Oral–aboral patterning and gastrulation of sea urchin embryos depend on sulfated glycosaminoglycans

MECHANISMS OF DEVELOPMENT

1 2 8 ( 2 0 1 1 ) 7 1 –8 9

available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/modo

Oral–aboral patterning and gastrulation of sea urchin embryos depend on sulfated glycosaminoglycans Karl-Frederik Bergeron *, Xing Xu, Bruce P. Brandhorst Molecular Biology and Biochemistry Department, Simon Fraser University, 8888 University Drive, Burnaby, British Columbia, Canada V5A 1S6

A R T I C L E I N F O

A B S T R A C T

Article history:

Glycosaminoglycans (GAGs) are a heavily sulfated component of the extracellular matrix

Received 30 June 2010

(ECM) implicated in a variety of cell signaling events involved in patterning of embryos.

Received in revised form

Embryos of the sea urchin Strongylocentrotus purpuratus were exposed to several inhibitors

30 October 2010

that disrupt GAG function during development. Treatment with chlorate, a general inhibi-

Accepted 1 November 2010

tor of sulfation that leads to undersulfated GAGs, reduced sulfation of the urchin blastocoe-

Available online 5 November 2010

lar ECM. It also prevented correct specification of the oral–aboral axis and mouth formation, resulting in a radialized phenotype characterized by the lack of an oral field,

Keywords: Secondary axis Oral–aboral Gastrulation TGF-beta Nodal

incomplete gastrulation and formation of multiple skeletal spicule rudiments. Oral markers were initially expressed in most of the prospective ectoderm of chlorate-treated early blastulae, but then declined as aboral markers became expressed throughout most of the ectoderm. Nodal expression in the presumptive oral field is necessary and sufficient to specify the oral–aboral axis in urchins. Several lines of evidence suggest a deregulation of Nodal signaling is involved in the radialization caused by chlorate: (1) Radial embryos resemble those in which Nodal expression was knocked down. (2) Chlorate disrupted local-

Chlorate Sulfation Glycosaminoglycan Extracellular matrix

ized nodal expression in oral ectoderm, even when applied after the oral–aboral axis is specified and expression of other oral markers is resistant to treatment. (3) Inhibition with SB-431542 of ALK-4/5/7 receptors that mediate Nodal signaling causes defects in ectodermal patterning similar to those caused by chlorate. (4) Intriguingly, treatment of embryos with a sub-threshold dose of SB-431542 rescued the radialization caused by low concentrations of chlorate. Our results indicate important roles for sulfated GAGs in Nodal signaling and oral–aboral axial patterning, and in the cellular processes necessary for archenteron extension and mouth formation during gastrulation. We propose that interaction of the Nodal ligand with sulfated GAGs limits its diffusion, and is required to specify an oral field in the urchin embryo and organize the oral–aboral axis.  2010 Elsevier Ireland Ltd. All rights reserved.

1.

Introduction

Establishment of the oral–aboral (OA) axis, aka ventro-dorsal axis, in the sea urchin embryo relies on transforming

growth factor-beta (TGF-beta) signaling events. Nodal is produced in the presumptive oral ectoderm of the early blastula embryo and has a key role necessary for establishment of the entire OA axis (Duboc et al., 2004; Flowers et al., 2004; Yaguchi

* Corresponding author. Tel.: +1 778 782 3021; fax: +1 778 782 5583. E-mail addresses: [email protected] (K.-F. Bergeron), [email protected] (X. Xu), [email protected] (B.P. Brandhorst). Abbreviations: ALK, activin-like kinase; BMP, bone morphogenetic protein; ClO3, chlorate; CS, chondroitin sulfate; DS, dermatan sulfate; ECM, extracellular matrix; GAG, glycosaminoglycan; HS, heparan sulfate; LSSW, low sulfate sea water; pNPX, 4-nitrophenyl-beta0 0 D-xylopyranoside; OA, oral–aboral; PAPS, 3 -phosphoadenosine 5 -phosphosulfate; SW, sea water; SeO4, selenate; SO4, sulfate; TGF, transforming growth factor 0925-4773/$ - see front matter  2010 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.mod.2010.11.001

72

MECHANISMS OF DEVELOPMENT

et al., 2010). Nodal signaling activity promotes nodal expression as well as the expression of downstream oral-specific patterning genes, e.g., lefty (aka antivin) and bmp2/4. The Nodal antagonist Lefty limits the spread of Nodal signaling activity beyond the border of the oral ectoderm territory (Duboc et al., 2004, 2008; Bolouri and Davidson, 2009) while BMP2/4 diffuses from this territory to specify the aboral ectoderm (Angerer et al., 2000; Duboc et al., 2004; Lapraz et al., 2009). Localized expression of nodal is necessary and sufficient for OA patterning (Duboc et al., 2004) and activates a gene regulatory network involved in OA specification (Su et al., 2009). Initiation of OA secondary axis specification is coordinated with animal–vegetal primary axis patterning through the regulation of transcription factor FoxQ2, a repressor of nodal expression (Yaguchi et al., 2008). Models of normal OA patterning were summarized by Duboc et al. (2004, Fig. 7; 2010, Fig. 9). Signaling events that pattern embryos take place in the extracellular space, in the fibrous mesh of the extracellular matrix (ECM) that surrounds and supports cells. The ECM is composed of proteins (mainly collagens) and glycosaminoglycans (GAGs), carbohydrate polymers that are usually attached to extracellular core proteins to form proteoglycans. For most GAGs, substitution with N- and O-linked sulfates and other groups results in highly modified, negatively charged disaccharide chains. The pattern of these modifications is thought to dictate the binding affinity of GAGs for specific signaling ligands, though a paradigm for GAG–cytokine interactions has yet to be identified (Mulloy and Rider, 2006). GAGs, heparan sulfate in particular, play roles as co-receptors for some cytokines, including TGF-beta superfamily ligands (Bernfield et al., 1999; Irie et al., 2003). GAGs are also thought to be important for the localization and stability of cytokines, acting as a repository and mediator of morphogen gradient formation along epithelia during development (Mulloy and Rider, 2006; Rozario and DeSimone, 2010). Sulfate is the most abundant anion in sea water after chloride, present at about 25 mM (Lyman and Fleming, 1940). It is an essential component of the defined culture medium (artificial sea water) for normal urchin development (Runnstro¨m et al., 1964; Karp and Solursh, 1974). PAPS (3 0 -phosphoadenosine 5 0 -phosphosulfate) is the universal sulfonate donor compound for all sulfotransferase reactions in the cell (Venkatachalam, 2003). Hence, PAPS biosynthesis is the limiting step in GAG sulfation (Sugahara and Schwartz, 1979). The sulfate analogs selenate and chlorate (ClO3) are competitive inhibitors of PAPS synthase (Baeuerle and Huttner, 1986; Ullrich and Huber, 2001). Selenate and ClO3 treatment are thought to principally interfere with GAG modification as these polymers bear the most sulfated groups (Wilson and Bandurski, 1958; Greve et al., 1988; Humphries and Silbert, 1988). In contrast, beta-xylopyranosides inhibit the attachment of GAG chains to proteoglycan core proteins resulting in the synthesis of free GAG chains and GAG-depleted proteoglycans (Carey, 1991; Tang et al., 1983). Various treatments are known to disrupt urchin OA patterning (e.g., Ho¨rstadius, 1973; Hardin et al., 1992; Akasaka et al., 1997; Flowers et al., 2004; Agca et al., 2009) but the molecular mechanisms behind these effects are poorly understood. In this study, we tested the role of sulfated GAGs, and by extension proteoglycans, in OA axis patterning. We

1 2 8 ( 2 0 1 1 ) 7 1 –8 9

treated Strongylocentrotus purpuratus embryos with GAG inhibitors in order to prevent normal GAG function during early development. These treatments caused defects in archenteron elongation and OA patterning. We focused on the general sulfation inhibitor ClO3 because of its specificity to OA patterning at low concentrations. ClO3 treatment led to a radial phenotype lacking an oral field. We characterized this phenotype by assessing gene and protein expression and cellular signaling events. Several lines of evidence indicate necessary roles for sulfated GAGs in Nodal signaling and OA axial specification. We propose that interaction of the Nodal ligand with sulfated GAGs in the ECM limits its diffusion, and is required to specify an oral field in the urchin embryo and organize the OA axis. Our results also suggest that archenteron extension and mouth formation during gastrulation are dependent on GAG sulfation.

2.

Results

2.1. Inhibitors of sulfation and GAG attachment to proteoglycans cause defects in oral–aboral patterning and gastrulation To investigate the role of sulfation during embryogenesis, S. purpuratus embryos were treated with the sulfation inhibitor ClO3. Consistent defects in development were observed in embryos continuously treated from 2 h post-fertilization (hpf) with 3–30 mM ClO3 in sea water. Archenteron extension was delayed and development arrested at the mid-to-late gastrula stage (Fig. 1B, B 0 , and C). No mouth or stomodeum were formed, while 5–10 rudimentary triradiate spicules were observed in a radial pattern around the equator of the blastocoel (Fig. 1B 0 ). At low ClO3 concentration (3 mM) the archenteron fully extended across the blastocoel and differentiated into distinct compartments, but failed to bend toward and fuse with the prospective oral ectoderm to form an oral opening (Fig. 1B); these arrested radial gastrulae displayed no OA or bilateral asymmetry, and possessed thick cuboidal ectoderm at the animal pole and thin squamous ectoderm in the vegetal half. Treated embryos later developed pigment cells distributed throughout the ectoderm (Supplementary Fig. S1). This radial phenotype is reminiscent of embryos in which Nodal activity has been blocked (Flowers et al., 2004; Duboc et al., 2004, 2008). In embryos treated with 30 mM ClO3 the archenteron extended to an average of 58% of the blastocoel diameter upon arrest (Fig. 1C). Mesenchyme differentiation was severely delayed in these embryos, but they later developed some pigment cells and short misshaped spicules (not shown). Urchin embryos treated with higher than 30 mM ClO3 arrested as morulae; these concentrations of ClO3 are detrimental to mammalian cell growth and viability (Humphries and Silbert, 1988). Embryos treated with ClO3 starting from the moment of fertilization elevated fertilization envelopes and cleaved normally but hatching was impaired. Thus, all treatments with ClO3 were begun 2 hpf or later. Selenate (SeO4) is another inhibitor of sulfation (Wilson and Bandurski, 1958). Treatment of S. purpuratus embryos with 3 mM SeO4 caused a defect in archenteron elongation

MECHANISMS OF DEVELOPMENT

1 2 8 ( 2 0 1 1 ) 7 1 –8 9

73

Fig. 1 – Gastrula-arrested embryos treated with inhibitors of sulfation, GAG attachment and Nodal receptor activity. Embryos were continuously treated with various inhibitors starting at 2 hpf unless otherwise noted (@ 24 hpf; panels B00 and F00 ) in the lower right of the panel. Panel A, untreated (UT); B, 3 mM ClO3; C, 30 mM ClO3; D, 3 mM SeO4; E, 1 mM pNPX; F, 5 lM SB-431542 (SB). Pictures were taken 96 hpf. Most of the embryos are viewed from the side, but some vegetal views are also provided (A 0 , B 0 and F 0 ). Whenever it was recognizable, the oral side of embryos was placed towards the right.

