of the notochordal process (np), which appears as a translucent strip anterior to the node area (anterior and posterior borders delineated by two broken lines in A and B) at stage 5. (C–E) At stage 6, the np is exposed almost completely free of hypoblast cells and shows on every cell a single cilium protruding from the ventral surface among numerous studlike microvilli (D, E). Horizontal lines in (B) indicate the approximate level of sections in Figs. 4A– 4C. Scale bar, 200 mm (B), 100 mm (C), 5 mm (D), 1 mm (E).

strated that Gsc signals were now confined to prechordal plate cells (Fig. 6C 0 ). In contrast, floor plate and notochordal plate as well as Hensen’s node, which was still easily recognizable by its thickened appearance and dense epithelium-like cell arrangement, were devoid of Gsc transcripts (Figs. 6C,6C 0 ). Gsc expression in mouse embryos of corresponding stages from late streak to early headfold is depicted in Figure 6D–6F (staging system by Downs and Davies, 1993). At the late streak stage, transcripts were found throughout a region at the distal tip of the egg cylinder (Fig. 6D), which, in histological sections, was about five to six cells wide along the dorsal to the ventral axis, and characterized by dense arrangement of cells (Figs. 6D 0 ,6D00 ; cf. Sulik et al., 1994; Bellomo et al., 1996). This region thus, both morphologically and molecularly, appeared homologous to Hensen’s node in the rabbit and we therefore choose the term ‘‘node’’ to distinguish this distinct area from the primitive streak posteriorly and floor plate or notochord anteriorly. Slightly later, at the zero bud stage, Gsc expression in the node had declined significantly (Figs. 6E,6E 0 ). A new domain became apparent anterior of the node and restricted to the ventral aspect of the egg cylinder (Figs. 6E,6E 0 ,6E00 ). These Gsc-positive cells constitute the prechordal

mesoderm, i.e., the first mesodermal cells leaving the node in the midline anteriorly, destined to become the prechordal plate at later stages. At the late headfold stage, Gsc continued to be expressed in the prechordal plate mesoderm and overlying neuroectoderm (Figs. 6F,6F 0 ). Signals were absent from the indentation at the distal aspect of the embryo, which, in the anterior direction, was continuous with the bilaminar epithelium of the floor plate and notochordal plate, and that abutted posteriorly against the node (Fig. 6F00 ). Comparison of this expression pattern with that in rabbit reveals that (1) the murine node is at all stages characterized by a distinct thickening and dense arrangement of epithelium-like cells and (2) Gsc marks the node (as the early organizer) at the late streak stage and the prechordal mesoderm from the zero bud stage onwards. Nodal expression encompasses the posterior notochord (PNC) in the vertebrate embryo The first expression of Nodal in gastrulation stages of the rabbit was seen in the anterior half of the primitive streak as it elongates during stage 3 (Fig. 7A). At stage 4, when Hensen’s node is first formed and presents itself as a marked thickening at the anterior end of the prim-

138

Fig. 4 Gradual emergence of Hensen’s node from the hypoblast (hy) during early somite stages in the rabbit embryo. hypoblast, node (n), and notochordal process (np) during stage (st.) 5 (A–C), st. 6 (D, E), and early somite (som) stages (F, G) as seen in transverse (A–C) and sagittal (D–G), semithin (A–C, D, F, G), and ultrathin (E) sections through the node area (C), the anterior and posterior part of the notochord (A, B), or both (D–G). Arrows mark the anterior border of the hypoblast layer. Asterisks mark artificial spaces that are typically created during histological processing and indicate the space between floor plate and np; double asterisks mark the space between the hypoblast and primitive node and streak in (F) and (G), respectively. At st. 5 (A–C), the anterior part of the np in (A, level of section indicated by line in Fig. 3B), is separated from the overlying floor plate epithelium by a narrow line of extracellular space dorsally, and ventrally covered by hypoblast

cells laterally only and free of hypoblast medially; near the node (B, level of section indicated by line in Fig. 3B), the np is separated from the overlying floor plate as in (A) and, again, ventrally covered at its lateral edges but only by thin cellular processes (carrying a dense microvillous coat) of hypoblast cells; the node itself (C), recognizable by the continuum between the pseudostraified epiblast layer and the dense cell mass of the node, is covered by hypoblast cells with prominent round nuclei. At st. 6 (D, E) and the onesomite stage (F), the hypoblast covers the node completely, while at the three-somite stage the hypoblast layer covering the node becomes interrupted (G). The ultrastructure of hypoblast cells is characterized by a dense coat of long irregular microvilli while node or notochordal cells (no in E) have only sparse and short microvilli. Arrowheads in (A) and (B) mark the lateral border of the gap in the hypoblast. Scale bar: 20 mm in A–D, F, G; 2.5 mm in E.

Fig. 5 The rabbit node is devoid of cilia. Immunohistochemical analysis of acetylated tubulin in node (n) and notochordal plate (np) of a stage (st.) 5 embryo. (A) Wholemount view with separation of node and np indicated by a dashed line. The vertical line indicates the plane of section in (B). (B) Brightfield view of median sagittal vibratome section (30 mm) of embryo in (A). (C) Confocal picture of stained vibratome section in (B) revealing cilia exclusively on np cells (enlargements shown in C 0 , C00 ; cf. Z-stack in Supplementary Video 1). Green, acetylated tubulin; red, nuclei (propidium iodide). fp, floor plate.

