HpSulf, a heparan sulfate 6-O-endosulfatase, is involved in the regulation of VEGF signaling during sea urchin development

MECHANISMS OF DEVELOPMENT

4 2 7 ( 2 0 1 0 ) 2 3 5 –2 4 5

available at www.sciencedirect.com

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

HpSulf, a heparan sulfate 6-O-endosulfatase, is involved in the regulation of VEGF signaling during sea urchin development Kazumasa Fujita a, Eriko Takechi a, Naoaki Sakamoto a, Noriko Sumiyoshi a, Shunsuke Izumi a, Tatsuo Miyamoto b, Shinya Matsuura b, Toko Tsurugaya c, Koji Akasaka c, Takashi Yamamoto a,* a

Department of Mathematical and Life Sciences, Graduate School of Science, Hiroshima University, 1-3-1 Kagamiyama, Higashi-Hiroshima 739-8526, Japan b Department of Radiation Biology, Research Institute for Radiation Biology and Medicine, Hiroshima University, 1-2-3 Kasumi, Minami-ku, Hiroshima 734-8553, Japan c Misaki Marine Biological Station, Graduate School of Sciences, University of Tokyo, 1024 Koajiro, Misaki, Miura, Kanagawa 238-0225, Japan

A R T I C L E I N F O

A B S T R A C T

Article history:

Cell surface heparan sulfate proteoglycans (HSPGs) play significant roles in the regulation

Received 28 October 2009

of developmental signaling, including vascular endothelial growth factor (VEGF), fibro-

Received in revised form

blast growth factor, Wnt and bone morphogenetic protein signaling, through modification

3 December 2009

of their sulfation patterns. Recent studies have revealed that one of the functions of hep-

Accepted 15 December 2009

aran sulfate 6-O-endosulfatase (Sulf) is to remove the sulfate from the 6-O position of

Available online 29 December 2009

HSPGs at the cell surface, thereby regulating the binding activities of heparan sulfate (HS) chains to numerous ligands and receptors in animal species. In this study, we

Keywords: Heparan sulfate 6-O-endosulfatase Vascular endothelial growth factor Heparan sulfate proteoglycan Cell surface Signaling Skeletogenesis Sea urchin

focused on the sea urchin Hemicentrotus pulcherrimus homolog of Sulf (HpSulf), and analyzed its expression pattern and functions during development. HpSulf protein was present throughout development and localized at cell surface of all blastomeres. In addition, the HS-specific epitope 10E4 was detected at the cell surface and partially colocalized with HpSulf. Knockdown of HpSulf using morpholino antisense oligonucleotides (MO) caused abnormal morphogenesis, and the development of MO-injected embryos was arrested before the hatched blastula stage, indicating that HpSulf is necessary for the early developmental process of sea urchin embryos. Furthermore, we found that injection of HpSulf mRNA suppressed the abnormal skeleton induced by overexpression of HpVEGF mRNA, whereas injection of an inactive form of HpSulf mRNA, containing mutated cysteines in the sulfatase domain, did not have this effect. Taken together, these results suggest that HpSulf is involved in the regulation of various signal transductions, including VEGF signaling, during sea urchin development.  2009 Elsevier Ireland Ltd. All rights reserved.

1.

Introduction

Morphogens and growth factors such as Wnt, bone morphogenetic protein, vascular endothelial growth factor (VEGF)

and fibroblast growth factor (FGF) play essential roles in axis formation and cell fate specification during sea urchin development (Angerer et al., 2000; Wikramanayake et al., 2004; Duloquin et al., 2007; Ro¨ttinger et al., 2008). These signaling

* Corresponding author. Tel.: +81 82 424 7446; fax: +81 82 424 7498. E-mail address: [email protected] (T. Yamamoto). 0925-4773/$ - see front matter  2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.mod.2009.12.001

236

MECHANISMS OF DEVELOPMENT

factors are locally distributed and form signaling gradients that induce specification and differentiation into various cell types (Angerer and Angerer, 2000; Angerer et al., 2000; Wikramanayake et al., 2004; Duloquin et al., 2007; Ro¨ttinger et al., 2008). In the sea urchin embryo, the spicules are specifically synthesized by the primary mesenchyme cells (PMCs) (Gustafson and Wolpert, 1967; Okazaki, 1975). At the mesenchyme blastula stage, prospective PMCs ingress and migrate, and then become positioned along the extracellular matrix that lines the blastocoel wall. At the gastrula stage, PMCs form a characteristic ring with two symmetrical clusters, followed by the formation of a tri-radiate spicule rudiment. Thereafter, the spicule rudiment increases in size and complexity (Decker and Lennarz, 1988). During this process, FGF and VEGF are reported to be involved in the morphogenesis of the embryonic skeleton (Duloquin et al., 2007; Ro¨ttinger et al., 2008). FGF signaling controls gastrulation and skeletogenesis via the regulation of several genes, including pea3, pax2/5/8, SM30 and SM50 (Ro¨ttinger et al., 2008). VEGF receptor (VEGFR) is expressed in all PMCs, whereas VEGF is restricted to two small areas of the ectoderm. Overexpression of VEGF by injection of its mRNA leads to aberrant multiple ramifications and duplication of the spicule rods at the pluteus stage. VEGFVEGFR signaling modulates specific genes in PMCs, such as SM30 and SM50, suggesting that it functions in the positioning and differentiation of the migrating PMCs during gastrulation through the interaction between ectoderm and these cells (Duloquin et al., 2007). However, the precise mechanisms for the regulation of VEGF and FGF signaling in the sea urchin are still not understood. It has been reported that cell surface heparan sulfate proteoglycans (HSPGs), which are composed of a core protein with covalently linked heparan sulfate (HS) glycosaminoglycan chains, play important roles in the regulation of various signaling pathways during embryogenesis (Lin, 2004). HS chains are synthesized and modified by numerous biosynthetic enzymes, including N-deacetylase/N-sulfotransferase, 2-O-sulfotransferase (2-O-ST), 3-O-ST, 6-O-ST and uridine 5 0 -diphosphate-glucose dehydrogenase (Shworak et al., 1997; Spicer et al., 1998; Aikawa and Esko, 1999; Bernfield et al., 1999; Rong et al., 2001; Smeds et al., 2003; Chapman et al., 2004; Lin, 2004). Recent reports have demonstrated that one of the functions of HS 6-O-endosulfatase (Sulf) is to remove the sulfate from the 6-O position of HSPGs at the cell surface, suggesting that Sulf modulates the sulfation pattern of HS chains within the extracellular region. Hence, Sulf family proteins are thought to regulate the binding activities of HSPGs to numerous ligands and receptors, including VEGF165, FGF, stromal cell-derived factor-1 and Noggin (Viviano et al., 2004; Uchimura et al., 2006), and consequently play roles in tumorigenesis, angiogenesis and embryogenesis (Wang et al., 2004; Narita et al., 2006; Freeman et al., 2008). In the present study, we examined the involvement of Sulf in the regulation of the signaling mechanisms required for sea urchin development. To this end, we analyzed the localization of the Hemicentrotus pulcherrimus homolog of Sulf (HpSulf) as well as that of HS in the sea urchin embryo. We also found that the knockdown of HpSulf using morpholino antisense oligonucleotides resulted in developmental arrest