and mid-gastrula arrest similar to embryos treated with 30 mM ClO3 (Fig. 1D); similar results have been reported previously (Sugiyama, 1972; Khurrum et al., 2004). SeO4-treated gastrulae displayed mesenchyme-like material in their blastocoels, but lacked pigment cells and spicules (not shown), suggesting additional effects of SeO4 on mesenchyme specification and/or differentiation. ClO3 treatment is thought to primarily interfere with sulfation of GAGs and, by extension, proteoglycans (Greve et al., 1988; Humphries and Silbert, 1988). We exposed urchin embryos to a beta-xylopyranoside in order to interfere with the synthesis of proteoglycans. Exogenous beta-xylosides compete as primers with the endogenous proteoglycan core proteins for galactosyltransferase I, an enzyme that participates in the synthesis of GAGs. This treatment results in the synthesis of free GAG chains and GAG-depleted proteoglycan core proteins (Carey, 1991). Treatment with several betaxylosides leads to a developmental arrest at the mesenchyme blastula stage in various urchin species, including S. purpuratus (Kinoshita and Saiga, 1979; Akasaka et al., 1980; Solursh et al., 1986), while lower doses gives rise to radialized gastrulae possessing multiple rudimentary spicules in some species (Kinoshita and Saiga, 1979). S. purpuratus embryos treated with 1 mM 4-nitrophenyl-beta-D-xylopyranoside (pNPX) starting at 2 hpf failed to complete gastrulation, possessed mesenchymelike material in their blastocoel (Fig. 1E), formed multiple small spicule rudiments in a radial pattern, and lacked pigment cells (not shown). Except for the lack of pigment cells, treatment with pNPX caused defects similar to those observed for embryos treated with ClO3, suggesting that ClO3 interferes with proteoglycan function via inhibition of sulfation of GAGs. Treatment with inhibitors of sulfation and GAG attachment led to similar mid-gastrula arrest phenotypes, suggesting that sulfated GAGs are necessary for the convergent-extension cell movements of archenteron elongation. Treatment with lower concentrations of the sulfation inhibitor ClO3 led to milder

phenotypes mostly involving OA ectoderm patterning and/or differentiation. The multiple defects observed suggest roles for sulfation in a number of different developmental processes. We focused our attention on 3 mM ClO3 treatment because of its consistent radialization effects while causing minimal mesenchyme and archenteron elongation defects compared to higher ClO3 concentrations and other inhibitors.

2.2.

Undersulfation causes the ClO3 radial phenotype

In order to directly visualize sulfation events, embryos were stained with Alcian Blue under conditions specific for sulfated groups (Pearse, 1960; Bjornsson, 1993). Gastrula embryos displayed uniform staining of the blastocoel (Fig. 2A). ClO3 treatment drastically diminished Alcian Blue staining in a concentration-dependent manner. In embryos treated with 30 mM ClO3, only the lumen of the archenteron remained stained, suggesting this specific sulfated material is extremely resistant to ClO3. Gastrulae arrested by treatment with 3 mM ClO3 showed intermediate staining of the blastocoel compared to control (Fig. 2A). Some blastocoelar proteins and proteins of the gut lining, including cellassociated proteoglycans, are membrane proteins. Membrane preparations from whole embryos were blotted onto a PVDF membrane and stained with Alcian Blue as described by Bjornsson (1998). Sulfation of membrane preparations was reduced in a dose-dependent manner by ClO3 treatment (Fig. 2B). If some of the ClO3-resistant sulfated material in the gut lumen is membrane bound, Alcian Blue staining on the dot blot is likely to be an overestimation of how much blastocoelar membrane associated sulfate remains in ClO3-treated embryos. To confirm that ClO3 treatment radializes embryos through inhibition of sulfation events, we cultured embryos in low sulfate sea water (LSSW) containing roughly 1/3 of

74

MECHANISMS OF DEVELOPMENT

1 2 8 ( 2 0 1 1 ) 7 1 –8 9

Fig. 2 – Staining for sulfate in whole embryos and membrane preparations following treatment with ClO3. Alcian Blue, a dye that binds specifically to sulfate groups in acidic conditions (pH 1.0), was used to stain whole gastrula embryos and membrane proteins. (A) Embryos were treated with 0, 3 and 30 mM ClO3 beginning at 24 hpf and collected at 72 hpf. Stained embryos were imaged using bright field optics. (B) Embryos were treated with 0, 3, 10 and 30 mM ClO3 beginning at 2 hpf and collected at 48 hpf (late gastrula stage). Membrane preparations (10 lg of protein) from 48 hpf embryos were dot blotted onto a PVDF filter that was then stained with Alcian Blue (detecting sulfate) or Coomassie Blue (detecting proteins).

the normal concentration of sulfate (SO4; 10 mM). These embryos were particularly sensitive to ClO3 treatment: 1 mM ClO3 treatment was sufficient to radialize nearly all embryos (instead of 3 mM in normal sea water). Furthermore, complementing the LSSW culture media with SO4 to the concentration of normal sea water rescued the radialization of embryos treated with 1 mM ClO3 (Fig. 3). However, this rescue was not complete as a mouth was not usually formed in these embryos. Increasing concentrations of SO4 above 50 mM caused early developmental perturbations and thus could not be used to attempt a rescue embryos reared in normal sea water and treated with 3 mM of ClO3 (not shown). We conclude that ClO3 inhibits sulfation in urchin embryos and that undersulfation is the cause of the radial phenotype.

2.3. Embryos are sensitive to ClO3 during oral–aboral axis specification To explore the developmental mechanism(s) by which undersulfation causes morphological defects, we determined the timing of embryos sensitivity to treatment with 3 mM ClO3 (Fig. 4). Gastrulation and OA patterning of embryos were disrupted in nearly all embryos if treatment began at or before mid-blastula stage (18 hpf). Embryos between 18 hpf (mid-blastula) and 36 hpf (early gastrula stage) showed increasing resistance to treatment. Many embryos treated beginning at 18 hpf arrested at prism stage and displayed mouth defects, while treatment beginning at 24 hpf and later yielded increasing fractions of embryos that formed normal plutei. When ClO3 treatment was initiated at the late gastrula stage (42 hpf) or later, more than 70% of the embryos gastrulated and developed normally into plutei (Fig. 4). Embryos were most sensitive to ClO3 before gastrulation began.

Specification of prospective oral and aboral ectoderm is thought begin around the sixth cleavage, after the founder cells for oral and aboral ectoderm lineages have formed (Cameron et al., 1990). Cell signaling is central to the OA specification process (Duboc et al., 2004; Flowers et al., 2004; Lapraz et al., 2009; Yaguchi et al., 2010). However, the presumptive OA axis is labile and commitment of cells to a particular fate along this axis does not occur until the onset of gastrulation (Hardin et al., 1992; Hardin and Armstrong, 1997; Flowers et al., 2004; Agca et al., 2009). Therefore, the ClO3 sensitivity period coincides roughly with the timing of OA specification during blastula stages.

2.4. Treatment with ClO3 or an inhibitor of Nodal signaling causes similar defects ClO3-treated arrested radial gastrulae are reminiscent of embryos in which Nodal signaling has been reduced by knocking down translation of nodal mRNA or overexpressing the Nodal antagonist Antivin/Lefty. These treatments lead to arrested late gastrulae having multiple spicule rudiments, a straight archenteron, and excess pigment cells (Flowers et al., 2004; Duboc et al., 2004, 2008). We compared ClO3treated embryos with embryos in which Nodal activity was inhibited by SB-431542. This small compound reduces the kinase activity of Activin receptor-like kinase (ALK)-4/5/7 receptors for TGF-betas, including Nodal and Univin (Inman et al., 2002; Range et al., 2007). SB-431542-treated embryos showed parallels with ClO3-treated embryos, with a radialized late gastrula arrest phenotype and 5–6 spicule rudiments. In contrast to ClO3-treated embryos however, SB-431542-treated embryos often displayed a conical shape with thick cuboidal ectoderm in the animal half and their guts displayed more

MECHANISMS OF DEVELOPMENT

Fig. 3 – Rescue of ClO3-treated embryos by sulfate. Embryos were reared in low sulfate (10 mM SO4) artificial SW and treated with 0, 1 and 3 mM ClO3, with or without a supplement of 20 mM SO4. Pictures were taken 72 hpf. Arrowheads point to the archenteron tip or mouth opening.

differentiated compartments (Fig. 1F and F 0 ). A similar phenotype has been reported for SB-431542-treated Paracentrotus lividus urchin embryos (Range et al., 2007). In an attempt to distinguish between OA specification and differentiation processes, we began inhibitor treatments at late blastula stage (24 hpf), when specification of the oral and aboral ectoderm is already under way (Angerer and Davidson, 1984) but OA ectoderm tissues are not yet distinguishable by morphology. Most embryos treated with either ClO3 or SB-431542 at 24 hpf failed to form a mouth and arrested as prisms with mouth defects. Though the archenteron bent toward the thickened, cuboidal presumptive oral ectoderm, and two bilaterally symmetrical spicules were usually observed, there was no stomodeal invagination and no tissue fusion between the archenteron tip and overlying ectoderm of the blastocoelar wall (Fig. 1B00 and F00 ). ClO3-treated embryos were more spherical, while gut compartments were more differentiated and spicule rudiments were larger in SB-432542treated embryos. We observed three distinct effects of ClO3 treatment on embryos: inhibition of archenteron extension (with 30 mM ClO3; Fig. 1C), disruption of the OA secondary

1 2 8 ( 2 0 1 1 ) 7 1 –8 9

75

Fig. 4 – Time course of developmental sensitivity of embryos to ClO3 treatment. Embryos were treated with 3 mM ClO3 beginning at different times post-fertilization and their phenotypes, scored at 96 hpf, were plotted as percentages of total embryos. Treatment times were roughly equivalent to the following developmental stages: 2 hpf, 2-cell (2C); 6 hpf, 16-cell (16C); 12 hpf, early blastula (EB); 18 hpf, hatched blastula (HB); 24 hpf, mesenchyme blastula (MB); 30 hpf, flattened blastula (FB); 36 hpf, early gastrula (EG); 42 hpf, late gastrula (LG); 48 hpf, prism (PR); 72 hpf, pluteus (PL). The number of embryos scored (n) is included for each treatment. The ‘‘Other/Abnormal’’ category comprises all phenotypes observed that did not fall into the ‘‘Early Pluteus’’, ‘‘Prism with Mouth Defect’’ or ‘‘Radial Gastrula’’ categories described in Results.