139

Fig. 6 Identification of the murine node by comparative analysis of goosecoid (Gsc) gene expression in rabbit and mouse. Ventral (A–C) and lateral (D–F) views and vibratome sections (C 0 –F 0 ) of embryos hybridized in whole mounts using specific antisense Gsc probes. (A– C) Gsc marks Hensen’s node (n) and the prechordal plate (pp) in the rabbit. At early stage (st.) 4 (A), Gsc is heavily transcribed in Hensen’s node and in the anterior third of the primitive streak (ps). Toward the end of st. 4 (B), Gsc marks the first prechordal mesoderm (pm) cells leaving the node anteriorly (arrow). At st. 6 (C, C 0 ), Gsc expression is confined to the pp. (D–F) Gsc marks the node and pp in the mouse. (D) At the late streak stage (LS), Gsc is expressed in the anterior primitive streak and node. (E) At the zero bud stage (0B) Gsc expression in the primitive streak and node has faded, and signals are strongest in the first presumtive prechordal cells leaving the node anteriorly (arrow). (F) At late headfold (LHF), Gsc expression is confined to the pp and overlying neuroectoderm. Parasagittal (D 0 , D00 ) and sagittal (E 0 , E00 , F 0 , F00 ) histological sections of embryos in (D–F) show the dense, epitheliumlike cell arrangement in the node area and the separation between floor plate and pm (E00 ) or the notochordal cells (F00 ) anterior to the node. Dashed lines in D00 –F00 mark anterior and posterior limits of the node area as determined by these histological characteristics of node and pm (E00 ) or notochordal plate (F00 ), respectively. Asterisks mark distal indentation (posterior notochordal plate) in (F, F 0 ). a, anterior; fp, floor plate; p, posterior.

itive streak, Nodal was found in the anteriormost primitive streak and node (Figs. 7B,7B 0 ). However, Nodal expression at the node was transient, as the signal declined toward the end of stage 4 (not shown) and was absent in early stage 5 embryos (Fig. 7C), when notochordal cells first emerge from the node. Toward the end of stage 5, concomittant with the epithelialization of the notochordal process into the notochordal plate, and at stage 6 during the elongation of the notochordal plate, the Nodal signal reappeared at the margin of the notochordal plate on both sides in a symmetrical fashion (Figs. 7D,7E). At the two to three somite stage, asymmetric expression of Nodal in the left lateral plate mesoderm (lpm) first appeared in a domain that, in its craniocaudal position, corresponded to the symmetrical expression accompanying the notochordal plate (Fig. 7E). A transiently increased expression in the left notochordal domain as compared with the right side occasionally became evident, but this occurred only after the left lpm expression was well established at the three to four somite stage (Figs. 7F,7F 0 ). The bilateral Nodal domains represented a longitudinal row of two to three cells width at the margin of the notochordal plate that differed from the intervening cells, as they were not in contact with the overlying floor plate (Fig. 7F 0 ). These nodal expressing cells mark the transition zone between the columnar epithelium of the notochordal plate and the flat epithelium of the neighboring endoderm (Figs. 7F 0 ,8C 0 ). These sites of Nodal gene expression correspond to the region with cilia seen in SEM analysis (Fig. 2). In the mouse, Nodal expression at gastrula or neurula stages (Figs. 7G–7I) closely resembles this expression pattern in the rabbit: at the zero bud stage, Nodal is found throughout the node, while no signal was found in the anteriorly located Gsc-positive prechordal cells (Figs. 7G,7G 0 ; cf. to Fig. 6E00 ). At the late bud (Fig. 7H) and early headfold stages (Fig. 7I), the Nodal domain appeared Y-shaped in ventral views. It included the node and encompassed the posterior aspect of the notochordal plate laterally (Figs. 7H,7I). Thus, in mouse Nodal marked the node until and slightly after the notochordal process formed at headfold stages, while at somite stages Nodal was restricted to the lateral borders of the PNC (Figs. 7G–7I,8B,8B 0 ). In Xenopus, bilateral Nodal signals were also found at the posterior end of the forming notochord from stage 15 up to early tadpole stages (Fig. 8A and data not shown). Histological examination identified Nodal expressing cells at the outer boundary of the gastrocoel roof plate (GRP), where they constituted one to two cells on the surface of the archenteron on both sides of the multilayered notochord (Fig. 8A 0 ). As revealed by anti-tubulin IHC (Fig. 8D), cilia reside in a triangular domain, extending from the blastopore posteriorly along the forming notochord. SEM analysis verified the monociliated nature of cells (Fig. 8G). Figures

140

8B,8E,8H and Figs. 8C,8F,8I depict comparable data for mouse and rabbit embryos, demonstrating the conserved nature of bilateral Nodal expression at the PNC at mid- to late-gastrula stages (Figs. 8A–8C), and the monociliated character of intervening cells (Figs. 8D– 8I). In conclusion, this comparative analysis establishes the PNC as a conserved embryological entity in the ventral midline of vertebrate embryos, despite differing modes of gastrulation and variable architectures of early embryos.

Discussion Definition of the PNC On the basis of our histological, electron microscopical, and gene expression analysis, five distinct regions can be