4 2 7 ( 2 0 1 0 ) 2 3 5 –2 4 5

before the hatched blastula stage. Furthermore, HpSulf suppressed VEGF-induced supernumerary spicules, while an inactive form of HpSulf with mutated cysteines in the sulfatase domain was unable to exert this effect. Our findings provide the first evidence that Sulf is an in vivo regulator of VEGF signaling during development.

2.

Results

2.1.

The structure of HpSulf

Using a PCR-based strategy, we isolated the cDNA for the sea urchin H. pulcherrimus homolog of Sulf (HpSulf). The nucleotide sequence of the HpSulf cDNA obtained in this study was 3759 bp in length and contained a single open reading frame of 2952 bp that encoded a 983-aa protein with a calculated molecular mass of 113 kDa. The prediction of protein sorting signals (PSORT) program (http://www.genome.ad.jp) estimated that the first 20 amino acids in the predicted protein for HpSulf comprise a signal sequence (Fig. 1A). A domain search analysis of HpSulf using the PFAM (protein family of alignments) database revealed the presence of a sulfatase domain (aa 60–449; Fig. 1A), within which a core region (aa 106–117) contains a cysteine residue that is conserved in the sulfatase family (Fig. 1A and B). This residue is modified to a-formylglycine by sulfatase modifying factor and is considered to be important for the enzymatic activity of the sulfatases (Schmidt et al., 1995; Hanson et al., 2004). A hydropathy analysis showed that the HpSulf protein contains a hydrophilic region at its C-terminus (aa 440–851) (Fig. 1A), which has also been reported in other Sulf proteins. In addition, there are six potential N-linked glycosylation and three furin cleavage sites in HpSulf protein that are present in the other Sulf family members (Fig. 1A). Indeed, five of the six potential N-linked glycosylation sites are conserved at the corresponding position in the vertebrate Sulfs. The amino acid sequence of the sulfatase domain of HpSulf was compared with those of other Sulf proteins. Excluding the 87-aa insertion found in its hydrophilic region, HpSulf shows 55.6–60.7% identities to vertebrate Sulfs and 51.4% identity to Caenorhabditis elegans Sulf. The core region of the sulfatase domain of HpSulf exhibits 83.3–91.7% identities to other Sulfs, 83.3% identity to human glucosamine-6-sulfatase (HG6S) and 58.3–66.7% identities to arylsulfatases (Fig. 1B). Next, we performed a phylogenic analysis for the predicted partial amino acid sequences of the sulfatase domains of HpSulf, other Sulfs and HG6S, a lysosomal sulfatase related to the Sulfs, using the neighbor-joining method. As shown in Fig. 1C, the branching of animal Sulfs matches the phylogenetic relationships of their respective species, suggesting that HpSulf is indeed the Sulf ortholog in H. pulcherrimus.

2.2. The temporal pattern of HpSulf mRNA and HpSulf protein expression during sea urchin development The HpSulf mRNA expression was examined by Northern blot analysis and whole-mount in situ hybridization (WMISH), but we were unable to detect clear signals because of the low amount of the mRNA (data not shown). As an alternative, we

MECHANISMS OF DEVELOPMENT

4 2 7 ( 2 0 1 0 ) 2 3 5 –2 4 5

237

Fig. 1 – Structure of HpSulf and an amino acid sequence comparison of Sulf family members. (A) Structure of the Hemicentrotus pulcherrimus homolog of Sulf (HpSulf). The signal sequence is indicated by a red box. The sulfatase domain (purple) was identified by analysis of the PFAM (protein family of alignments) database and contains a core region (black). A hydropathy analysis revealed a hydrophilic region at the C-terminus (yellow). The open and filled circles denote potential N-linked glycosylation and furin cleavage sites, respectively. The bottom panel shows the results of the hydropathy analysis of HpSulf protein. (B) Alignment of the amino acid sequences of a core region of the sulfatase domain among sulfatase family members. The asterisk indicates a cysteine, that is required for catalytic activity in the sulfatase family. HpSulf, Hemicentrotus pulcherrimus Sulf; HSulf-1, human Sulf-1; HSulf-2, human Sulf-2; MSulf-1, mouse Sulf-1; MSulf-2, mouse Sulf-2; RSulfFP-1, rat SulfFP-1; RSulfFP-2, rat SulfFP-2; QSulf-1, quail Sulf-1; QSulf-2, quail Sulf-2; XtSulf-1, Xenopus tropicalis Sulf-1; ZfSulf-1, zebrafish Sulf-1; ZfSulf-2, zebrafish Sulf-2; CeSulf-1, Caenorhabditis elegans Sulf-1; HG6S, human glucosamine-6-sulfatase; HArsA, human arylsulfatase A; HpArs, Hemicentrotus pulcherrimus arylsulfatase. (C) Phylogenetic tree analysis of the members of the Sulf family based on the amino acid sequences of the sulfatase domain. examined the HpSulf mRNA expression by semiquantitative reverse transcription-polymerase chain reaction (RT-PCR) (Fig. 2A). HpSulf mRNA expression was detected in the unfertilized egg, and then began to gradually decrease toward the hatched blastula stage. HpSulf mRNA expression was found to increase from the mesenchyme blastula stage to the prism stage. Next, we examined the expression of HpSulf protein during development by Western blot analysis using purified anti-HpSulf antibodies. HpSulf protein was detected as three bands of different sizes with estimated molecular weights of about 70, 85 and 110 kDa (Fig. 2B, arrows). The signal for the 110-kDa band gradually decreased from the morula stage to the mesenchyme blastula stage, while the signals for the 70- and 85-kDa bands increased from the unhatched blastula stage, with peaks at the hatched blastula stage and prism stage, respectively. The signals for these bands were abolished or reduced by the addition of competing peptides (Fig. 2C). On the other hand, the signals for the bands exceeding 200 kDa were not reduced by the addition of peptides, and these bands were therefore considered to be non-specific (Fig. 2C, asterisk). We also examined the localization of HpSulf protein by immunostaining with anti-HpSulf polyclonal antibodies. As shown in Fig. 3A, HpSulf protein was detected on the cell surface of all blastomeres at the hatched blastula stage. At the prism larva stage, HpSulf protein was observed in