axis, and interference with mouth formation. The last two effects were phenocopied by exposure to SB-431542. Embryos treated with either ClO3 or SB-431542 had mesenchyme cells within their blastocoel (Fig. 1B and F). Some nonpigmented mesenchyme cells expressing the SP1 epitope were found scattered throughout the blastocoel of ClO3- and SB-431542-treated late embryos (144 hpf; Fig. 5B 0 and B00 ). This is reminiscent of early stages of pigment cell development (Gibson and Burke, 1985). This also suggests a delay or defect in aboral mesenchyme and/or ectoderm differentiation, as presumptive pigment cells are normally restricted to, and embedded in, aboral ectoderm by the completion of gastrulation (48 hpf in S. purpuratus). Immunostaining for Spec1, an early marker of aboral ectoderm specification (Angerer and Davidson, 1984; Klein et al., 1990), showed that embryos treated with ClO3 or SB-431542 express it in a large radialized zone of the ectoderm, with most intense staining in the vegetal half. Sharp boundaries of differential Spec1 protein expression between oral and aboral ectoderm were absent in treated gastrulae, but staining was gradually lost towards the animal pole (Fig. 5A 0 , A00 , B 0 , and B00 ). ClO3-treated embryos did not express the archenteron-specific Endo1 epitope at 96 hpf, but it was expressed at 144 hpf (Fig. 5A 0 and A00 ), indicating a delay in endoderm differentiation. Cells immunostained with a monoclonal antibody against serotonin (Fig. 5C, C 0 , and C00 )

76

MECHANISMS OF DEVELOPMENT

1 2 8 ( 2 0 1 1 ) 7 1 –8 9

Fig. 5 – Ectoderm, gut, pigment cell, and neuron formation in embryos treated with ClO3 or SB-431542. Single confocal slices of 144 hpf embryos untreated (UT), or continuously treated with 3 mM ClO3 or 5 lM SB-431542 (SB) (as shown in the lower left of each panel) beginning at 2 hpf, and stained with various antibodies as noted in the top left of each panel: A, anti-Spec1 (magenta) and anti-Endo1 (green) double staining; B, anti-Spec1 (magenta) and SP1 (green) double staining; C, anti-serotonin (magenta) staining and non-specific tissue autofluorescence (green). The animal pole of embryos is at the top. Whenever possible (for untreated embryos) the oral side is towards the right (A and B) or facing the reader (C).

were restricted to the apical organ of normal larva, and were found at the animal pole of ClO3- and SB-431542-treated embryos, indicating that neurogenesis occurs in the appropriate domain. Differentiation of pigment cells, aboral ectoderm and, to a lesser extent, endoderm tissue were similarly perturbed in ClO3- and SB-431542-treated embryos. Moreover, a similar range of phenotypes (radial late gastrulae and prisms with mouth defects) was obtained with both inhibitors depending on the timing of treatment. The sensitivity period to SB431542 closely paralleled that of ClO3 but the transition between all radial gastrulae and mostly normal plutei was sharper (Supplementary Fig. S2). These results suggest ClO3 interferes with the same processes as SB-431542 and raise the possibility that Nodal signaling is perturbed when GAG sulfation is inhibited by ClO3 treatment.

2.5. ClO3 downregulates the expression of genes involved in oral–aboral patterning ClO3 treatment disrupted bilateral spiculogenesis and mouth formation, processes that depend on oral ectoderm differentiation (Guss and Ettensohn, 1997; Hardin and Armstrong, 1997). Moreover, treatment with the TGF-beta receptor inhibitor SB-431542 led to similar abnormalities.

Because the TGF-beta ligand Nodal and its antagonist Lefty have critical roles in sea urchin OA patterning (Duboc et al., 2004, 2008; Flowers et al., 2004; Yaguchi et al., 2010), we assessed their mRNA expression levels throughout embryogenesis using quantitative real-time polymerase chain reaction (qRT-PCR). As previously reported (Duboc et al., 2004), a decline in nodal mRNA (24 hpf) occurred after its initial peak of expression (18 hpf), and was followed by a second, lower peak of expression persisting through gastrulation (48 hpf) during normal development (Fig. 6A). In ClO3-treated embryos, the initial peak of nodal expression was observed on schedule, but at levels reduced 2- to 4-fold (18 and 24 hpf, respectively). The second peak of nodal RNA was also observed on time but then fell to much lower levels of expression toward the end of gastrulation (48 hpf; Fig. 6A). The pattern of expression of lefty RNA paralleled nodal (Fig. 6B) although its expression was up to 9 times higher, suggesting the Lefty antagonist protein is also found at higher levels than Nodal in the urchin embryo. z12-1, a zinc finger transcription factor, was used as a reference gene for each qRT-PCR. Northern blot analysis at various times during development (18 and 48 hpf) showed that z12-1 RNA levels were unaffected by treatment with up to 30 mM ClO3 (Supplementary Fig. S3). Maintenance of nodal and lefty expression after their temporal expression peaks (18–24 hpf and 42–48–72 hpf; Fig. 6A and B) was

MECHANISMS OF DEVELOPMENT

1 2 8 ( 2 0 1 1 ) 7 1 –8 9

77

Fig. 6 – Quantification of OA marker mRNA expression levels in ClO3-treated embryos. Gene expression was quantified by two-step qRT-PCR on RNA extracted at various times during development from embryos untreated or treated with 3 mM ClO3 beginning at 2 hpf. Genes assessed included the TGF-beta family cytokines nodal (panel A) and lefty (B). Included tables show the mean [Cttarget–Ctreference] for each treatment and timepoint. Panel C shows the relative quantity of various oral (nodal, lefty, gsc, bmp2/4) and aboral (tbx2/3, cyIIIa, spec1) ectoderm markers at 24 hpf. Times at which RNA was extracted were roughly equivalent to the following developmental stages: 12 hpf, early blastula (EB); 18 hpf, hatched blastula (HB); 24 hpf, mesenchyme blastula (MB); 30 hpf, flattened blastula (FB); 36 hpf, early gastrula (EG); 42 hpf, late gastrula (LG); 48 hpf, prism (PR); 72 hpf, pluteus (PL).

undermined in ClO3-treated embryos. As their continued expression relies on a nodal positive feedback loop (Nam et al., 2007; Range et al., 2007; Bolouri and Davidson, 2009), our results imply that this autoregulatory mechanism is defective when the ECM is undersulfated.

The biggest reduction in nodal and lefty RNA in ClO3-treated embryos compared to untreated controls occurred in mesenchyme blastulae (24 hpf), prior to their second peaks of expression. Expression of additional oral ectoderm markers (gsc, bmp2/4) as well as aboral ectoderm markers (tbx2/3,

78

MECHANISMS OF DEVELOPMENT

1 2 8 ( 2 0 1 1 ) 7 1 –8 9

Fig. 7 – Spatial expression of gene markers of OA ectoderm specification throughout early development of ClO3-treated embryos. Gene expression was analyzed by WMISH on embryos untreated (UT) or treated with 3 mM ClO3 (as shown in the lower left of each panel) beginning at 2 hpf. Embryos were fixed at the times noted in the lower right of each panel. For all panels, the animal pole is at the top and the presumptive oral side of each embryo is toward the right. The hybridization probe used is shown at the top right of each panel. Genes assessed included oral ectoderm markers nodal (nod) (panels A and B) and lefty (lty) (C and D), as well as the aboral ectoderm marker spec1 (ec) (E and F).

MECHANISMS OF DEVELOPMENT

1 2 8 ( 2 0 1 1 ) 7 1 –8 9

79

Fig. 8 – Spatial expression of markers of endodermal and ectodermal territories in embryos treated with ClO3. Gene expression was analyzed by WMISH on embryos untreated (UT) or treated with 3 or 30 mM ClO3 (as shown in the lower left of each panel) beginning at 2 hpf unless otherwise noted in the lower right of each panel. All embryos were fixed at 48 hpf. For all panels, the animal pole is at the top and the presumptive oral side of each embryo is toward the right. The hybridization probe used is shown at the top right of each panel. Genes assessed included oral ectoderm markers goosecoid (gsc) (panel A), lefty (lty) (E) and nodal (nod) (F); onecut/hnf6 (oc) expressed in the ciliated band (B); nk2.1 (nk) and foxq2 (q2) expressed in the animal plate (C and G, respectively); brachyury (bra) expressed in the vegetal plate and the stomodeum (D); and the endoderm marker gatae (gte) (H).

cyIIIa, spec1) was also reduced at 24 hpf (Fig. 6C), as expected from the perturbed expression of nodal normally necessary for the establishment of the entire OA axis (Duboc et al., 2004; Flowers et al., 2004).

2.6.

ClO3 disrupts oral–aboral patterning of the ectoderm

To assess the effect of undersulfation on OA patterning, we investigated the spatial expression of genes involved in OA specification in embryos treated with 3 mM ClO3 beginning at 2 hpf by whole mount in situ hybridization (WMISH). Nodal (nod) RNA was detected in prospective oral ectoderm cells of both treated and untreated 12 hpf early blastulae (Fig. 7A1 and B1), and was restricted to oral ectoderm through gastrulation in untreated embryos (Fig. 7A2–A5). By 18 hpf, the expression domain of nodal was expanded to include the entire animal half of the ectoderm of ClO3-treated embryos (Fig. 7B2 and B3). Between 24 and 36 hpf expression of nodal in ClO3-treated embryos became delocalized, with no specific subset of ectoderm cells stained (Fig. 7B4 and B5). Similar changes in expression pattern were observed for lefty (lty) (Fig. 7C and D). Goosecoid (gsc) RNA is normally restricted to oral ectoderm (Fig. 8A), but showed faint diffuse staining of all the ectoderm of late gastrula embryos (48 hpf; Fig. 8A 0 ). Brachyury (bra) RNA was not detected in prospective stomodeal ectoderm as in normal late gastrulae, while its vegetal ring of expression around the blastopore was enlarged (48 hpf; Fig. 8D and D 0 ; vegetal pole views in Supplementary

Fig. S5A). Expression of oral ectoderm markers was expanded in treated blastulae while no specific subset of cells expressed them in treated gastrulae. The expression of early markers of aboral ectoderm cyIIIa (Supplementary Fig. S4A–D) and spec1 (ec) was slightly delayed and expanded to include much of the prospective ectoderm with more intense staining near the vegetal pole (Fig. 7E and F; Supplementary Fig. S4I and J). Staining for onecut/hnf6 (oc) RNA, a marker of the proneural ciliated band that forms at the boundary of oral and aboral ectoderm (Otim et al., 2004), was concentrated at the animal pole (Fig. 8B and B 0 ). Expression of nk2.1 (nk), a marker of the apical plate neurogenic ectoderm (Takacs et al., 2004), was also concentrated at the animal pole of ClO3-treated gastrula embryos (Fig. 8C and C 0 ). Expression of aboral ectoderm markers was expanded to include most of the vegetal ectoderm in ClO3-treated embryos, blastulae and gastrulae alike, while ectoderm at the animal pole expressed neurogenic ectoderm markers, consistent with the immunostaining results (Fig. 5A 0 , B 0 , and C 0 ). Diffusible Nodal signaling mediated by Smad2/3 suppresses neural differentiation of ectoderm except in the animal plate (Yaguchi et al., 2006, 2007). To a large extent this suppression of neurogenesis, restricting it to the apical organ, appears to be operating in ClO3-treated embryos. The presence of paired triradiate spicules indicated that an OA axis and bilateral symmetry were maintained in some embryos treated with ClO3 beginning at the late mesenchyme blastula stage (treated at 24 hpf; Figs. 1B00 and 4). We assessed