distinguished along the ventral midline during gastrulation in the rabbit and the mouse; these are, from cranial to caudal, (1) the prechordal plate, (2) the anterior notochord (at the level of the somites), (3) the PNC, (4) the node, and (5) the primitive streak. Of these, the PNC stands out by (a) being encompassed by bi-lateral Nodal expression and by (b) the regular appearance of single cilia on the ventral aspect of every notochordal cell. Furthermore, the PNC presents an entity distinct from the node or organizer as it does not express the organizer gene Gsc. Interspecific analysis, which applies the same parameters (ciliation and gene expression) to Xenopus as the primary modern amphibian model for organizer activity, reveals the GRP to be a structure strikingly similar to the PNC of the mammalian embryo and suggests that the definition of a posterior subsegment in the notochord involved in laterality determination may be a conserved feature among vertebrates. Although PNC and notochord are continuous and commonly express notochordal marker genes such as Foxa2, Brachyury, and noto as well as ciliary markers such as lrd, polaris, and Foxj1 (Wilkinson et al., 1990; Herrmann and Kispert, 1994; Supp et al., 1997; Brody et al., 2000; Murcia et al., 2000), they differ in several important aspects: (1) the PNC is significantly wider than the remaining (anterior) part of the notochord (cf. Fig. 2B); (2) the PNC is bordered by Nodal transcription on both sides; and (3) the PNC is the site of a ciliaFig. 7 Nodal expression defines the early node and the posterior notochordal plate (np) in rabbit and mouse embryos. Whole-mount in situ hybridization of rabbit and mouse embryos using specific antisense Nodal probes. (A–F) Nodal expression in the node and bilateral notochordal domain in the rabbit. All specimens are shown in ventral view, sections with dorsal side up. (A) Nodal expression in the anterior primitive streak (ps) at stage (st.) 3. (B) Nodal expression throughout Hensen’s node (n) and in an anterior segment of primitive streak at st. 4. Plane of transverse section in (B 0 ) is indicated by horizontal line. (C) Absence of Nodal signals at early st. 5. (D–F) Bilaterally symmetrical Nodal expression from st. 6 onwards, marking the transition between the columnar epithelium of the np and the flat epithelium of the neighboring endoderm (F 0 , cf. Fig. 8C 0 ). (E, F) Asymmetric Nodal expression in the left (l) lateral plate mesoderm (lpm). Asymmetric Nodal signals in the left lpm at the two to three somite (som) stage (E). At three to four somites (F), the left lateral plate domain is accompanied by a left asymmetry in the np domain (F 0 ). Nodal-positive cells on both sides of the np are distinct from intervening notochordal cells due to lack of contact with the overlying floor plate (fp). (G–I) Nodal expression in node and bilateral notochordal domain in the mouse. Embryos shown in lateral view, anterior to the left. (G) At the zero bud stage (0B), Nodal expression is restricted to the node. The sagittal section in (G 0 ) shows the same histological characteristics of the node area and of the prechordal mesoderm as found in the semithin sections (Fig. 4,6D 0 ); dashed lines indicate the anterior and the posterior border of the node area accordingly. During late bud (LB) and early headfold (EHF) stages (H, I), distal views of egg cylinders (insets in H and I) reveal Nodal signals in node and two lateral wings (together the so-called ‘‘crown cells’’; (Bellomo et al., 1996)), extending from the node anteriorly (arrowheads). Asterisks in (I) indicate the location of emerging ventral indentation. Dashed lines in the insets of (H) and (I) highlight the boundary between node and posterior np. pp, prechordal plate; r, right.

141

Fig. 8 Conserved bilateral Nodal expression encompassing a monociliated notochordal domain in frog, mouse, and rabbit embryos. (A–C) Ventral views of embryos hybridized with specific Nodal probes in whole mounts demonstrate bilateral Nodal expression encompassing the posterior aspect of the notochord (no) in all three species. Domains representing the gastrocoel roof plate (GRP) in frog (A) and the posterior notochord (PNC) in mouse and rabbit (B, C) are indicated by square brackets. (A 0 –C 0 ) Transverse sections of embryos, levels indicated by horizontal lines in (A–C), confirm the Nodal expression to reside in the ventral cell layer adjacent to the notochord. Sections oriented with dorsal side up. Immunohistochemistry for acetylated tubulin (D–F) and scanning electron

microscopy (G–I) reveal cilia on the ventral surface of GRP/PNC, as defined by Nodal expression. Note that GRP cells are much larger than cells in PNC. Insets in (D–F) show higher magnifications of cilia. The inset in (D) presents a confocal image of a transverse vibratome section. Frog: Posterior parts of embryos were cut transversally at about the middle of the embryos. (A) Lumen represents the archenteron; the dorsal side is facing upwards. (D, G) Close-up views from archenteron onto GRP. Mouse: Embryos presented in distal views (B, E, H) to allow a close-up examination of cilia on the PNC (E, H). Rabbit: overview (C) and close-ups (F, I) of the ciliated region in ventral views. a, anterior; en, endoderm; p, posterior.

driven leftward fluid flow (Nonaka et al., 1998; Hamada et al., 2002; Hirokawa et al., 2006).

itive streak. Intriguingly, Hensen (1876) had already noticed that the ventral cell layer ‘‘was inseparable at the site of the node.’’ At that time, the ventral layer of the early mammalian gastrula was still considered to be the forerunner of the endoderm and this led Hensen to state that ‘‘the germ layers’’ were inseparable at the node. As the early lineage studies by Gardner and his colleagues showed that the endoderm arises from the epiblast during gastrulation (Gardner and Rossant, 1979; cf. Tam et al., 2003), endoderm cannot be present before gastrulation and, therefore, the present results (cf. Fig. 4) and previous morphological work on the mouse (Sulik et al., 1994; Bellomo et al., 1996) suggest

Morphology Our histological analysis and comparison with published work (Sulik et al., 1994; Bellomo et al., 1996) reveals that our criteria for the rabbit node hold for mouse as well, namely the combination of bottle shaped cells at the level of the ingressing epiblast dorsally with the dense epithelium-like cell arrangement ventrally in which layering is indistinct and that together produce the thickened structure at the anterior end of the prim-

142

that Hensen must have tried to remove the hypoblast layer (visceral endoderm in the mouse) and found it to be fixed to the node area, therefore covering it ventrally. With regard to the distinct border between bottle cells and the densely packed cells in the interior of the mammalian node, Hausen and Riebesell (1991), in their most detailed description of early Xenopus development, have noted a similar phenomenon for the circumblastoporal region, a part of which is the dorsal lip tissue, in which mesodermal and neurogenic cells could not be distinguished. Cilia are a common phenomenon on the apical surface of early embryonic epithelia and, on the ventral surface of the mouse egg cylinder, are thought to play an important role during determination of laterality in the mouse embryo (Nonaka et al., 1998; Hamada et al., 2002; Hirokawa et al., 2006). Cilia have previously been described at the ventral surface of the dorsal blastopore at stage 13/14 and the GRP at stage 14 (Essner et al., 2002; Shook et al., 2004), which suggests that in Xenopus, too, fluid flow could be involved in laterality determination. The topographical correlation with a symmetrical Nodal expression domain lateral to the PNC now narrows down the region in which cilia may be involved in laterality determination across species boundaries, be it as active fluid flow generators (Hirokawa et al., 2006) or as passive fluid flow sensors (McGrath et al., 2003; Tabin and Vogan, 2003).