all cell lineages and found to be localized at the cell surface (Fig. 3C and C 0 ). The signals were reduced by the addition of competing peptides (Fig. S1A–D). To confirm the cell surface localization of HpSulf protein, we transfected an expression construct for His-tagged HpSulf into HEK293T cells and examined its localization by live staining of nonpermeabilized HEK293T cells using an anti-His tag antibody. Histagged HpSulf protein was also detected at the cell surface of HEK293T cells where several spots were observed (Fig. 3E).

2.3.

Localization of HS chains in the sea urchin embryo

HS has been identified in sea urchin embryos (Solursh and Katow, 1982). However, the localization of HS in the sea urchin has not determined. Thus, we examined the localization of HS in the sea urchin embryos by immunostaining with the 10E4 antibody, which recognizes native HS with sulfated glucosamine residues (Yip et al., 2002; Ai et al., 2003; Narita et al., 2006). The 10E4-positive signals were observed as dots at the cell surface and blastocoel, and were partially colocalized with those of HpSulf protein (Fig. 4D and D 0 ). In addition, the reactivity with the 10E4 antibody was abolished by treatment with heparitinase I (Fig. 4H), which specifically cleaves HS chains. These results indicated that the HS and HpSulf are both localized at the cell surface in the sea urchin embryo.

238

MECHANISMS OF DEVELOPMENT

4 2 7 ( 2 0 1 0 ) 2 3 5 –2 4 5

Fig. 2 – Temporal pattern of HpSulf mRNA and HpSulf protein expressions during sea urchin development. (A) Semiquantitative RT-PCR analysis of HpSulf at various developmental stages in the sea urchin. The upper and lower panels show HpSulf and HpMitCOI mRNAs, respectively. The total PCR cycles were as follows: HpSulf, 35 cycles; HpMitCOI, 27 cycles. UF, unfertilized egg; 16, 16-cell stage; Mo, morula; UHB, unhatched blastula; HB, hatched blastula; MB, mesenchyme blastula; eG, early gastrula; lG, late gastrula; Pri, prism; Plu, pluteus. (B) Western blot analysis. Aliquots containing 1 lg of protein prepared from sea urchin embryos at various developmental stages were analyzed in parallel. (C) An absorption assay using a synthetic peptide evaluated by Western blot analysis. The arrows and asterisk in (B) and (C) indicate three bands of different sizes corresponding to HpSulf proteins and nonspecific bands, respectively.

2.4. Knockdown of HpSulf results in the arrest of developmental processes To further examine the roles of Sulf during sea urchin development, we designed experiments to perturb the embryos by inhibiting the translation of HpSulf through the injection of fertilized eggs with HpSulf morpholino antisense oligonucleotides (MO). First, we confirmed the intended effects of the HpSulf MO on the translation of this gene using a GFP reporter mRNA containing the target sequence for the HpSulf MO. As shown in Fig. 5D, GFP fluorescence was effectively blocked by the HpSulf MO at the mesenchyme blastula stage compared with the control MO. In addition, the effect of the HpSulf MO was confirmed at the hatched blastula stage by Western blot analysis using anti-HpSulf antibodies (Fig. 5E). The amounts of the 70 and 85 kDa of HpSulf proteins were decreased by injection of the HpSulf MO, whereas that of 110-kDa HpSulf protein was not. These results indicate that the HpSulf MO specifically blocked the synthesis of the 70- and 85-kDa HpSulf proteins, but had no effect on the synthesis of the 110-kDa HpSulf protein. The latter findings may arise because the 110-kDa protein is maternally supplied form HpSulf. This possibility is supported by our observations that the 110-kDa protein was abundant during the cleavage stage and gradually decreased toward the mesenchyme blastula stage (Fig. 2B).