80

MECHANISMS OF DEVELOPMENT

1 2 8 ( 2 0 1 1 ) 7 1 –8 9

Fig. 9 – Smad activation by Nodal or BMP2/4 signaling in blastula embryos following ClO3 treatment. Single confocal slices of 15 hpf, 18 and 24 hpf blastula embryos untreated (UT) and treated with 3 mM ClO3 beginning at 2 hpf, and stained with an anti-phospho-Smad3/Smad1 antibody. Smad2/3 phosphorylation downstream of Nodal signaling (15 and 18 hpf) was blocked by a 1 h treatment with 10 lM SB-431542 (1 h SB), whereas Smad1/5/8 activation downstream of BMP2/4 signaling (24 hpf) was not sensitive to this inhibitor. The presumptive oral side of each embryo is toward the right and the vegetal pole of 24 hpf embryos is toward the bottom. Arrowheads show the extent of nuclear staining. The fainter nuclear staining at 15 and 18 hpf compared to 24 hpf required longer exposure (resulting in higher background).

expression of OA ectoderm marker genes in these embryos to investigate ectoderm differentiation events. Expression patterns of spec1, nk2.1, onecut/hnf6, gsc (not shown), bra (Fig. 8D00 ) and lefty (Fig. 8E) transcripts were normal in embryos treated with 3 mM ClO3 beginning at 24 hpf. However, staining for nodal mRNA was weak and diffuse, detectable in few, if any, cells (Fig. 8F). Thus, the OA axis and bilateral symmetry, as well as proper expression of oral and aboral markers, appear to be specified in embryos before gastrulation in a way that is often resistant to ClO3 treatment, but nodal

expression and formation of the oral opening remain sensitive to ClO3 treatment until later in development. Expansion of the expression domain of OA ectoderm markers in ClO3-treated embryos is consistent with a loss of OA polarity. Interestingly, oral markers are first expressed in most of the prospective ectoderm of ClO3-treated early blastulae, but then decline while aboral markers become expressed throughout. This suggests that oral and aboral genes are transiently co-expressed in presumptive ectoderm cells of ClO3treated mesenchyme blastulae, an event that does not occur

MECHANISMS OF DEVELOPMENT

1 2 8 ( 2 0 1 1 ) 7 1 –8 9

81

embryos appear to be normally patterned along the AV axis but endoderm morphogenesis and/or differentiation are somewhat defective as some presumptive endoderm cells had not yet entered the archenteron by 48 hpf at all concentrations of ClO3 tested (Fig. 8H 0 and H00 ).

2.8.

Fig. 10 – Rescue of ClO3-treated embryos by partial inhibition of Nodal receptor activity. Embryos were treated with 0, 1.0, 1.5 and 2.0 mM ClO3 beginning at 2 hpf in the presence or absence of suboptimal doses of SB-431542 (SB; 0.5 lM), and their phenotypes, scored at 96 hpf, were plotted as percentages of total embryos. The number of embryos scored (n) is included for each treatment. The ‘‘Other/ Abnormal’’ category comprises all phenotypes observed that did not fall into the ‘‘Early Pluteus’’, ‘‘Prism with Mouth Defect’’ or ‘‘Radial Gastrula’’ categories described in Section 2.

in control embryos of the same stage. Our WMISH results are consistent with the suppression of ectoderm genes detected by qRT-PCR (Fig. 6); this may not be discernable from the figures provided as the images were selected from among the most intensely stained ClO3-treated embryos to illustrate spatial expression.

2.7. axis

ClO3 does not affect patterning of the animal–vegetal

To test whether ClO3 treatment is specific to OA patterning, or if it also affects other embryo signaling/patterning processes, we investigated patterning of the animal–vegetal (AV) axis in embryos treated with 3 mM ClO3 beginning at 2 hpf. Restriction of foxQ2 expression to the animal plate is crucial in coordinating AV patterning with the initiation of OA axis specification (Yaguchi et al., 2008). Expression of foxQ2 (q2) mRNA was normal in gastrula embryos (Fig. 8G and G 0 ), consistent with normal AV patterning. Gatae expression defines three compartments in the archenteron of control embryos (two zones of expression separated by a zone with little expression; Fig. 8H) that were observed in ClO3-treated embryos as well (Fig. 8H 0 and H00 ). Endo16 expression in the archenteron was similar in control and treated embryos (Supplementary Fig. S5D). However, in arrested gastrulae treated with ClO3, a ring of cells expressing gatae (Fig. 8H 0 and H00 ; vegetal pole views in Supplementary Fig. S5F) or endo16 (Supplementary Fig. S5E) surrounded the blastopore. This expanded expression correlates with the enlarged vegetal domain of bra expression reported above (Fig. 8D 0 ; vegetal pole views in Supplementary Fig. S5A). Tissues of ClO3-treated

ClO3 alters the patterns of Smad activation

Based on our results suggesting an effect of ClO3 on TGFbeta signaling, we examined effectors downstream of Nodal and BMP receptors. Use of an antibody against phosphoSmad3/Smad1 allowed us to observe both Nodal-dependent (Smad2/3) and BMP2/4-dependent (Smad1/5/8) activation of Smads in nuclei (Yaguchi et al., 2007). However, the visualization of phospho-Smad2/3 downstream of Nodal was impeded once BMP2/4 signaling had begun. After 21 hpf, staining of phospho-Smad1/5/8 overpowers the fainter staining of Nodal-dependent Smads. SB-431542, an inhibitor specific for the TGF-beta type I receptors including ALK-4/5/7 (Inman et al., 2002), is useful in distinguishing activation of Smad2/ 3 from Smad1/5/8: a 1 h exposure to this compound specifically extinguishes Smad2/3 activation. Early Smad2/3 phosphorylation (15 and 18 hpf; Fig. 9A, A 0 , C, and C 0 ) on the presumptive oral side of blastulae was sensitive to SB-431542 as expected (Yaguchi et al., 2007). ClO3 treatment induced an expansion of a weak Smad2/3 phosphorylation domain in 18 hpf blastulae (Fig. 9C 0 ), consistent with expanded but diluted Nodal signaling activity accompanying the delocalized expression of nodal. Intense, SB-431542resistant Smad1/5/8 phosphorylation staining was observed on the presumptive aboral side of untreated mesenchyme blastulae (24 hpf; Fig. 9A00 and B00 ) but expanded to the animal pole and in a few cells at the vegetal pole of ClO3-treated embryos (Fig. 9C00 and D00 ).

2.9. Partial inhibition of TGF-beta signaling rescues oral– aboral patterning in embryos treated with suboptimal concentrations of ClO3 As both ClO3 and SB-431542 interfere with Nodal signaling and lead to OA patterning defects, we tested for possible interactions between sub-threshold concentrations of these inhibitors (Fig. 10). OA patterning was disrupted in a fraction of embryos exposed to suboptimal concentrations (1.0– 2.0 mM) of ClO3 starting at 2 hpf but was rescued by simultaneous exposure to a very low concentration (0.5 lM) of SB-431542 (Fig. 10), a concentration that does not alter OA patterning by itself. This treatment presumably inhibited expansion of the domain of Nodal signaling activity caused by ClO3 treatment so that the OA territory was more correctly specified and maintained. Treatment with 0.5 lM SB-431542 did not rescue embryos treated with 3 mM ClO3, while higher SB-431542 concentrations caused some of these embryos to exogastrulate (not shown).

3.

Discussion

We examined the effect of inhibitors of sulfation GAG attachment on sea urchin embryo gastrulation and patterning

82

MECHANISMS OF DEVELOPMENT

1 2 8 ( 2 0 1 1 ) 7 1 –8 9

Fig. 11 – Model of the impact of GAG undersulfation on TGF-beta ligand diffusion/activity and embryo OA patterning. ClO3 treatment leads to two main predicted effects: (1) GAG undersulfation and concomitant expansion of Nodal ligand diffusion/ signaling later followed by (2) an extinguishment of Nodal activity. A mitochondria-based redox gradient in the egg leads to polarized expression of TGF-beta ligand Nodal in the untreated blastula. Nodal maintains its own expression (through a positive autoregulatory loop) and triggers the expression of Lefty and BMP2/4. For clarity, two intermediate blastulae stages are shown, the first showing the Lefty ligand diffusion pattern (but not the BMP2/4 ligand) and the second showing the BMP2/ 4 ligand (but not the Lefty ligand). Lefty diffuses more than the GAG-bound Nodal and antagonizes Nodal activity, effectively restricting its signaling to the presumptive oral field. FoxQ2 and beta-Catenin activity limit the range of animal and vegetal expression of Nodal, respectively. BMP2/4 is co-expressed with the TGF-beta ligands on the presumptive oral side of the blastula but diffuses across the blastocoel to specify aboral structures. Gastrula tissues differentiate into recognizable oral (Goosecoid) and aboral (Spec1) ectoderm, separated by a neurogenic ciliary band (Onecut/Hnf6). In ClO3-treated blastulae however, Nodal cannot bind to undersulfated GAGs and diffuses at least as readily as Lefty. The domain of Nodal signaling is initially expanded. BMP2/4 is expressed wherever Nodal signaling can trigger it. Perhaps because increased Nodal diffusion eventually leads to a dilution of its activity, Nodal expression/activity is lowered as gastrulation begins and most of the ectoderm is converted to aboral, expressing Spec1. The neurogenic ciliary band marker Onecut/Hnf6 is only expressed at the animal pole as no oral field, and therefore no oral/aboral boundary, forms.

along two orthogonal axes of symmetry: the primary AV axis and the secondary OA axis. All the inhibitors used led to defects in archenteron elongation and mouth formation yet did not affect AV patterning. Low concentrations of the broad sulfation inhibitor ClO3 led to defects mostly specific to the OA axis. We present a model in which restriction of Nodal signaling to the oral territory depends on sulfated GAGs/ proteoglycans (Fig. 11 and Section 3.2).

3.1. ClO3 treatment may impair the function of sulfated GAGs/proteoglycans of the ECM in cell signaling The observation that pNPX, SeO4 and ClO3 treatments can lead to almost identical radialized phenotypes (Fig. 1C, D, and E) suggests that sulfated GAG-decorated proteoglycans are the main functional component of the ECM that is being disrupted by our inhibitors. Indeed, GAGs

and proteoglycans are heavily sulfated constituents of the ECM that have been shown to be highly sensitive to ClO3 treatment (Greve et al., 1988; Humphries and Silbert, 1988). Cell-associated proteoglycans, present in membrane protein preparations, are particularly interesting candidates for having roles in OA patterning. These proteoglycans are known to play important roles in cell signaling by several ligands (including Wnt, FGF, Hh, PDGF/VEGF and TGF-betas) and in the establishment of morphogenetic gradients during development of several animals (Kramer and Yost, 2003; Lin, 2004; Hacker et al., 2005). The ECM can bind soluble/secreted factors, maintaining them in the extracellular space and thereby function as a repository. The consequence of such interactions may be to restrict or promote access of ligands to cognate cell-surface receptors, to modulate the spatial distribution of a diffusible morphogen, or to sequester and stabilize factors for subsequent release

MECHANISMS OF DEVELOPMENT

(Rozario and DeSimone, 2010). The OA patterning defects observed in our ClO3-treated radial embryos, combined with the central role of TGF-beta ligands in specification and patterning of the urchin OA axis (Duboc et al., 2004, 2008, 2010; Flowers et al., 2004; Lapraz et al., 2009; Yaguchi et al., 2010), suggests a necessary role for sulfated GAGs/proteoglycans of the ECM in maintaining the expression, stability, localization and/or activity of these ligands in the prospective oral field. In cell cultures, treatment with ClO3 is used for the production of GAGs with defined structural alterations (Safaiyan et al., 1999); sulfation of heparan sulfate is less reduced than that of chondroitin sulfate or the related GAG dermatan sulfate (Greve et al., 1988; Humphries and Silbert, 1988). These GAGs, probably in association with proteoglycan core proteins, have been shown to constitute the major sulfated macromolecules within the blastocoel and basement membranes of S. purpuratus embryos (Solursh and Katow, 1982), with dermatan sulfate being most prevalent during the mesenchyme blastulae to early gastrula stages (Vilela-Silva et al., 2001) when the OA axis is being determined. Interestingly, the TGF-beta ligand Nodal has been found to bind to chondroitin sulfate in vitro (Oki et al., 2007). Treatment with ClO3 may disrupt sulfation of dermatan and/or chondroitin within urchin embryos, thereby altering their possible interaction with TGF-beta ligands like Nodal.