Gene expression: Gsc The best marker gene for the full gastrula organizer to date is the homeobox transcription factor Gsc (Cho et al., 1991; Blum et al., 1992; De Robertis et al., 1992; Izpisua-Belmonte et al., 1993; Schulte-Merker et al., 1994). The onset of transcription in the dorsal lip of Xenopus gastrula embryos at stage 10.5 coincides with the development of Spemann’s organizer activity (Cho et al., 1991). Likewise, Gsc marks Hensen’s node in chick and rabbit (Izpisua-Belmonte et al., 1993, this work), and the embryonic shield of the zebrafish gastrula embryo (De Robertis et al., 1992), which is considered the equivalent of the node in the teleost fish larva (Oppenheimer, 1936; Ho, 1992). The absence of Gsc transcripts in the PNC, i.e., in the distal indentation of the mouse E7.5 egg cylinder, argues against organizer activity in this structure, in line with the experimental analysis of Tam and colleagues (for a review, see, Robb and Tam, 2004), while the expression at the anterior end of the primitive streak from early to mid-gastrula (about E6.4–E6.8) parallels the potency to induce axial tissue in homo- and heterotypic grafts (Blum et al., 1992; Kinder et al., 2001). Three criteria could thus be used to localize the embryonic organizer in vertebrate embryos: (1) the

expression of Gsc; (2) a bulk of epithelium-like intermingled cells; and (3) axis-inducing potency. We further suggest to reserve the term ‘‘node’’ to characterize tissue in which all three criteria are fulfilled. In mouse embryos, these criteria are met by the mid-gastrula organizer (MGO), as defined by Tam. In contrast, the EGO, while expressing Gsc, lacks a morphological node and head-patterning activity (Kinder et al., 2001; Robb and Tam, 2004). Gene epression: Nodal Drawing on Nodal expression for the definition of the PNC appears to be appropriate as Nodal is, to date, the only gene described so far to outline a homologous posterior segment of the notochord in several vertebrate species (mouse, rabbit, frog). Being the first indicator of successfully established laterality in the mammalian embryo (Hamada et al., 2002; Hirokawa et al., 2006; Raya and Belmonte, 2006), Nodal is a good candidate as a target gene in the cascade linking the initial symmetry breaking to molecularly established laterality, with symmetry breakage likely occurring at around the time when the node is formed in the mammalian embryo. Possible function of the PNC The three specific PNC characteristics mentioned above may bear on the specific function of this structure, namely its role in the specification of laterality. PNC mutant mice in which the formation or structure of the PNC is affected consistently display laterality defects but do not affect the two main embryonic axes (dorsal– ventral and anterior–posterior). These mutants may be grouped into those affecting basic PNC architecture (dll1, Brachyury, noto; Herrmann and Kispert, 1994; Krebs et al., 2003; Przemeck et al., 2003; Abdelkhalek et al., 2004), ciliogenesis (Kif3B and polaris; Nonaka et al., 1998; Murcia et al., 2000), ciliary polarization (inv; Okada et al., 2005), ciliary motility (iv; Okada et al., 1999), or as yet unknown ciliary functions (Pkd2; Pennekamp et al., 2002). All these latter genes function in setting up the vectorial fluid flow known as ‘‘Nodal Flow’’ or are suspected to act in its perception and relay to the left-specific Nodal signaling cascade. Interestingly, the Brachyury and noto mouse mutants, which both affect the architecture of the PNC, are characterized by disruption of tail formation (Herrmann and Kispert, 1994; Abdelkhalek et al., 2004), which suggests that the PNC may also be directly involved in secondary neurulation and tail morphogenesis. The formal possibility, that the distal indentation (PNC) of the E7.5–E8.5 mouse egg cylinder represents the node at later stages, seems unlikely for the following reasons. Morphologically, mid- to late-gastrula embryos retain a small patch of tissue between PNC and

143

primitive streak in which germ layers cannot be distinguished (cf. Fig. 5), which is one of the defining histological characteristics of the node. In addition, many mutants are known that affect the PNC, for example dll1, Brachyury, noto, Kif3B, or lrd (Herrmann and Kispert, 1994; Supp et al., 1997; Nonaka et al., 1998; Krebs et al., 2003; Przemeck et al., 2003; Abdelkhalek et al., 2004), which all show distinct phenotypes at later embryogenesis but develop normally through gastrulation and neurulation. Ablation experiments performed on E7.5 mouse embryos that included PNC as well as adjacent tissue had no effect on anterior–posterior patterning, providing further evidence for the absence of organizer activity in the PNC (Davidson et al., 1999). The definitive experiment to exclude organizer activity from PNC tissue seems difficult to achieve, as this would require a clean surgical separation of node and PNC in mouse embryos, which appears rather challenging due to the embryo’s small size. Perhaps, an experiment can be designed using rabbit embryos, along the lines performed by Knoetgen et al. (2000), as rabbit blastodiscs are considerably larger and experimental manipulations of blastodiscs can be more easily performed as compared with the mouse egg cylinder. Phylogenetic significance of the PNC Comparison of mammalian and amphibian embryos suggests the presence of a structural module characterized by (1) bi-lateral Nodal expression; (2) motile monocilia; and (3) a cilia-driven leftward flow, the function of which primarily seems to be the correct development of laterality. Using this functional characteristic the PNC/ GRP may be regarded as a laterality organizing center. However, in the chick, one of the best-characterized model organisms for LR development, cilia-driven leftward flow has not been found nor has an equivalent of the PNC been described. This may have two reasons: (1) molecular asymmetries have been found at much earlier stages in the chick than in mouse or frog, i.e., before development of the notochord (Levin et al., 1995), and (2) the chick notochord forms as a more or less solid rod, retains this shape through to late somite stages, and is not inserted into the lower layer (hypoblast or endoderm) or exposed to the cavity below the lower layer at any stage. Interestingly, nodal gene expression at the anterior aspect of Hensen’s node, lateral to the posterior base of the notochord, was reported in chick as well (Levin et al., 1995). Perhaps, the focus of analysis has been on stages preceding stages with a functional PNC equivalent. However, cilia-driven leftward flow in other organisms and asymmetric Nodal expression in the left lpm observed in the chick, too (Nodal transcription in the left lpm starts at stage 8; Levin et al., 1995) suggests that the time window for possible cilia-driven leftward flow in chick might be at around a similar stage (stage 7) as in mammals. Alternatively, chick embryos might