Fig. 3 – The HpSulf protein is localized at the cell surface. (A–D, C 0 and D 0 ) Embryos were fixed, immunostained with anti-HpSulf polyclonal antibodies and observed using a confocal laser scanning microscope. (A and B) HB, hatched blastula stage; (C, D, C 0 and D 0 ) Pri, prism stage. Magnified images of the boxed areas in (C) and (D) are shown in (C 0 ) and (D 0 ), respectively. The arrows indicate the localization of HpSulf protein at the cell surface. (E and F) Live staining of nonpermeabilized HEK293T cells expressing His-tagged HpSulf detected using an anti-His tag antibody. The arrowheads indicate positive signals for His-tagged HpSulf protein. Epifluorescence images (A, C, C 0 and E) and bright field images (B, D, D 0 and F) are shown. Scale bars: 10 lm. Next, we injected several doses of the HpSulf MO into fertilized sea urchin eggs and examined their morphological effects during development. The embryos injected with the HpSulf MO developed normally until the unhatched blastula stage (Fig. 6A, D, G and J). However, these morphants exhibited anomalous development after the hatched blastula stage in a dose-dependent manner. When injections of 0.6 and 1.2 fmol/embryo were performed, a delay in development and the presence of abnor-

MECHANISMS OF DEVELOPMENT

4 2 7 ( 2 0 1 0 ) 2 3 5 –2 4 5

239

Fig. 4 – The heparin sulfate (HS) chains of heparan sulfate proteoglycans (HSPGs) are localized at the cell surface. (A–D and A 0 – D 0 ) Embryos were fixed, immunostained with an anti-heparan sulfate monoclonal antibody (10E4; B and B 0 ) and anti-HpSulf polyclonal antibodies (C and C 0 ), and observed using a confocal laser scanning microscope. Magnified images of the boxed areas in (A) to (D) are shown in (A 0 ) to (D 0 ), respectively. (E–H) Embryos were treated with heparitinase I or untreated. (A–H) G, gastrula. Bright field images (A, A 0 , E and G) and fluorescence images (B–D, B 0 –D 0 , F and H) are shown. Scale bars: 25 lm.

mal cells in the blastocoel were observed at 13.5 h post-fertilization (hpf) (Fig. 6M). In addition, when the control embryos reached to the mesenchyme blastula stage (Fig. 6C), an increase in the number of abnormal cells in the blastocoel and dissociation of the blastomeres were observed (Fig. 6F and I). When a dose of 0.3 fmol/embryo was injected, most of the HpSulf MOinjected embryos developed normally until the hatched blastula stage (Fig. 6J and K). However, abnormal cells in the blastocoel were detectable in these embryos when the control embryos reached the mesenchyme blastula stage (Fig. 6C) and gastrulation failed to occur (Fig. 6L). Hence, all the embryos injected with the HpSulf MO seemed to arrest their development, although the timing of this effect was dependent on the dose of the injected MO. In addition, we carried out a rescue experiment using HpSulf mRNA without the HpSulf MO target site. Co-injection of the HpSulf mRNA with the HpSulf MO decreased the number of abnormal cells in the blastocoel compared with injection of the HpSulf MO alone, although a developmental delay was observed (Fig. 6N–P). However, gastrulation was not observed in these embryos. These results suggest that the HpSulf mRNA injection was able to partially rescue the morphological effects induced by HpSulf MO injection. Therefore, to examine

the cause of the partial rescue, we carried out Western blot analyses using anti-HpSulf antibodies. As shown in Fig. 6Q, injection of the HpSulf mRNA led to a marked increase in the 110-kDa HpSulf protein, whereas the 70-kDa HpSulf protein was only slightly increased and the 85-kDa HpSulf protein was not increased. These findings indicate that the 70- and 85-kDa HpSulf proteins, whose syntheses are blocked by the HpSulf MO injection, were not efficiently produced from the injected HpSulf mRNA, suggesting that the partial rescue may be caused by incomplete supply of the 70- and 85-kDa HpSulf proteins. Therefore, although the phenotypes induced by the HpSulf MO injection were not completely rescued by the injection of HpSulf mRNA, it is possible that the morphological effects were caused by defects in the translation of the HpSulf mRNA. Taken together, our results suggest that HpSulf is required for the early development of the sea urchin embryo.

2.5. HpSulf suppresses VEGF-induced supernumerary spicules in the sea urchin embryo In the sea urchin species Paracentrotus lividus, it has been reported that the homolog of VEGF is expressed in ectodermal

240

MECHANISMS OF DEVELOPMENT

4 2 7 ( 2 0 1 0 ) 2 3 5 –2 4 5

jected the HpVEGF mRNA with the HpSulf mRNA into fertilized eggs, the severe effects were reduced to occurrence in 30% of the larvae (Fig. 7C, D and I), and the percentage of embryos

Fig. 5 – Effects of the HpSulf MO on translational inhibition. (A–D) A GFP reporter mRNA containing the 5 0 UTR of the HpSulf mRNA was co-injected with the control MO or HpSulf MO into fertilized sea urchin eggs. GFP fluorescence is detected at the mesenchyme blastula stage in control MOinjected embryos (B), but not in the HpSulf MO-injected embryos (D). Bright field images (A and C) and GFP fluorescence images (B and D) are shown. (E) Western blot analysis using anti-HpSulf antibodies. The embryos were injected with the control MO or HpSulf MO. Total protein extracts prepared from the embryos at the hatched blastula stage were analyzed in parallel. The arrows indicate three bands of different sizes corresponding to HpSulf proteins.

cells and regulates skeletogenesis (Duloquin et al., 2007). In the present study, our aim was to examine the involvement of HpSulf in the regulation of VEGF signaling in the sea urchin embryo. Consequently, we isolated the cDNA for the H. pulcherrimus homolog of VEGF (HpVEGF), and examined its expression and the effects of its overexpression in H. pulcherrimus embryos. (Fig. S2A–K). Consistent with the previous results in P. lividus (Duloquin et al., 2007), HpVEGF mRNA was localized at two restricted areas of the ectoderm from the hatched blastula stage to the gastrula stage (Fig. S2B–D). In addition, supernumerary spicules were induced at the pluteus stage by overexpression of HpVEGF (Fig. S2H–K). Based on these results, we speculate that VEGF is involved in skeletogenesis in the H. pulcherrimus embryo. Next, we examined the effects of HpSulf mRNA injection on the morphology of the supernumerary spicules induced by overexpression of HpVEGF. When we injected the HpVEGF mRNA, severe effects on the skeleton were observed in 85% of the prism larvae (Fig. 7A, B and I) similar to the case for the pluteus larvae (Fig. S2H–K). In contrast, when we co-in-