3.2. A model for the regulation of Nodal diffusion through its affinity for sulfated GAGs Interaction of Nodal with sulfated GAGs in mouse embryos is suggested to facilitate ligand transport from its site of secretion and/or its stability (Oki et al., 2007). Diffusion of Dpp (a Drosophila TGF-beta homolog) to form a morphogen gradient that patterns the wing disk depends on Dally, a proteoglycan core protein (Belenkaya et al., 2004). This diffusion depends in turn on the secreted factor Pentagone, without which Dpp remains tightly bound to proteoglycans close to its site of secretion (Vuilleumier et al., 2010). Thus, the association of urchin Nodal with sulfated GAGs/proteoglycans may normally mediate its diffusion and inhibition of sulfation might undermine this process. We propose that interaction of urchin Nodal with chondroitin/dermatan sulfate is required to limit its diffusion and maintain a center of Nodal signaling in the oral field at a sufficient local concentration and activity to positively autoregulate its own expression after the mid-blastula stage (see our model in Fig. 11). In ClO3-treated embryos, Nodal activity is spread out and diluted, resulting in initial expansion of oral markers, defective differentiation of oral ectoderm and subsequent aboralization of the ectoderm. This model is consistent with the limited Nodal diffusion previously inferred (Yaguchi et al., 2007; Duboc et al., 2010). Expansion of Nodal signaling is presumably not as pronounced in embryos treated with 1.0–2.0 mM ClO3 (Fig. 10). Most of these embryos were rescued by co-treatment with low doses of the inhibitor of TGF-beta signaling SB-431542. The inhibition of low levels of ectopic Nodal signaling in these embryos may be sufficient to downregulate ectopic nodal expression and yet maintain an autoregulatory center of Nodal signaling that specifies the oral field on one side of the embryo.

1 2 8 ( 2 0 1 1 ) 7 1 –8 9

83

3.3. Aberrant Nodal signaling and expression in ClO3treated embryos Expression of the nodal gene is the earliest known transcriptional event in the specification of the oral ectoderm. Beginning at fifth cleavage, nodal is expressed and quickly restricted to the presumptive oral ectoderm where it plays a critical role in OA axis specification (Duboc et al., 2004; Flowers et al., 2004). The timing and spatial expression of nodal was normal in early blastula embryos treated with ClO3 (12 hpf; Fig. 7B1), when it is under transcriptional control of the p38 stress-activated protein kinase and a redox anisotropy across the prospective OA axis of the early cleavage egg (Bradham and McClay, 2006; Coffman et al., 2009). However, nodal expression was later disrupted (18–48 hpf; Fig. 7B2, B3, B4, and B5). Staining for phospho-Smad (Fig. 9) indicates that early Nodal signaling (15 hpf) began with a spatially normal pattern in ClO3-treated embryos, but that it soon expanded within the ectoderm (18 hpf). The spatial patterns of nodal and lefty expression were also expanded in treated midblastulae (18 hpf; Fig. 7B2 and D2), consistent with Nodal’s Smad-dependent autoregulatory positive feedback loop playing an active role in its expression and that of its antagonist Lefty (Nam et al., 2007; Range et al., 2007; Bolouri and Davidson, 2009; Su et al., 2009). The first observed effects of ClO3 treatment occurred after the initiation of asymmetric Nodal signaling, suggesting that earlier OA axis patterning processes (redox gradient, p38 kinase activity) are not significantly affected by GAG undersulfation. The response of other oral ectoderm genes to ClO3 treatment differs from that of nodal: their spatial expression was not altered in most gastrulae when ClO3 treatment began at 24 hpf (Fig. 8D00 and E); only nodal expression was delocalized (Fig. 8F). Continuing, localized expression of nodal depends on a positive autoregulatory feedback loop (Nam et al., 2007; Range et al., 2007; Bolouri and Davidson, 2009), and is sensitive to ClO3 treatment in late blastulae and post-gastrulae (24 and 42–72 hpf, respectively; Fig. 6A). This complementary evidence suggests that the primary effect of reduced sulfation is on localization and/or maintenance of Nodal activity and consequently on proper nodal expression. If nodal expression is too low and too delocalized to positively autoregulate its own expression, its second peak of expression in ClO3-treated early gastrulae (36 hpf; Fig. 6A) cannot depend on that autoregulation. The TGF-beta ligand Univin is an activator of nodal transcription whose zygotic expression is independent of Nodal signaling (Range et al., 2007). Thus, Univin may help promote the second temporal peak of nodal expression in ClO3-treated gastrulae. Ectopic Nodal appears to trigger Smad2/3 activation 1–2 cells from its origin (Yaguchi et al., 2007), implying diffusion of this ligand is limited. However, recent evidence suggests Nodal can diffuse 7–9 cell diameters to reach endomesoderm precursors within the vegetal plate of the blastula (Duboc et al., 2010). In contrast, the Nodal antagonist Lefty diffuses more freely in urchin embryos (Duboc et al., 2008), as does BMP2/4 (Lapraz et al., 2009). The properties of Nodal and Lefty expression in urchin embryos create a reaction–diffusion system predicted to pattern all three germ layers along the OA axis based on a slight initial asymmetry of expression (Duboc

84

MECHANISMS OF DEVELOPMENT

et al., 2008, 2010). Modeling suggests that higher levels of Lefty expression dampen Nodal expression and that its greater mobility confines Nodal to a domain of uniform expression in the oral ectoderm territory (Bolouri and Davidson, 2009). It is possible that the antagonistic effect of Lefty on Nodal activity is impeded by treatment with ClO3. Lefty mRNA was reduced and spatially expanded prior to gastrulation in embryos treated with ClO3 from 2 hpf, so it may be less effective in restricting Nodal signaling to the oral field, promoting expansion of Nodal activity and embryo radialization. However, when translation of lefty mRNA is specifically knocked down, nodal RNA is expressed in a broad equatorial ring but not notably reduced per embryo (Duboc et al., 2008), in contrast to embryos treated with ClO3 in which it is considerably reduced and becomes delocalized. Though we cannot rule out the possibility of an additional effect of ClO3 on Lefty activity in treated embryos, the model we propose for the regulation of Nodal ligand diffusion by sulfated GAGs is the simplest one we could devise that describes an initial expansion of Nodal signaling followed by a downregulation of its expression. Inhibition of Lefty activity alone cannot explain the complex phenotype we obtain by ClO3 treatment. Nodal signaling is sufficient to promote expression of oral markers (including itself) in ClO3-treated blastulae, though at reduced levels and with initially expanded domains. We suggest that the diminished levels of nodal expression are the result of a dependence of the positive autoregulation loop of Nodal signaling and expression on sulfation. Sulfated GAGs may play a role in the activity or stability of Nodal and/or activation of its receptor ALK-4/5/7; or they could act as co-receptors as for fibroblast growth factor (FGF) signaling (Yayon et al., 1991). Alternatively, Nodal may diffuse away from the oral field if it is not sufficiently anchored to undersulfated GAGs, reducing its concentration below a threshold required to promote differentiation of oral ectoderm. We favor this latter scenario as it is consistent with the early expansion of nodal expression and Nodal-dependent Smad2/3 activation, and the subsequent expression of aboral markers in much of the ectoderm as a result of the decline in Nodal expression throughout the embryo.

3.4. Aboral features of the ectoderm in ClO3-treated embryos The ectoderm of embryos treated with ClO3 beginning at 2 hpf expresses various territorial markers in disrupted spatial patterns. Oral markers are first expressed throughout most of the blastula ectoderm but then decline, presumably because Nodal signaling is impaired in ClO3-treated embryos. Expression of the early aboral markers spec1 (Figs. 5 and 7F) and cyIIIa (Supplementary Fig. S4A–D, I, and J) then take over much of the vegetal ectoderm that later assumes the squamous epithelial morphology of aboral ectoderm. This conversion to presumptive aboral ectoderm probably contributes to the decline of expression of oral-specific genes. Treatments that interfere with oral ectoderm specification, such as the knockdown of deadringer (Amore et al., 2003), onecut/hnf6 (Otim et al., 2004) or gsc (Angerer et al., 2001) cause expression of the aboral marker spec1 to spread through the non-polar ectoderm. The underlying gene regulatory networks normally

1 2 8 ( 2 0 1 1 ) 7 1 –8 9

insure the mutually exclusive expression of genes specific to the oral and aboral ectoderm territories (Su et al., 2009; Duboc et al., 2010). Treatment with ClO3 also diminishes early expression of tbx2/3 that encodes a transcription factor essential for aboral ectoderm specification (Gross et al., 2003). We suggest that although aboral specification is initially perturbed, much of the ectoderm of ClO3-treated postblastula embryos eventually differentiates, at least partially, into aboral ectoderm because perturbed Nodal signaling is unable to maintain the oral field. BMP2/4 is required for specification and differentiation of aboral ectoderm. This ligand is synthesized and secreted by cells of the oral ectoderm and diffuses as a morphogen to specify the aboral ectoderm (Angerer et al., 2000; Duboc et al., 2004; Lapraz et al., 2009). The radial pattern of inferred BMP2/4-dependent Smad activity observed at 24 hpf in ClO3treated embryos (Fig. 9), combined with possible loss of counteracting oralizing activities, may be sufficient to promote the creation of the broadened, radialized ectoderm territory marked by cyIIIa and spec1 expression. Sulfated GAGs/proteoglycans enhance BMP ligand activity (Takada et al., 2003) and mediate its diffusion (Belenkaya et al., 2004). Expression of the proteoglycan glypican-5 is restricted to the aboral ectoderm of P. lividus late blastulae and may participate in a positive feedback loop maintaining BMP signaling on the aboral side of the embryo (Lapraz et al., 2009). However, inhibition of sulfation did not mimic effects of perturbation of BMP2/4 signaling reported by Lapraz et al. (2009) for sea urchin embryos. The BMP antagonist Chordin prevents BMP2/4 from specifying aboral ectoderm in its oral domain of expression in P. lividus (Lapraz et al., 2009), but chordin expression is reduced and delocalized in ClO3-treated embryos (not shown), likely contributing to the expansion of aboral ectoderm.