have lost a PNC-based mechanism, as the chick node displays early morphological asymmetries that might have superseded the need for a cilia-driven leftward flow in LR determination. Analysis of other avian embryos might prove worthwhile in that respect. How are these morphological and functional aspects conserved in vertebrates other than mammalian embryos? Our analysis in Xenopus shows that two of the above PNC characters, bi-lateral Nodal expression (Lowe et al., 1996) and monocilia (Shook et al., 2004), are conserved in amphibian embryos (Fig. 7). In Xenopus, Keller and colleagues have characterized the respective structure, which they call gastrocoel roof plate (GRP; Shook et al., 2004), a term reminiscent of the previously used ‘‘archenteron’’ for the mouse PNC (Theiler, 1972). The GRP in Xenopus is derived from surface ectoderm (Shook et al., 2004). Like the PNC, it is continuous with the notochord, which, in Xenopus, forms from the GRP and the deep cells (Shook et al., 2004). The main function of these surface-derived cells of the GRP transiently exposed to the archenteron might thus be to set up a cilia-driven leftward flow and promote LR development. In zebrafish, a homologous module to the PNC/GRP has been characterized with Kupffer’s vesicle (KV). KV is a transient structure derived from the dorsal forerunner cells, which do not involute during gastrulation, but remain in a posterior position with respect to the forming notochord. Its relatedness to PNC/GRP is threefold: (1) KV is bordered by Nodal expression (one of the three zebrafish homologs, southpaw; Long et al., 2003); (2) it harbors motile monocilia, which (3) produce a cilia-driven leftward flow required for LR development (Essner et al., 2005; Kramer-Zucker et al., 2005). The homology with the mouse PNC is further supported by genetic evidence. The one zebrafish mutant that does not develop a KV is no tail (ntl; Melby et al., 1996). Similarly, the archenteron (PNC) was reported absent in Brachyury mutant mouse embryos (Fujimoto and Yanagisawa, 1983), a finding that we have confirmed recently (P.A. and M.B., in preparation). It will be interesting to analyze whether GRP/ cilia-driven leftward flow is affected in Xbra loss-offunction experiments in Xenopus, and whether and how laterality is altered in treated embryos. Terminology Our definition of the PNC in the rabbit and mouse embryo suggests that a ‘‘re-renaming’’ of the distal indentation of the E7.5–E8.5 mouse egg cylinder, commonly referred to as the mouse ‘‘node,’’ may be appropriate. Until its renaming in 1991 by Beddington et al. (1992), the PNC in mouse was referred to as ‘‘archenteron’’ (Theiler, 1972). Theiler, in his first complete account on development and normal stages of the house mouse, described the cylindrical cells in the ventral midline of the E7.5 presomite stage embryo, which

144

‘‘. . . posteriorly . . . form the notochordal plate which is slightly indented. This indentation has also been called the archenteron’’ (Theiler, 1972). He went on to clarify that ‘‘the archenteron is a transitory structure which has nothing to do with the formation of the hindgut’’ (Theiler, 1972). The continuity with the notochord anteriorly was referred to by Theiler in his description of the E8 embryo (one to seven somites), where he described the notochord as ‘‘a long strand of columnar cells (which) still has a shallow groove posteriorly (the archenteron)’’ (Theiler, 1972). The renaming from ‘‘archenteron’’ into ‘‘node’’ was introduced by Beddington at a symposium on postimplantation development in the mouse in 1991 (Beddington et al., 1992). ‘‘This (the archenteron) is a misnomer as it is not equivalent to the archenteron in amphibians but, as far as we can tell, corresponds to the dorsal blastopore lip of Xenopus or Hensen’s node of the chick. Therefore, I would suggest that we call it the ‘node’’’ (Beddington et al., 1992). This renaming reflected the search for a landmark at the anterior end of the primitive streak, similar to Hensen’s discovery of the node as a prominent thickening in the rabbit blastodisc. This now seems obsolete as the use of this definition unites two entities (the node and the PNC) that can, on the basis of present knowledge, be distinguished both morphologically and functionally. We propose to abandon the term ‘‘node’’ for the characterization of the distal indentation for the following reasons. The hallmarks of the primary embryonic—or Spemann—organizer have been defined by classical homo- and heterotypic transplantation experiments to include (1) self-differentiation into notochordal tissue; (2) neural induction of ventral ectoderm; (3) dorsalization of ventro-lateral mesoderm; and (4) induction of gastrulation movements (Joubin and Stern, 1999; Kinder et al., 2001; Niehrs, 2004; De Robertis, 2006). By these criteria, organizer function has been unequivocally assigned to the dorsal lip of the amphibian blastopore and to Hensen’s node in chick and rabbit gastrulae (cf. Waddington, 1933; Waddington and Schmidt, 1933; Viebahn, 2001). The first account of murine organizer activity reported inducing capacity in distal fragments of early-gastrula egg cylinders between 6.4 and 6.8 days post coitum (d.p.c.) in Einsteck experiments into the blastocoel of Xenopus recipient blastulae (Blum et al., 1992). Donor embryos in these experiments lacked PNC, notochord, and prechordal plate due to developmental stage (Blum et al., 1992). The first report of homotypic grafts, which resulted in axis duplications in about 35%–40% of cases, used E7.5 embryos (Beddington, 1994). Although these embryos might have had a PNC, the experimental details suggest that just the node/organizer was transplanted (Beddington, 1994). A careful and detailed analysis of the inductive potency of graft tissue over time revealed that the MGO of the E7.0 mouse egg cylinder harbored full