MECHANISMS OF DEVELOPMENT

showing a normal skeleton and a mild phenotype was consequently increased (Fig. 7I). On the other hand, when we injected the HpSulf mRNA alone, more than 95% of the embryos exhibited normal development and normal skeletogenesis was observed (Fig. 7G–I). These results suggest that the incorrect signaling caused by overexpression of HpVEGF is countered by the forced expression of HpSulf. It has been reported that the enzyme activity of Sulf is involved in the regulation of Wnt and FGF signaling (Dhoot et al., 2001; Wang et al., 2004). Therefore, we examined whether the inactive form of HpSulf (mutHpSulf), in which two point mutations were introduced at conserved cysteine residues (C106A and C107A) to block the a-formylglycine modification of the amino acid, inhibits the formation of the supernumerary spicules caused by overexpression of HpVEGF. In the embryos co-injected with the HpVEGF and mutHpSulf mRNAs, no suppression of the supernumerary spicules was observed (Fig. 7E, F and I). As a control, embryos were injected with the mutHpSulf mRNA alone and these embryos developed normally (data not shown). These results suggest that the enzyme activity of HpSulf is required for the regulation of VEGF signaling.

3.

Discussion

In the present study, we isolated the HpSulf cDNA and showed that the predicted amino acid sequence contains a signal motif and a sulfatase domain as well as the C-terminal hydrophilic regions that are highly conserved in Sulf family proteins. These findings suggest that the broad structural organization of the Sulfs is well conserved in the sea urchin. Sulf family proteins undergo furin cleavage in various species (Morimoto-Tomita et al., 2002; Tang and Rosen, 2009),

b Fig. 6 – Morphological effects of the HpSulf MO during early sea urchin development. (A–C) Embryos injected with the control MO. Normal PMC ingression is observed. UHB, unhatched blastula; HB, hatched blastula; MB, mesenchyme blastula. (D–L) Embryos injected with the HpSulf MO. Anomalous development is observed in a dose-dependent manner. The embryos were observed at 10 h postfertilization (hpf) (A, D, G and J), 13.5 hpf (B, E, H and K) and 23 hpf (C, F, I and L). (M) Percentages of normal embryos at 13.5 hpf following injection with the HpSulf MO. (N–P) A rescue experiment using an HpSulf mRNA without the HpSulf MO target site. The embryos were injected with 1.2 fmol of the control MO (N) or 1.2 fmol of the HpSulf MO without (O) or with (P) 3.2 pg of HpSulf mRNA. The embryos were observed at 19 hpf (N–P). MB, mesenchyme blastula. Bright field images are shown (A–L and N–P). (Q) Western blot analysis using anti-HpSulf antibodies. The embryos were injected with 1.2 fmol of the HpSulf MO without or with 3.2 pg of the HpSulf mRNA. Total protein extracts prepared from the embryos were analyzed in parallel. The total amounts of proteins in the lanes were evaluated by Coomassie Brilliant Blue R-250 (CBB) staining. The arrows indicate three bands of different sizes corresponding to HpSulf proteins.

4 2 7 ( 2 0 1 0 ) 2 3 5 –2 4 5

241

which generates a variety of different sizes for these molecules. In the present study, we observed three different bands for HpSulf by Western blot analysis. Since the HpSulf protein contains potential furin cleavage sites, it is possible that posttranslational modification of HpSulf by furin occurs in the sea urchin embryo. In addition, the potential N-linked glycosylation sites conserved in HpSulf may also generate differently sized proteins. In fact, it has been reported that changes in the molecular weight of quail Sulf-1 (QSulf-1) occur because of N-linked glycosylation (Ambasta et al., 2007). Sulf is localized at the extracellular regions where it modulates the sulfation pattern of HS chains (Ai et al., 2003; Freeman et al., 2008). In the present study, we detected the presence of HpSulf at the surface of cell lineages such as ectodermal cells and mesenchymal cells in the sea urchin embryo. In addition, HS was also detected at the cell surface of these cells. On the other hand, our preliminary data showed that HpSulf modulates the sulfation pattern in cultured cells (data not shown). Taken together, it is suggested that HpSulf functions as an extracellular sulfatase and is involved in various signaling pathways through modification of the sulfation pattern of HSPGs on the cell surface in the sea urchin embryo. In Xenopus tropicalis, Sulf-1 is expressed in the fin, somites, floor plate, pericardium and branchial arches (Freeman et al., 2008). In the mouse, Sulf-1 and Sulf-2 are broadly expressed throughout the embryo, with partially overlapping regions in the telencephalic vesicle, nasal placodes, somites, tip of the tail and floor plate (Ratzka et al., 2008). In contrast to these tissue-specific Sulf expression patterns, HpSulf protein was detected at the cell surface of all blastomeres, and cell lineage-specific expression was not observed in our analyses until the prism stage. Knockdown analysis of HpSulf caused severe effects such as developmental arrest and dissociation of blastomeres at the hatched blastula stage. On the other hand, overexpression of HpSulf mRNA did not appear to influence development. These results indicate that the expression of HpSulf is required for all blastomeres, and that excess HpSulf protein does not affect cell specification and differentiation during sea urchin development. Sulf proteins function to promote glial cell line-derived neurotrophic factor and Wnt signaling, and some of them inhibit FGF2 signaling (Ai et al., 2003, 2007; Dai et al., 2005; Freeman et al., 2008; Lamanna et al., 2008). QSulf-1 depresses the binding of Wnt to HS to promote the Wnt–Frizzled interaction (Ai et al., 2003). X. tropicalis Sulf-1 promotes the activity of Wnt11 in the early embryo, and acts as an endogenous inhibitor that modulates the BMP-mediated regulation of genes, including Otx2 and Pax6, during development (Freeman et al., 2008). These previous reports suggest that the Sulf proteins play important roles in the regulation of various signaling pathways. In the present study, we observed severe effects resulting from knockdown of HpSulf during early development in the sea urchin. In fact, many types of signaling are required for cell fate specification in the sea urchin embryo. For example, Wnt8 is expressed during the 16- to 60-cell stage, exhibits restricted expression to the veg2 tier and plays a critical role in the differentiation of the endomesoderm lineages (Wikramanayake et al., 2004), while ActivinB, which accumulates throughout the sea urchin embryo at the early cleavage and early blastula stages, has an essential role in endomeso-