3.5. Endomesoderm patterning and gastrulation defects in embryos treated with ClO3 Nodal and BMP2/4 also have essential roles in OA patterning of the endoderm and mesoderm (Duboc et al., 2010). Consistent with its disruption of nodal expression, ClO3 treatment resulted in radialized endomesoderm patterning as well. For example, cyIIa is normally expressed on the oral side of prospective secondary mesenchyme cells at the tip of the archenteron during gastrulation (Arnone et al., 1998). As the cyIIIa actin gene encodes a protein almost identical to that encoded by the cyIIa gene (Shott et al., 1984), our cyIIIa probe hybridizes to both cyIIIa and oral mesoderm-specific cyIIa mRNAs in gastrulae. In ClO3 treated embryos, all of the cells at the tip of the gut express cyIIa (Supplementary Fig. S4B4 and D3). Conversely, gcm is expressed in presumptive aboral mesoderm of mesenchyme blastula embryos (Ransick and Davidson, 2006) and its expression is lost following ClO3 treatment (24 hpf; Supplementary Fig. S5B and C). The expansion of an oral mesenchyme marker (cyIIa) at the expense of an aboral one (gcm) in late blastulae and early gastrulae is consistent with our proposed initial expansion of Nodal signaling and oral features, though it is delayed relative to ectoderm patterning. Since pigment cells, derivatives of aboral secondary mesenchyme, eventually form in ClO3-treated embryos, we suggest that as in the case of ectoderm specification,

MECHANISMS OF DEVELOPMENT

aboral mesenchyme features later take over from oral ones. This process might contribute to the observed delay in mesenchyme differentiation (Figs. 1B 0 and 5B 0 ). The expression patterns of endoderm markers gatae and endo16 confirmed a delay or defect in the internalization of archenteron cells observed in developing ClO3-treated embryos. A ring of cells expressing these endoderm-specific genes around the blastopore indicates some presumptive endoderm cells had failed to internalize by 48 hpf (Fig. 8H 0 and H00 ; Supplementary Fig. S5E and F). Normally, bra is expressed in a tight ring of cells around the blastopore during gastrulation, but in ClO3-treated embryos the ring was expanded (Fig. 8D 0 ; Supplementary Fig. S5A). There was no overlap of expression Endo1 and Spec1 in older embryos (144 hpf; Fig. 5A 0 ), indicating that the presumptive endoderm cells outside of the blastopore of gastrulae treated with 3 mM ClO3 eventually became part of the archenteron. Extension of the archenteron was considerably reduced in embryos treated with 30 mM ClO3 compared to 3 mM (Fig. 1B and C). It is not clear what processes are inhibited by the higher concentration of ClO3 that restrict extension of the archenteron. ActivinB signaling is involved in the specification of endomesoderm and disrupting it delays gastrulation (Sethi et al., 2009). ActivinB signals through the same ALK-4/ 5/7 receptor as Nodal, suggesting it may also depend on sulfated GAGs disrupted by ClO3. However, expression of RNA markers indicated that presumptive endoderm cells remained correctly specified along the AV axis at all concentrations of ClO3 tested (Fig. 8H 0 , and H00 ; Supplementary Fig. S5D), though gastrulation was delayed at high concentrations (30 mM ClO3). The ECM is required for normal cell movements during development (reviewed in VanSaun and Matrisian (2006)), suggesting that inhibition of sulfation may have interfered with the cellular rearrangements required for convergent extension of the archenteron. Extension of the archenteron the final 1/3 of the distance across the blastocoel of untreated embryos depends on the extension of filopodia from SMCs at the tip of the gut that recognize a binding target on the inner surface of the oral ectoderm (Hardin and McClay, 1990). Treatment of embryos with ClO3 impeded the maintenance of an oral field eliminating the filopodial target. However, this cannot explain why gut extension was inhibited only at high concentrations of ClO3.

3.6. Oral ectoderm differentiation and mouth formation in ClO3-treated embryos The hallmark of the oral side of an animal is the presence of a mouth opening. Formation of the urchin embryonic mouth involves invagination of oral ectoderm to form the stomodeum, attachment of the archenteron tip to the stomodeum prior to fusion, and perforation of the two fusing epithelial sheets and the hyaline layer (the ECM covering the apical surface of ectoderm cells) to form the oral aperture (Gustafson and Wolpert, 1963). This tissue fusion process is similar to dorsal closure in Drosophila, eyelid fusion in vertebrates and wound healing. Little is known about the dependence of these processes on sulfation or the ECM (Martin and Parkhurst, 2004). No oral opening or stomodeal invagination was observed by light microscopy in embryos treated

1 2 8 ( 2 0 1 1 ) 7 1 –8 9

85

with ClO3 (Fig. 1B). Bra mRNA is a marker for the prospective stomodeum (Tagawa et al., 1998; Gross and McClay, 2001); it was not seen in the ectoderm of the animal hemisphere of embryos treated with ClO3 beginning at 2 hpf (Fig. 8D 0 ) but it was observed in the oral ectoderm of embryos treated from 24 hpf (Fig. 8D00 ), though no stomodeal invagination or mouth were formed (Fig. 1B00 ). Ectoderm and endoderm tissues were correctly patterned in many embryos treated with ClO3 beginning 24 hpf; only the expression of nodal was seriously disturbed. This suggests that the presumptive stomodeum is specified but does not differentiate properly in embryos treated with ClO3 beginning at 24 hpf. In many of these embryos, the archenteron extended across the blastocoel and bent toward the cuboidal ectoderm of the presumptive oral field, indicating that the tip of the archenteron recognized this region of ectoderm as oral. However, the archenteron tip failed to fuse with the overlying oral ectoderm where presumptive stomodeal cells expressed bra. Thus the mouth formation defect does not appear to be due to a failure of oral tissue specification. Moreover, though OA polarity was restored, mouth formation was usually not rescued by addition of SO4 to ClO3-treated embryos (Fig. 3). Therefore, ClO3 treatment prevents stomodeal invagination and the oral tissue fusion event but undersulfation might not be direct the cause of the observed mouth formation defect.

3.7.

Developmental roles of sulfated GAGs

Taken together, our experiments provide evidence that sulfation is necessary for the correct function of GAGs and proteoglycans in establishment and/or perception of the TGF-beta signals released by ectodermal cells leading to normal OA patterning in the developing urchin embryo. Our results show that proper nodal expression and Nodal signaling is dependent on continuing sulfation, probably through an effect of sulfated GAGs on the diffusion of this TGF-beta ligand. We propose that interaction of Nodal with sulfated GAGs is required to maintain an organizing center of Nodal signaling in the oral field at a sufficient local concentration to positively autoregulate its own expression and promote the specification and differentiation of oral tissue and proper patterning of aboral ectoderm. Additionally, sulfation may play essential roles in convergent-extension, tissue fusion and/or cell adhesion events involved in gastrulation.

4.

Experimental procedures

4.1.

Animal collection, culture and treatments

Adult S. purpuratus sea urchins from Vancouver Island, British Columbia, were induced to shed gametes by electrostimulation. Embryos (50/ml or fewer) were cultured in SW at 12 C. Developing embryos were treated with 0, 1, 1.5, 2, 3, 10 and 30 mM NaClO3 (ClO3; Sigma) at different times post-fertilization. Embryos were also treated with 3 mM Na2SeO4 (SeO4; Sigma), 20 mM Na2SO4 (SO4; Fisher Scientific), 1 mM 4-nitrophenyl-beta-D-xylopyranoside (pNPX; Sigma), or 0.5 and 5.0 lM SB-431542 (SB; EMD Chemicals). Artificial SW

86

MECHANISMS OF DEVELOPMENT

was prepared according to Kester et al. (1967). Low sulfate SW, containing only 10 mM sulfate (SO4), was prepared by replacing part of the Na2SO4 with NaCl.

4.2.

Synthesis of riboprobes

DNA fragments of at least 500 bp of coding sequence were PCR amplified and cloned in the pBluescript II KS+ plasmid. RNA probes were synthesized with T7 or T3 RNA polymerase (Ambion), 2 lg linear template DNA and the DIG RNA labeling mix (Roche) according to manufacturer instructions. Template DNA was degraded using RNAse-free DNAse I (Roche). Probes were passed through a Micro Bio-spin 30 Column (Bio-Rad) before use.

4.3.

Whole mount in situ hybridization

This protocol, adapted from Minokawa et al. (2004), does not use protease digestion. The ECM-preserving fixative was used as described. Embryos were then washed once with 0.1 M MOPS, pH 7.0, 1.0 M NaCl; dehydrated in a graded series of ethanol; and preserved at 20 C in 70% ethanol. Hybridization was also performed as described, except the 3 h posthybridization wash which was replaced with 3 successive 45 min MOPS buffer (0.1 M MOPS, pH 7.0, 0.5 M NaCl, 0.1% Tween 20) washes at 50 C. The embryos were incubated with anti-DIG-AP fragments (Roche) as described and stained in NBT/BCIP liquid substrate system (Sigma) up to 24 h.

4.4.

Alcian Blue staining

The cationic dye Alcian Blue reacts specifically with sulfated functional groups at pH lower than 2.5 (Pearse, 1960). Staining conditions were derived from Bjornsson (1993). Whole gastrula embryos treated at 24 hpf with increasing concentration of ClO3 were fixed 1 h in SW containing 4% paraformaldehyde. The fixed embryos were washed three times with GT buffer (0.4 M guanidine-HCl, 18 mM sulfuric acid, 0.25% Triton X-100, pH 1.0) and stained overnight at RT in GT buffer containing 0.25% Alcian Blue 8GX (ESBE Scientific). The stained embryos were then thoroughly washed in GT buffer. Cell membranes were prepared according to Williams et al. (1984). Gastrula embryos were washed twice in PBS, 0.05% EDTA and their cells lysed with hypotonic borate buffer as described with one change: PMSF was replaced with cOmplete protease inhibitor cocktail (Roche). Next, the membrane preparations were immobilized on PVDF according to Karlsson et al. (2000) with some modifications. The PVDF membrane was derivatized by incubation in 1% CTAB, 30% propanol and rinsed extensively in 0.15 M NaCl. Membrane preparations (10 lg) were allowed to pass through pre-rinsed wells of a dot blot apparatus in 300 ll of buffer solution (0.15 M NaCl, 50 mM Tris–acetate, pH 7.4) containing 0.25% SDS. For total protein staining, the membrane was incubated in Coomassie stain (0.1% Coomassie brilliant blue R250, 50% methanol, 10% acetic acid). For sulfate staining, the membrane was incubated in Alcian stain (0.25% Alcian Blue 8GX, 0.4 M guanidine-HCl, 18 mM sulfuric acid, 0.25% Triton X-100, pH 1.0) (Reagent 2B; Bjornsson, 1998).

4.5.