organizer activity, while later embryos progressively lost the capacity to induce anterior midline structures such as brain, cranial mesenchyme, heart, and anterior mesendoderm (Kinder et al., 2001; Robb and Tam, 2004). Taken together, these studies demonstrate that appearance of organizer activity clearly preceded the formation of PNC, notochord, and prechordal plate during mouse gastrulation.

Conclusion In conclusion, we have shown that in mammalian gastrula embryos node and PNC are distinct entities with unique morphological, embryological, and molecular properties. Ciliation and neighboring Nodal expression point to the possibility that the PNC acts as laterality organizing center while the node remains responsible to organize the longitudinal body axis. Our concept should help to identify and characterize homologous organizer and signaling centers of the vertebrate embryo and might provide a tool to investigate the evolution of chordate laterality. Acknowledgments We thank Patrick Tam for valuable discussions and suggestions, Gert Sonntag and Jochen Tham (Zeiss Germany) for their continuous help and support with all aspects of microscopy, Werner Amselgruber for access to his SEM, Sybille Wolf (Stuttgart) and Johan W.M. Hekking (Maastricht) for help in SEM analysis, and Andreas Miething (Bonn), Peter Schwartz, and Heike Hu¨hn (Go¨ttingen) for critical parts of the TEM analysis. This work was supported by DFG grants to M.B.; a fellowship of the Elitefo¨rderungsprogramm Baden-Wu¨rttemberg to A.F.; a PhD. fellowship of the Boehringer Ingelheim Fonds to K.M.; the Sonderforschungsbereich 495; and DFG and DAAD grants to C.V.

Supplementary Material The following supplementary material is available for this article: Video Clip S1. Exclusive presence of cilia on notochordal plate cells. This Video shows a Z-series of confocal sections of a stage 5 embryo stained for nuclei (propidium iodide, red) and alpha-acetylated tubulin (green). The series starts at the ventral side and advances through the node and notochordal plate. Progressing along the dorso-ventral axis, cilia (in transverse section) become visible as green dots on the small cells of the notochordal plate (np), while the node region (n) does not reveal ciliary structures. This material is available as part of the online article from: http://www.blackwell-synergy.com/doi/abs/ 10.1111/j.1432-0436.2006.00124.x (This link will take you to the article abstract). Please note: Blackwell Publishing is not responsible for the content or functionality of any supplementary

145

materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

References Abdelkhalek, H.B., Beckers, A., Schuster-Gossler, K., Pavlova, M.N., Burkhardt, H., Lickert, H., Rossant, J., Reinhardt, R., Schalkwyk, L.C., Muller, I., Herrmann, B.G., Ceolin, M., Rivera-Pomar, R. and Gossler, A. (2004) The mouse homeobox gene Not is required for caudal notochord development and affected by the truncate mutation. Genes Dev 18:1725–1736. Beddington, R.S. (1994) Induction of a second neural axis by the mouse node. Development 120:613–620. Beddington, R.S., Jaenisch, R., Smith, J.C., McLaren, A.L., Lawson, K.A. and Herrmann, B.G. (1992) Three-dimensional representation of gastrulation in the mouse. Ciba Found Symp 165:55–60. Beddington, R.S. and Robertson, E.J. (1999) Axis development and early asymmetry in mammals. Cell 96:195–209. Bellomo, D., Lander, A., Harragan, I. and Brown, N.A. (1996) Cell proliferation in mammalian gastrulation: the ventral node and notochord are relatively quiescent. Dev Dyn 205:471–485. Blum, M., Gaunt, S.J., Cho, K.W., Steinbeisser, H., Blumberg, B., Bittner, D. and De Robertis, E.M. (1992) Gastrulation in the mouse: the role of the homeobox gene goosecoid. Cell 69: 1097–1106. Boettger, T., Knoetgen, H., Wittler, L. and Kessel, M. (2001) The avian organizer. Int J Dev Biol 45:281–287. Brody, S.L., Yan, X.H., Wuerffel, M.K., Song, S.K. and Shapiro, S.D. (2000) Ciliogenesis and left–right axis defects in forkhead factor HFH-4-null mice. Am J Respir Cell Mol Biol 23: 45–51. Cho, K.W., Blumberg, B., Steinbeisser, H. and De Robertis, E.M. (1991) Molecular nature of Spemann’s organizer: the role of the Xenopus homeobox gene goosecoid. Cell 67:1111–1120. Collignon, J., Varlet, I. and Robertson, E.J. (1996) Relationship between asymmetric nodal expression and the direction of embryonic turning. Nature 381:155–158. Conlon, F.L., Lyons, K.M., Takaesu, N., Barth, K.S., Kispert, A., Herrmann, B. and Robertson, E.J. (1994) A primary requirement for nodal in the formation and maintenance of the primitive streak in the mouse. Development 120:1919–1928. Davidson, B.P., Kinder, S.J., Steiner, K., Schoenwolf, G.C. and Tam, P.P. (1999) Impact of node ablation on the morphogenesis of the body axis and the lateral asymmetry of the mouse embryo during early organogenesis. Dev Biol 211:11–26. Davidson, B.P. and Tam, P.P. (2000) The node of the mouse embryo. Curr Biol 10:R617–R619. De Robertis, E.M. (2006) Spemann’s organizer and self-regulation in amphibian embryos. Nat Rev Mol Cell Biol 7:296–302. De Robertis, E.M., Blum, M., Niehrs, C. and Steinbeisser, H. (1992) Goosecoid and the organizer. Dev (Suppl): 167–171. De Robertis, E.M., Larrain, J., Oelgeschlager, M. and Wessely, O. (2000) The establishment of Spemann’s organizer and patterning of the vertebrate embryo. Nat Rev Genet 1:171–181. Downs, K.M. and Davies, T. (1993) Staging of gastrulating mouse embryos by morphological landmarks in the dissecting microscope. Development 118:1255–1266. Essner, J.J., Amack, J.D., Nyholm, M.K., Harris, E.B. and Yost, H.J. (2005) Kupffer’s vesicle is a ciliated organ of asymmetry in the zebrafish embryo that initiates left–right development of the brain, heart and gut. Development 132:1247–1260. Essner, J.J., Vogan, K.J., Wagner, M.K., Tabin, C.J., Yost, H.J. and Brueckner, M. (2002) Conserved function for embryonic nodal cilia. Nature 418:37–38.