242

MECHANISMS OF DEVELOPMENT

derm specification mediated by micromeres and their progeny (Sethi et al., 2009). Furthermore, BMP2/4 transcripts are preferentially expressed in the presumptive ectoderm, and BMP2/4 is important for oral/aboral ectodermal polarity and the differentiation of ectodermal cell types (Angerer et al., 2000). Therefore, we consider that the knockdown of HpSulf causes simultaneous disruption of the regulatory systems of

4 2 7 ( 2 0 1 0 ) 2 3 5 –2 4 5

various signaling pathways and leads to the arrest of the developmental process. In the sea urchin species P. lividus, it has been reported that VEGF-VEGFR signaling between the ectoderm and PMCs is associated with skeletogenesis through the regulation of the positioning and differentiation of the migrating PMCs during gastrulation (Duloquin et al., 2007). Consistent with this stage, we observed that HpSulf is localized at the cell surface of PMCs. This finding suggests the possibility that HpSulf may regulate the VEGF-VEGFR signaling at the cell surface of PMCs. It has been reported that heparin-binding angiogenic factors such as VEGF165 stimulate the proliferation of human umbilical vein endothelial cells (HUVECs) through the activation of ERK phosphorylation, and that this effect is not observed in HUVECs with suppressed expression of human Sulf-1 (Narita et al., 2006). That earlier report also suggested that Sulfs play roles in the negative regulation of VEGF signaling. In the present study, we have shown that HpSulf protein is localized at the cell surface of PMCs, and that HpSulf represses the supernumerary spicules induced by overexpression of VEGF, suggesting that it has a repressive function in VEGF signaling and that the Sulf function is conserved in the sea urchin embryo. We have also shown that co-injection of a mutated HpSulf mRNA fails to suppress the supernumerary spicules caused by overexpression of VEGF mRNA. This observation is consistent with a previous report that Sulfs with cysteine mutations within the catalytic domain are unable to function as regulators of various signaling pathways in the quail (Dhoot et al., 2001; Ai et al., 2003; Wang et al., 2004). On the other hand, it has been reported that human Sulf-2 (HSulf-2) abolishes the binding of VEGF165 to heparin through removal of the 6-O sulfate from heparin, and that a mutated HSulf-2 does not affect this binding (Uchimura et al., 2006). Given these earlier findings, it is reasonable to speculate that HpSulf functions as an extracellular enzyme that modifies HSPGs and thereby participates in the regulation of VEGF signaling. However, since the embryos injected with the HpSulf mRNA alone developed normally in the present study, HpSulf does not seem to affect the normal levels of VEGF signaling. A similar phenomenon

b Fig. 7 – HpSulf suppresses the supernumerary spicules caused by overexpression of HpVEGF. (A and B) Embryos injected with 0.4 pg of HpVEGF mRNA. (C and D) Embryos coinjected with 2.5 pg of HpSulf mRNA and 0.4 pg of HpVEGF mRNA. (E and F) Embryos co-injected with 2.5 pg of mutHpSulf mRNA and 0.4 pg of HpVEGF mRNA. (G and H) Embryos injected with 2.5 pg of HpSulf mRNA. The injected embryos were observed under a bright field microscope (A, C, E and G) and a dark field microscope (B, D, Fand H) at 50 h after fertilization. (I) Frequencies of the supernumerary spicules in embryos injected under the different conditions. The levels of the supernumerary spicules are classified into three groups: severe defect (multiple supernumerary spicules; purple), mild defect (one supernumerary spicule; yellow) and normal (no supernumerary spicules; blue).

MECHANISMS OF DEVELOPMENT

has been reported for the regulation of FGF signaling in the Xenopus embryo (Wang et al., 2004), since QSulf-1 rescues the elongated and differentiated mesoderm induced by FGF2 in Xenopus embryos, whereas QSulf-1 alone does not induce abnormal morphogenesis or the expression of mesodermal markers. These findings may suggest that Sulf plays roles in the maintenance of correct VEGF signaling during development by repressing any excess signaling.

4.

Experimental procedures

4.1.

Culture of the sea urchin embryos

Gametes of the sea urchin (H. pulcherrimus) harvested at Seto inland sea were obtained by coelomic injections of 0.55 M KCl, and fertilized eggs were subsequently cultured in the filtered sea water (FSW) at 16 C.

4.2.

Cloning of cDNAs for HpSulf and HpVEGF

The HpSulf and HpVEGF cDNA fragments were amplified by PCR from a H. pulcherrimus cDNA library with following PCR primer sets: forward 5 0 -AATGCCTTTGTGACCACGCCCATG-3 0 and reverse 5 0 -GGTCCGGACACAGTAGTACGAGTT-3 0 for HpSulf; forward 5 0 -GACGACGCCGGACTTACGGCGGAA-3 0 and reverse 5 0 -CACGTTTCAGAATAAATAGTAAGAGTCC-3 0 for HpVEGF, respectively. To obtain the 5 0 and 3 0 ends of HpSulf and HpVEGF cDNAs, rapid amplification of cDNA ends was performed using the GeneRacer Kit (Invitrogen). The HpSulf and HpVEGF fulllength cDNAs were sequenced using DTCA Quick Start kit (Beckman Coulter) and CEQ8000 (Beckman Coulter). The sequences of HpSulf and HpVEGF cDNAs have been deposited in GenBank (Accession No. AB513622 for HpSulf and No. AB513617 for HpVEGF).

4.3.