1 2 8 ( 2 0 1 1 ) 7 1 –8 9

Whole mount immunostaining

Mouse monoclonals anti-Endo1 (1:100; Wessel and McClay, 1985), anti-SP1 (1:100; Gibson and Burke, 1985) and rabbit polyclonals anti-Spec1 (1:100; Wikramanayake et al., 1995), anti-serotonin (1:500; Sigma S5545), and anti-phoshoSmad3/Smad1 (1:500; Cell Signaling Technology #9514) were used as primary antibodies. Urchin embryos were fixed 1 h with 4% paraformaldehyde in SW, rinsed twice in PBS, 0.1% Triton X-100 (PBST) and blocked 2 h with 10% goat serum, 10% BSA in PBST. Embryos were incubated with primary antibody overnight at 4 C and thoroughly washed in PBST. Fluorescent secondary antibodies anti-mouse Alexa 488 and/or anti-rabbit Alexa 568 (1:500– 1:1000, Molecular Probes) were added for 2 h and thoroughly washed in PBS. For phospho-Smad staining, embryos were fixed for 10 min only and transferred to cold methanol. Samples were mounted in Vectashield (Vector Labs) for viewing.

4.6.

Microscopy

Bright field and differential interference photomicroscopy were performed using a Vanox AHBS3 light microscope (Olympus) equipped with 20· and 40· objectives and a Sony PowerHAD 3CCD color video camera. Immunofluorescent images were captured using an Axio Observer.Z1 microscope (Carl Zeiss International) equipped with a CSU10 Nipkow confocal spinning disk unit (Yokogawa Electronic), 10· and 25· Zeiss objectives and a C9100-13 EM-CCD imaging camera (Hamamatsu Photonics). Single confocal slices were processed with Volocity software (version 5.2; PerkinElmer).

4.7.

Quantitative real-time PCR (qRT-PCR)

Total RNA was extracted from 5000 embryos using the RNeasy mini kit (QIAgen). Genomic DNA contamination was eliminated from the extracted total RNA with the DNA-free kit (Ambion). Complementary DNA was prepared from 1 lg total RNA hybridized to 0.1 nmol poly dT20 with 100 U M-MLV reverse transcriptase (Ambion). The reverse transcriptase was heat inactivated and the RNA degraded with 2.5 U RNAse H (Ambion). The synthesized cDNA was extracted with phenol:chloroform:isoamyl alcohol (25:24:1) then ethanolprecipitated in the presence of 0.1 g/L linear acrylamide. Quantitative RT-PCRs were performed on the StepOne Real-Time PCR System with Power SYBR Green Master Mix (Applied Biosystems). Each reaction was performed in triplicate, using 20 ng of cDNA/reaction and z12-1 as an endogenous control. Primer sequences for z12-1, nodal, lefty, bmp2/ 4, gsc, cyIIIa, tbx2/3 and spec1 were taken from Agca et al. (2009). The numbers of z12-1 mRNAs per single embryo have previously been determined as 2300, 2100, 1600, 1900, 1200, 1100, and 1600 molecules for egg, 4, 12, 24, 48, 54, and 72 h, respectively (Wang et al., 1995). In the present study, we used 1600 molecules for 12 and 18 h, 1900 molecules for 24, 30 and 36 h, 1200 molecules for 42 and 48 h, and 1600 molecules for 72 h as standard numbers for z12-1 mRNA per embryo, and calculated the estimated number of transcripts of interest

MECHANISMS OF DEVELOPMENT

using the formula from Otim et al. (2004). Briefly, the number of transcripts of gene targets per embryo at any given stage (Qtarget) was calculated using the known values for z12-1 (Qz12), with the formula Q target ¼ Q z12  2ðCt ½targetCt ½z12Þ .

Acknowledgements This research was supported by a NSERC Discovery Grant (RGPIN 45977) to B.P.B. We thank Corina Dumitrescu for her initial experiments with ClO3 treatment of Lytechinus pictus embryos, and Greg Murray for technical help. We also thank Robert and Lynne Angerer for a plasmid containing the gsc cDNA, Robert Burke for the SP1 antibody, David McClay for the Endo1 antibody and Shunsuke Yaguchi for advice on the use of the phospho-Smad antibody.

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.mod.2010.11.001.

R E F E R E N C E S

Agca, C., Klein, W.H., Venuti, J.M., 2009. Respecification of ectoderm and altered Nodal expression in sea urchin embryos after cobalt and nickel treatment. Mech. Dev. 126, 430–442. Akasaka, K., Amemiya, S., Terayama, H., 1980. Scanning electron microscopical study of the inside of sea urchin embryos (Pseudocentrotus depressus): effects of aryl beta-xyloside, tunicamycin and deprivation of sulfate ions. Exp. Cell Res. 129, 1–13. Akasaka, K., Uemoto, H., Wilt, F., Mitsunaga-Nakatsubo, K., Shimada, H., 1997. Oral–aboral ectoderm differentiation of sea urchin embryos is disrupted in response to calcium ionophore. Dev. Growth Differ. 39, 373–379. Amore, G., Yavrouian, R.G., Peterson, K.J., Ransick, A., McClay, D.R., Davidson, E.H., 2003. Spdeadringer, a sea urchin embryo gene required separately in skeletogenic and oral ectoderm gene regulatory networks. Dev. Biol. 261, 55–81. Angerer, R.C., Davidson, E.H., 1984. Molecular indices of cell lineage specification in sea urchin embryos. Science 226, 1153– 1160. Angerer, L.M., Oleksyn, D.W., Logan, C.Y., McClay, D.R., Dale, L., Angerer, R.C., 2000. A BMP pathway regulates cell fate allocation along the sea urchin animal–vegetal embryonic axis. Development 127, 1105–1114. Angerer, L.M., Oleksyn, D.W., Levine, A.M., Li, X., Klein, W.H., Angerer, R.C., 2001. Sea urchin goosecoid function links fate specification along the animal–vegetal and oral–aboral embryonic axes. Development 128, 4393–4404. Arnone, M.I., Martin, E.L., Davidson, E.H., 1998. Cis-regulation downstream of cell type specification: a single compact element controls the complex expression of the CyIIa gene in sea urchin embryos. Development 125, 1381–1395. Baeuerle, P.A., Huttner, W.B., 1986. Chlorate – a potent inhibitor of protein sulfation in intact cells. Biochem. Biophys. Res. Commun. 141, 870–877. Belenkaya, T.Y., Han, C., Yan, D., Opoka, R.J., Khodoun, M., Liu, H., Lin, X., 2004. Drosophila Dpp morphogen movement is independent of dynamin-mediated endocytosis but regulated by the glypican members of heparan sulfate proteoglycans. Cell 119, 231–244.

1 2 8 ( 2 0 1 1 ) 7 1 –8 9

87

Bernfield, M., Go¨tte, M., Park, P.W., Reizes, O., Fitzgerald, M.L., Lincecum, J., Zako, M., 1999. Functions of cell surface heparan sulfate proteoglycans. Ann. Rev. Biochem. 68, 729–777. Bjornsson, S., 1993. Simultaneous preparation and quantitation of proteoglycans by precipitation with Alcian Blue. Anal. Biochem. 210, 282–291. Bjornsson, S., 1998. Quantitation of proteoglycans as glycosaminoglycans in biological fluids using an Alcian Blue dot blot analysis. Anal. Biochem. 256, 229–237. Bolouri, H., Davidson, E.H., 2009. The gene regulatory network basis of the ‘‘community effect’’, and analysis of a sea urchin embryo example. Dev. Biol. 340, 170–178. Bradham, C.A., McClay, D.R., 2006. P38 MAPK is essential for secondary axis specification and patterning in sea urchin embryos. Development 133, 21–32. Cameron, R.A., Fraser, S.E., Britten, R.J., Davidson, E.H., 1990. Segregation of oral from aboral ectoderm precursors is completed at fifth cleavage in the embryogenesis of Strongylocentrotus purpuratus. Dev. Biol. 137, 77–85. Carey, D.J., 1991. Biological functions of proteoglycans: use of specific inhibitors of proteoglycan synthesis. Mol. Cell. Biochem. 104, 21–28. Coffman, J.A., Coluccio, A., Planchart, A., Robertson, A.J., 2009. Oral–aboral axis specification in the sea urchin embryo III. Role of mitochondrial redox signaling via H2O2. Dev. Biol. 330, 123–130. Duboc, V., Ro¨ttinger, E., Besnardeau, L., Lepage, T., 2004. Nodal and BMP2/4 signaling organizes the oral–aboral axis of the sea urchin embryo. Dev. Cell 6, 397–410. Duboc, V., Lapraz, F., Besnardeau, L., Lepage, T., 2008. Lefty acts as an essential modulator of Nodal activity during sea urchin oral–aboral axis formation. Dev. Biol. 320, 49–59. Duboc, V., Lapraz, F., Saudemont, A., Bessodes, N., Mekpoh, F., Haillot, E., Quirin, M., Lepage, T., 2010. Nodal and BMP2/4 pattern the mesoderm and endoderm during development of the sea urchin embryo. Development 137, 223–235. Flowers, V.L., Courteau, G.R., Poustka, A.J., Weng, W., Venuti, J.M., 2004. Nodal/activin signaling establishes oral–aboral polarity in the early sea urchin embryo. Dev. Dyn. 231, 727–740. Gibson, A.W., Burke, R.D., 1985. The origin of pigment cells in embryos of the sea urchin Strongylocentrotus purpuratus. Dev. Biol. 107, 414–419. Greve, H., Cully, Z., Blumberg, P., Kresse, H., 1988. Influence of chlorate on proteoglycan biosynthesis by cultured human fibroblasts. J. Biol. Chem. 263, 12886–12892. Gross, J.M., McClay, D.R., 2001. The role of Brachyury (T) during gastrulation movements in the sea urchin Lytechinus variegatus. Dev. Biol. 239, 132–147. Gross, J.M., Peterson, R.E., Wu, S.Y., McClay, D.R., 2003. LvTbx2/3: a T-box family transcription factor involved in formation of the oral/aboral axis of the sea urchin embryo. Development 130, 1989–1999. Guss, K.A., Ettensohn, C.A., 1997. Skeletal morphogenesis in the sea urchin embryo: regulation of primary mesenchyme gene expression and skeletal rod growth by ectoderm-derived cues. Development 124, 1899–1908. Gustafson, T., Wolpert, L., 1963. The cellular basis of morphogenesis and sea urchin development. Int. Rev. Cytol. 15, 139–214. Hacker, U., Nybakken, K., Perrimon, N., 2005. Heparan sulphate proteoglycans: the sweet side of development. Nat. Rev. Mol. Cell Biol. 6, 530–541. Hardin, J., Armstrong, N., 1997. Short-range cell–cell signals control ectodermal patterning in the oral region of the sea urchin embryo. Dev. Biol. 182, 134–149. Hardin, J., Coffman, J.A., Black, S.D., McClay, D.R., 1992. Commitment along the dorsoventral axis of the sea urchin embryo is altered in response to NiCl2. Development 116, 671–685.