Fischer, A., Viebahn, C. and Blum, M. (2002) FGF8 acts as a right determinant during establishment of the left–right axis in the rabbit. Curr Biol 12:1807–1816. Fujimoto, H. and Yanagisawa, K.O. (1983) Defects in the archenteron of mouse embryos homozygous for the T-mutation. Differentiation 25:44–47. Gardner, R.L. and Rossant, J. (1979) Investigation of the fate of 4–5 day post-coitum mouse inner cell mass cells by blastocyst injection. J Embryol Exp Morphol 52:141–152. Hamada, H., Meno, C., Watanabe, D. and Saijoh, Y. (2002) Establishment of vertebrate left–right asymmetry. Nat Rev Genet 3:103–113. Hamburger, V. and Hamilton, H.L. (1992) A series of normal stages in the development of the chick embryo. Dev Dyn 195: 231–272. Hausen, P. and Riebesell, M. (1991) The early development of Xenopus laevis. Verlag der Zeitschrift fu¨r Naturforschung, Tu¨bingen. Hensen, V. (1876) Beobachtungen u¨ber die Befruchtung und Entwicklung des Kaninchens und Meerschweinchens. Z Anat Entwicklungsgeschichte 1:213–273; 353–423. Herrmann, B.G. and Kispert, A. (1994) The T genes in embryogenesis. Trends Genet 10:280–286. Hirokawa, N., Tanaka, Y., Okada, Y. and Takeda, S. (2006) Nodal flow and the generation of left–right asymmetry. Cell 125:33–45. Ho, R.K. (1992) Axis formation in the embryo of the zebrafish, Brachydanio rerio. Semin Dev Biol 3:53–64. Idkowiak, J., Weisheit, G. and Viebahn, C. (2004) Polarity in the rabbit embryo. Semin Cell Dev Biol 15:607–617. Izpisua-Belmonte, J.C., De Robertis, E.M., Storey, K.G. and Stern, C.D. (1993) The homeobox gene goosecoid and the origin of organizer cells in the early chick blastoderm. Cell 74:645–659. Joubin, K. and Stern, C.D. (1999) Molecular interactions continuously define the organizer during the cell movements of gastrulation. Cell 98:559–571. Kinder, S.J., Tsang, T.E., Wakamiya, M., Sasaki, H., Behringer, R.R., Nagy, A. and Tam, P.P. (2001) The organizer of the mouse gastrula is composed of a dynamic population of progenitor cells for the axial mesoderm. Development 128:3623–3634. Knoetgen, H., Teichmann, U., Wittler, L., Viebahn, C. and Kessel, M. (2000) Anterior neural induction by nodes from rabbits and mice. Dev Biol 225:370–380. Kramer-Zucker, A.G., Olale, F., Haycraft, C.J., Yoder, B.K., Schier, A.F. and Drummond, I.A. (2005) Cilia-driven fluid flow in the zebrafish pronephros, brain and Kupffer’s vesicle is required for normal organogenesis. Development 132: 1907–1921. Krebs, L.T., Iwai, N., Nonaka, S., Welsh, I.C., Lan, Y., Jiang, R., Saijoh, Y., O’Brien, T.P., Hamada, H. and Gridley, T. (2003) Notch signaling regulates left–right asymmetry determination by inducing Nodal expression. Genes Dev 17:1207–1212. Levin, M., Johnson, R.L., Stern, C.D., Kuehn, M. and Tabin, C. (1995) A molecular pathway determining left–right asymmetry in chick embryogenesis. Cell 82:803–814. Long, S., Ahmad, N. and Rebagliati, M. (2003) The zebrafish nodal-related gene southpaw is required for visceral and diencephalic left–right asymmetry. Development 130:2303–2316. Lowe, L.A., Supp, D.M., Sampath, K., Yokoyama, T., Wright, C.V., Potter, S.S., Overbeek, P. and Kuehn, M.R. (1996) Conserved left–right asymmetry of nodal expression and alterations in murine situs inversus. Nature 381:158–161. Mangold, O. (1933) U¨ber die Induktionsfa¨higkeit der verschiedenen Bezirke der Neurulavon Urodelen. Wilhelm Roux’ Arch Entwicklungsmech 21:761–766. McGrath, J., Somlo, S., Makova, S., Tian, X. and Brueckner, M. (2003) Two populations of node monocilia initiate left–right asymmetry in the mouse. Cell 114:61–73. Melby, A.E., Warga, R.M. and Kimmel, C.B. (1996) Specification of cell fates at the dorsal margin of the zebrafish gastrula. Development 122:2225–2237.