Semiquantitative RT-PCR

Total RNA was isolated from H. pulcherrimus embryos at various developmental stages using ISOGEN (Wako). Two micrograms of total RNA, which had been treated with DNase I was reverse-transcribed using a random hexanucleotide primer with ThermoScript RT-PCR system (Invitrogen). The cDNAs were amplified by PCR with Ex taq DNA polymerase (TaKaRa Bio). An aliquot of the RT reaction was then used in PCRs with the appropriate primers as follows; HpSulf (5 0 -CG AGATCACCAGAAGAAATGGGAC-3 0 and 5 0 -ATCTTCTTCAGGTT CACCCACGAT-3 0 ), HpMitCOI (5 0 -GGCACAGCTATGAGTGTAATT ATCC-3 0 and 5 0 -GATAGTTCATCCAGTCCCTGCTC-3 0 ). All comparisons were performed in the linear range of amplification. The PCR products were then separated by electrophoresis on 2% agarose gels and visualized by ethidium bromide staining.

4.4.

Western blot analysis

Polyclonal anti-HpSulf antibodies were raised in rabbits against the synthetic peptide CLDRTRGETPNRGRTN by immunizing the rabbit. The peptide was coupled with a CNBr-activated Sepharose (GE Healthcare) in coupling buffer (0.2 M NaHCO3, 0.5 M NaCl, pH 8.3) at 4 C overnight, followed by

4 2 7 ( 2 0 1 0 ) 2 3 5 –2 4 5

243

blocking with 0.2 M glycine, pH 8.0, at 4 C overnight. After washing, the Sepharose was incubated with HpSulf antiserum in incubating buffer (75 mM Tris–HCl, pH 8.0) at 4 C overnight. After washing with incubating buffer, the bound antibody was eluted with 100 mM glycine, pH 2.5, and then immediately neutralized by 1 M Tris–HCl, pH 8.0. Embryos sonicated in PBS for 30 s at 4 C were dissolved in sample buffer (final concentration: 0.35 M Tris–HCl, pH 6.8, 10.28% sodium dodecyl sulfate, 36% glycerol, 5% b-mercaptoethanol, 0.012% bromphenol blue), and boiled for 5 min. Proteins were separated by electrophoretic device (ATTO), and then blotted to PVDF membranes (Millipore). For detection of HpSulf, the membranes were blocked with TBST (Tris-buffered saline containing 0.05% Tween 20, 5% (w/v) skim milk, 1.5% goat serum) and incubated with purified polyclonal anti-HpSulf antibodies at 4 C overnight. After washing with TBST, the membranes were incubated for 1 h at room temperature with horseradish peroxidase-conjugated antirabbit IgG (1:5000; Pierce). The chemiluminescence of Super Signal West Dura Extended Duration Substrate (Pierce) by hydrolysis with horseradish peroxidase was detected by X-ray film. Molecular weights were determined using molecular weight standards (Precision Plus Protein Standards Dual color; Bio-Rad).

4.5.

Immunohistochemical analysis

Embryos were fixed with 4% paraformaldehyde–FSW for 16 h at 4 C. After washing with MOPS buffer, fixed embryos were blocked with 1% BSA in PBS for 1 h at room temperature. Embryos were then incubated with purified polyclonal antiHpSulf antibodies and a mouse anti-HS monoclonal antibody (10E4; 1:150; Seikagaku) at 4 C overnight. This was followed by washing with PBS and incubation with an Alexa Fluor 546 fluorescent-conjugated chicken anti-rabbit IgG antibodies (1:1000; Molecular Probes) and Alexa Fluor 488 fluorescentconjugated anti-mouse IgM secondary antibodies (1:600; Molecular Probes) in 1% BSA in PBS for 1 h at room temperature. After washing with PBS, embryos were observed under a fluorescence microscope (FV-1000; Olympus).

4.6. Heparitinase digestion of HS chains in sea urchin embryo Fixed embryos were treated with 10 mU heparitinase I (Seikagaku) for 2 h at 37 C in a buffer containing 50 mM sodium acetate and 10 mM calcium chlorite (pH 7.0).

4.7.

Cell culture

HEK293T cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM; WAKO) supplemented with 10% fetal bovine serum, 1% nonessential amino acids and 1% streptomycin/ penicillin in 100-mm dishes at 37 C under 5% CO2.

4.8.

DNA transfection and live staining

HEK293T cells were transfected with a pcDNA3.1/mycHisA expression vector (Invitrogen) harboring the HpSulf cDNA. At 2 days after the transfection, the cells were incu-

244

MECHANISMS OF DEVELOPMENT

bated with a mouse anti-His tag antibody (1:300; Novagen) for live staining in serum-free DMEM for 1 h at 37 C. After washing with PBS, the cells were fixed with 4% paraformaldehyde in PBS for 20 min at room temperature. After washing with PBS, the cells were blocked with 1% BSA in PBS for 1 h at room temperature, and then incubated with Alexa Fluor 488 fluorescent-conjugated anti-mouse IgG secondary antibodies (1:400; Molecular Probes) in 1% BSA in PBS at 4 C overnight. After further washing with PBS, the cells were observed under a fluorescence microscope (IX81; Olympus).

4.9. Microinjection of synthetic mRNA and antisense morpholino oligonucleotides into sea urchin eggs A substitution mutant of HpSulf (mutHpSulf) was generated using a PCR strategy with a mutated PCR primer set: forward, 5 0 -GCTCCTTCTCGGAGCAGTATCTTG-3 0 and reverse, 5 0 -GGCCATGGGCGTGGTCACAAAG-3 0 . To generate DNA templates for in vitro RNA synthesis, the plasmids containing HpSulf, mutHpSulf and HpVEGF cDNA inserts were linearized with restriction enzymes. Capped mRNAs were synthesized using the T7 mMessage mMachine kit and Cap Analog [m7G(5 0 )pppG; Ambion] in accordance with the manufacturer’s instructions. The mRNAs were extracted with phenol–chloroform and suspended in 40% glycerol. The mRNAs concentration and integrity of these preparations was determined by spectrophotometry and electrophoresis through agarose gels. MOs complementary to the sequence containing the 5 0 upstream sequence of the translation start site of HpSulf mRNA (5 0 -GTGTTTGGTGTGGCTACTGCAGTC-3 0 ) and a standard control MO (5 0 -CCTCTTACCTCAGTTACAATTTATA-3 0 ) were obtained from Gene Tools. Oligonucleotides were dissolved in 40% glycerol at a concentration of 3.6 · 108 molecules/pl. Microinjection of sea urchin eggs was carried out as described (Rast, 2000) with some previously reported modifications (Ochiai et al., 2008). For the knockdown of HpSulf, 2 pl of HpSulf MO solution was injected into fertilized sea urchin eggs. For the rescue assays, 2 pl of a 200 fg/pl dose of HpVEGF mRNA with 1.25 pg/pl of HpSulf mRNA or 1.25 pg/pl of mutHpSulf mRNA was injected into the fertilized sea urchin eggs. For Western blot analysis, these embryos were dissolved in sample buffer and boiled for 5 min. The total amounts of proteins in the lanes were evaluated by Coomassie Brilliant Blue R-250 (CBB) staining.