88

MECHANISMS OF DEVELOPMENT

Hardin, J., McClay, D.R., 1990. Target recognition by the archenteron during sea urchin gastrulation. Dev. Biol. 142, 86– 102. Ho¨rstadius, S., 1973. Experimental Embryology of Echinoderms. Clarendon Press, Oxford. Humphries, D.E., Silbert, J.E., 1988. Chlorate: a reversible inhibitor of proteoglycan sulfation. Biochem. Biophys. Res. Commun. 154, 365–371. Inman, G.J., Nicola´s, F.J., Callahan, J.F., Harling, J.D., Gaster, L.M., Reith, A.D., Laping, N.J., Hill, C.S., 2002. SB-431542 is a potent and specific inhibitor of transforming growth factor-beta superfamily type I activin receptor-like kinase (ALK) receptors ALK4, ALK5, and ALK7. Mol. Pharmacol. 62, 65–74. Irie, A., Habuchi, H., Kimata, K., Sanai, Y., 2003. Heparan sulfate is required for bone morphogenetic protein-7 signaling. Biochem. Biophys. Res. Commun. 308, 858–865. Karlsson, M., Edfors-Lilja, I., Bjornsson, S., 2000. Binding and detection of glycosaminoglycans immobilized on membranes treated with cationic detergents. Anal. Biochem. 286, 51–58. Karp, G.C., Solursh, M., 1974. Acid mucopolysaccharide metabolism, the cell surface, and primary mesenchyme cell activity in the sea urchin embryo. Dev. Biol. 41, 110–123. Kester, D.R., Duedall, I.W., Connors, D.N., Pytkowicz, R.M., 1967. Preparation of artificial sea water. Limnol. Oceanogr. 12, 176–179. Khurrum, M., Hernandez, A., Eskalaei, M., Badali, O., CoyleThompson, C., Oppenheimer, S.B., 2004. Carbohydrate involvement in cellular interactions in sea urchin gastrulation. Acta Histochem. 106, 97–106. Kinoshita, S., Saiga, H., 1979. The role of proteoglycan in the development of sea urchins. I. Abnormal development of sea urchin embryos caused by the disturbance of proteoglycan synthesis. Exp. Cell Res. 123, 229–236. Klein, W.H., Tomlinson, C.R., Zhang, W., Xiang, M., Kozlowski, M., Gan, L., Nguyen, T., Wessel, G.M., 1990. Ectoderm differentiation and the Spec gene family of sea urchins. In: Davidson, E., Ruderman, J., Posakony, J. (Eds.), Developmental Biology, UCLA Symposium on Molecular and Cellular Biology, New Series, vol. 125. Alan R. Liss Inc., New York, pp. 123–134. Kramer, K.L., Yost, H.J., 2003. Heparan sulfate core proteins in cellcell signaling. Annu. Rev. Genet. 37, 461–484. Lapraz, F., Besnardeau, L., Lepage, T., 2009. Patterning of the dorsal-ventral axis in echinoderms: insights into the evolution of the BMP-Chordin signaling network. PLoS Biol. 7. doi:10.1371/journal.pbio.1000248. Lin, X., 2004. Functions of heparin sulfate proteoglycans in cell signaling during development. Development 131, 6009–6021. Lyman, J., Fleming, R.H., 1940. Composition of sea water. J. Mar. Res. 3, 134–146. Martin, P., Parkhurst, S.M., 2004. Parallels between tissue repair and embryo morphogenesis. Development 131, 3021–3034. Minokawa, T., Rast, J.P., Arenas-Mena, C., Franco, C.B., Davidson, E.H., 2004. Expression patterns of four different regulatory genes that function during sea urchin development. Gene Expr. Patterns 4, 449–456. Mulloy, B., Rider, C.C., 2006. Cytokines and proteoglycans: an introductory overview. Biochem. Soc. Trans. 34, 409–413. Nam, J., Su, Y.-H., Lee, P.Y., Robertson, A.J., Coffman, J.A., Davidson, E.H., 2007. Cis-regulatory control of the nodal gene, initiator of the sea urchin oral ectoderm gene network. Dev. Biol. 306, 860–869. Oki, S., Hashimoto, R., Okui, Y., Shen, M.M., Mekada, E., Otani, H., Saijoh, Y., Hamada, H., 2007. Sulfated glycosaminoglycans are necessary for Nodal signal transmission from the node to the left lateral plate in the mouse embryo. Development 134, 3893– 3904. Otim, O., Amore, G., Minokawa, T., McClay, D.R., Davidson, E.H., 2004. SpHnf6, a transcription factor that executes multiple functions in sea urchin embryogenesis. Dev. Biol. 273, 226–243.

1 2 8 ( 2 0 1 1 ) 7 1 –8 9

Pearse, A.G.E., 1960. Histochemistry: Theorical and Applied. Little, Brown, and Company, Boston. pp. 672–673. Range, R., Lapraz, F., Quirin, M., Marro, S., Besnardeau, L., Lepage, T., 2007. Cis-regulatory analysis of nodal and maternal control of dorsal-ventral axis formation by Univin, a TGF-beta related to Vg1. Development 134, 3649–3664. Ransick, A., Davidson, E.H., 2006. Cis-regulatory processing of Notch signaling input to the sea urchin glial cells missing gene during mesoderm specification. Dev. Biol. 297, 587–602. Rozario, T., DeSimone, D.W., 2010. The extracellular matrix in development and morphogenesis: a dynamic view. Dev. Biol. 341, 126–140. Runnstro¨m, J., Ho¨rstadius, S., Immers, J., Fudge-Mastrangelo, M., 1964. An analysis of the role of sulfate in the embryonic differentiation of the sea urchin (Paracentrotus lividus). Rev. Suisse Zool. 71, 21–31. Safaiyan, F., Kolset, S.O., Prydz, K., Gottfridsson, E., Lindahl, U., Salmivirta, M., 1999. Selective effects of sodium chlorate treatment on the sulfation of heparan sulfate. J. Biol. Chem. 274, 36267–36273. Sethi, A.J., Angerer, R.C., Angerer, L.M., 2009. Gene regulatory network interactions in sea urchin endomesoderm induction. PLoS Biol. 7. doi:10.1371/journal.pbio.1000029. Shott, R.J., Lee, J.J., Britten, R.J., Davidson, E.H., 1984. Differential expression of the actin gene family of Strongylocentrotus purpuratus. Dev. Biol. 101, 295–306. Solursh, M., Katow, H., 1982. Initial characterization of sulfated macromolecules in the blastocoels of mesenchyme blastulae of Strongylocentrotus purpuratus and Lytechinus pictus. Dev. Biol. 94, 326–336. Solursh, M., Mitchell, S.L., Katow, H., 1986. Inhibition of cell migration in sea urchin embryos by beta-D-xyloside. Dev. Biol. 118, 325–332. Su, Y.-H., Li, E., Geiss, G.K., Longabaugh, W.J.R., Kra¨mer, A., Davidson, E.H., 2009. A perturbation model of the gene regulatory network for oral and aboral ectoderm specification in the sea urchin embryo. Dev. Biol. 329, 410–421. Sugahara, K., Schwartz, N.B., 1979. Defect in 3 0 phosphoadenosine 5 0 -phosphosulfate formation in brachymorphic mice. Proc. Natl. Acad. Sci. USA 76, 6615–6618. Sugiyama, K., 1972. Occurence of mucopolysaccharides in the early development of the sea urchin embryo and its role in gastrulation. Dev. Growth Differ. 14, 63–73. Tagawa, K., Humphreys, T., Satoh, N., 1998. Novel pattern of Brachyury gene expression in hemichordate embryos. Mech. Dev. 75, 139–143. Takacs, C.M., Amore, G., Oliveri, P., Poustka, A.J., Wang, D., Burke, R.D., Peterson, K.J., 2004. Expression of an NK2 homeodomain gene in the apical ectoderm defines a new territory in the early sea urchin embryo. Dev. Biol. 269, 152–164. Takada, T., Katagiri, T., Ifuku, M., Morimura, N., Kobayashi, M., Hasegawa, K., Ogamo, A., Kamijo, R., 2003. Sulfated polysaccharides enhance the biological activities of bone morphogenetic proteins. J. Biol. Chem. 278, 43229– 43235. Tang, S.S., Trackman, P.C., Kagan, H.M., 1983. Reaction of aortic lysyl oxidase with beta-aminopropionitrile. J. Biol. Chem. 258, 4331–4338. Ullrich, T.C., Huber, R., 2001. The complex structures of ATP sulfurylase with thiosulfate, ADP and chlorate reveal new insights in inhibitory effects and the catalytic cycle. J. Mol. Biol. 313, 1117–1125. VanSaun, M.N., Matrisian, L.M., 2006. Matrix metalloproteinases and cellular motility in development and disease. Birth Defects Res. C Embryo Today 78, 69–79. Venkatachalam, K.V., 2003. Human 3 0 -phosphoadenosine 5 0 phosphosulfate (PAPS) synthase: biochemistry, molecular biology and genetic deficiency. IUBMB Life 55, 1–11.

MECHANISMS OF DEVELOPMENT

Vilela-Silva, A.C., Werneck, C.C., Valente, A.P., Vacquier, V.D., Moura˜o, P.A., 2001. Embryos of the sea urchin Strongylocentrotus purpuratus synthesize a dermatan sulfate enriched in 4-O- and 6-O-disulfated galactosamine units. Glycobiology 11, 433–440. Vuilleumier, R., Springhorn, A., Patterson, L., Koidl, S., Hammerschmidt, M., Affolter, M., Pyrowolakis, G., 2010. Control of Dpp morphogen signalling by a secreted feedback regulator. Nat. Cell Biol. 12, 611–617. Wang, D.G., Britten, R.J., Davidson, E.H., 1995. Maternal and embryonic provenance of a sea urchin embryo transcription factor, SpZ12-1. Mol. Mar. Biol. Biotechnol. 4, 148–153. Wessel, G.M., McClay, D.R., 1985. Sequential expression of germlayer specific molecules in the sea urchin embryo. Dev. Biol. 111, 451–463. Wikramanayake, A.H., Brandhorst, B.P., Klein, W.H., 1995. Autonomous and non-autonomous differentiation of ectoderm in different sea urchin species. Development 121, 1497–1505. Williams, L.T., Tremble, P.M., Lavin, M.F., Sunday, M.E., 1984. Platelet-derived growth factor receptors form a high affinity state in membrane preparations. Kinetics and affinity crosslinking studies. J. Biol. Chem. 259, 5287–5294.

1 2 8 ( 2 0 1 1 ) 7 1 –8 9

89

Wilson, L.G., Bandurski, R.S., 1958. Enzymatic reactions involving sulfate, sulfite, selenate, and molybdate. J. Biol. Chem. 233, 975–981. Yaguchi, S., Yaguchi, J., Burke, R.D., 2006. Specification of ectoderm restricts the size of the animal plate and patterns neurogenesis in sea urchin embryos. Development 133, 2337– 2346. Yaguchi, S., Yaguchi, J., Burke, R.D., 2007. Sp-Smad2/3 mediates patterning of neurogenic ectoderm by nodal in the sea urchin embryo. Dev. Biol. 302, 494–503. Yaguchi, S., Yaguchi, J., Angerer, R.C., Angerer, L.M., 2008. A Wnt– FoxQ2–nodal pathway links primary and secondary axis specification in sea urchin embryos. Dev. Cell 14, 97–107. Yaguchi, S., Yaguchi, J., Angerer, R.C., Angerer, L.M., Burke, R.D., 2010. TGFbeta signaling positions the ciliary band and patterns neurons in the sea urchin embryo. Dev. Biol. 347, 71– 81. Yayon, A., Klagsbrun, M., Esko, J.D., Leder, P., Ornitz, D.M., 1991. Cell surface, heparin-like molecules are required for binding of basic fibroblast growth factor to its high affinity receptor. Cell 64, 841–848.