146 Murcia, N.S., Richards, W.G., Yoder, B.K., Mucenski, M.L., Dunlap, J.R. and Woychik, R.P. (2000) The Oak Ridge Polycystic Kidney (orpk) disease gene is required for left–right axis determination. Development 127:2347–2355. Niehrs, C. (2004) Regionally specific induction by the Spemann– Mangold organizer. Nat Rev Genet 5:425–434. Nonaka, S., Shiratori, H., Saijoh, Y. and Hamada, H. (2002) Determination of left–right patterning of the mouse embryo by artificial nodal flow. Nature 418:96–99. Nonaka, S., Tanaka, Y., Okada, Y., Takeda, S., Harada, A., Kanai, Y., Kido, M. and Hirokawa, N. (1998) Randomization of left–right asymmetry due to loss of nodal cilia generating leftward flow of extraembryonic fluid in mice lacking KIF3B motor protein. Cell 95:829–837. Okada, Y., Nonaka, S., Tanaka, Y., Saijoh, Y., Hamada, H. and Hirokawa, N. (1999) Abnormal nodal flow precedes situs inversus in iv and inv mice. Mol Cell 4:459–468. Okada, Y., Takeda, S., Tanaka, Y., Belmonte, J.C. and Hirokawa, N. (2005) Mechanism of nodal flow: a conserved symmetry breaking event in left–right axis determination. Cell 121:633–644. Oppenheimer, J.M. (1936) Processes of localization in developing Fundulus. J Exp Biol 73:405–444. Pennekamp, P., Karcher, C., Fischer, A., Schweickert, A., Skryabin, B., Horst, J., Blum, M. and Dworniczak, B. (2002) The ion channel polycystin-2 is required for left–right axis determination in mice. Curr Biol 12:938–943. Przemeck, G.K., Heinzmann, U., Beckers, J. and Hrabe de Angelis, M. (2003) Node and midline defects are associated with left–right development in Delta1 mutant embryos. Development 130:3–13. Raya, A. and Belmonte, J.C. (2006) Left–right asymmetry in the vertebrate embryo: from early information to higher-level integration. Nat Rev Genet 7:283–293. Robb, L. and Tam, P.P. (2004) Gastrula organiser and embryonic patterning in the mouse. Semin Cell Dev Biol 15:543–554. Schier, A.F. and Talbot, W.S. (1998) The zebrafish organizer. Curr Opin Genet Dev 8:464–471. Schulte-Merker, S., Hammerschmidt, M., Beuchle, D., Cho, K.W., De Robertis, E.M. and Nusslein-Volhard, C. (1994) Expression of zebrafish goosecoid and no tail gene products in wild-type and mutant no tail embryos. Development 120:843–852. Shook, D.R., Majer, C. and Keller, R. (2004) Pattern and morphogenesis of presumptive superficial mesoderm in two closely related species, Xenopus laevis and Xenopus tropicalis. Dev Biol 270:163–185.

Spemann, H. (1931) U¨ber den Anteil von Implantat und Wirtskeim an der Orientierung und Beschaffenheit der induzierten Embryonalanlagen. Wilhelm Roux’ Arch Entwicklungsmech 123:389–517. Spemann, H. and Mangold, H. (1924) U¨ber Induktion von Embryonalanlagen durch Implantation artfremder Organisatoren. Arch Mikrosk Anat Entwicklungsmech 100:599–638. Sulik, K., Dehart, D.B., Iangaki, T., Carson, J.L., Vrablic, T., Gesteland, K. and Schoenwolf, G.C. (1994) Morphogenesis of the murine node and notochordal plate. Dev Dyn 201: 260–278. Supp, D.M., Witte, D.P., Potter, S.S. and Brueckner, M. (1997) Mutation of an axonemal dynein affects left–right asymmetry in inversus viscerum mice. Nature 389:963–966. Tabin, C.J. and Vogan, K.J. (2003) A two-cilia model for vertebrate left–right axis specification. Genes Dev 17:1–6. Tam, P.P., Kanai-Azuma, M. and Kanai, Y. (2003) Early endoderm development in vertebrates: lineage differentiation and morphogenetic function. Curr Opin Genet Dev 13: 393–400. Tam, P.P. and Steiner, K.A. (1999) Anterior patterning by synergistic activity of the early gastrula organizer and the anterior germ layer tissues of the mouse embryo. Development 126: 5171–5179. Theiler, K. (1972) The house mouse. Springer, Berlin Heidelberg New York. Viebahn, C., Mayer, B. and Miething, A. (1995) Morphology of incipient mesoderm formation in the rabbit embryo: a light- and retrospective electron-microscopic study. Acta Anat (Basel) 154:99–110. Viebahn, C., Stortz, C., Mitchell, S.A. and Blum, M. (2002) Low proliferative and high migratory activity in the area of Brachyury expressing mesoderm progenitor cells in the gastrulating rabbit embryo. Development 129:2355–2365. Waddington, C.H. (1933) Induction by the primitive streak and its derivatives in the chick. J Exp Biol 10:38–46. Waddington, C.H. and Schmidt, G.A. (1933) Induction by heteroplastic grafts of the primitive streak in birds. Wilhelm Roux’ Arch Entwicklungsmech 128:522–563. Wilkinson, D.G., Bhatt, S. and Herrmann, B.G. (1990) Expression pattern of the mouse T gene and its role in mesoderm formation. Nature 343:657–659. Zhou, X., Sasaki, H., Lowe, L., Hogan, B.L. and Kuehn, M.R. (1993) Nodal is a novel TGF-beta-like gene expressed in the mouse node during gastrulation. Nature 361:543–547.