4.10.

Whole-mount in situ hybridization (WMISH)

WMISH was performed as originally described by Arenas-Mena et al. (2000) and modified by Minokawa et al. (2004). DIG-labeled sense and antisense RNA probes were prepared with the Ambion Megascript kit using DIG-11-UTP (Ambion).

Acknowledgments The authors thank Dr. M. Kiyomoto (Tateyama Marine Laboratory, Ochanomizu University) for supplying live sea urchins. This work was supported by Grants-in-Aid for Scientific Re-

4 2 7 ( 2 0 1 0 ) 2 3 5 –2 4 5

search (C) from the Ministry of Education, Science, Sports and Culture, Japan.

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.mod.2009.12.001.

R E F E R E N C E S

Ai, X., Do, A.T., Lozynska, O., Kusche-Gullberg, M., Lindahl, U., Emerson Jr., C.P., 2003. QSulf1 remodels the 6-O sulfation states of cell surface heparan sulfate proteoglycans to promote Wnt signaling. J. Cell Biol. 162, 341–351. Ai, X., Kitazawa, T., Do, A.T., Kusche-Gullberg, M., Labosky, P.A., Emerson Jr., C.P., 2007. SULF1 and SULF2 regulate heparan sulfate-mediated GDNF signaling for esophageal innervation. Development 134, 3327–3338. Aikawa, J., Esko, J.D., 1999. Molecular cloning and expression of a third member of the heparan sulfate/heparin GlcNAc Ndeacetylase/ N-sulfotransferase family. J. Biol. Chem. 274, 2690–2695. Ambasta, R.K., Ai, X., Emerson Jr., C.P., 2007. Quail Sulf1 function requires asparagine-linked glycosylation. J. Biol. Chem. 282, 34492–34499. Angerer, L.M., Angerer, R.C., 2000. Animal–vegetal axis patterning mechanisms in the early sea urchin embryo. Dev. Biol. 218, 1–12. 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. Arenas-Mena, C., Cameron, A.R., Davidson, E.H., 2000. Spatial expression of Hox cluster genes in the ontogeny of a sea urchin. Development 127, 4631–4643. 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. Annu. Rev. Biochem. 68, 729–777. Chapman, E., Best, M.D., Hanson, S.R., Wong, C.H., 2004. Sulfotransferases: structure, mechanism, biological activity, inhibition, and synthetic utility. Angew. Chem. Int. Ed. Engl. 43, 3526–3548. Dai, Y., Yang, Y., MacLeod, V., Yue, X., Rapraeger, A.C., Shriver, Z., Venkataraman, G., Sasisekharan, R., Sanderson, R.D., 2005. HSulf-1 and HSulf-2 are potent inhibitors of myeloma tumor growth in vivo. J. Biol. Chem. 280, 40066–40073. Decker, G.L., Lennarz, W.J., 1988. Skeletogenesis in the sea urchin embryo. Development 103, 231–247. Dhoot, G.K., Gustafsson, M.K., Ai, X., Sun, W., Standiford, D.M., Emerson Jr., C.P., 2001. Regulation of Wnt signaling and embryo patterning by an extracellular sulfatase. Science 293, 1663–1666. Duloquin, L., Lhomond, G., Gache, C., 2007. Localized VEGF signaling from ectoderm to mesenchyme cells controls morphogenesis of the sea urchin embryo skeleton. Development 134, 2293–2302. Freeman, S.D., Moore, W.M., Guiral, E.C., Holme, A.D., Turnbull, J.E., Pownall, M.E., 2008. Extracellular regulation of developmental cell signaling by XtSulf1. Dev. Biol. 320, 436–445. Gustafson, T., Wolpert, L., 1967. Cellular movement and contact in sea urchin morphogenesis. Biol. Rev. Camb. Philos. Soc. 42, 442–498. Hanson, S.R., Best, M.D., Wong, C.H., 2004. Sulfatases: structure, mechanism, biological activity, inhibition, and synthetic utility. Angew. Chem. Int. Ed. 43, 5736–5763.

MECHANISMS OF DEVELOPMENT

Lamanna, W.C., Frese, M.A., Balleininger, M., Dierks, T., 2008. Sulf loss influences N-, 2-O-, and 6-O-sulfation of multiple heparan sulfate proteoglycans and modulates fibroblast growth factor signaling. J. Biol. Chem. 283, 27724–27735. Lin, X., 2004. Functions of heparan sulfate proteoglycans in cell signaling during development. Development 131, 6009–6021. 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. Morimoto-Tomita, M., Uchimura, K., Werb, Z., Hemmerich, S., Rosen, S.D., 2002. Cloning and characterization of two extracellular heparin-degrading endosulfatases in mice and humans. J. Biol. Chem. 277, 49175–49185. Narita, K., Staub, J., Chien, J., Meyer, K., Bauer, M., Friedl, A., Ramakrishnan, S., Shridhar, V., 2006. HSulf-1 inhibits angiogenesis and tumorigenesis in vivo. Cancer Res. 66, 6025–6032. Ochiai, H., Sakamoto, N., Momiyama, A., Akasaka, K., Yamamoto, T., 2008. Analysis of cis-regulatory elements controlling spatio-temporal expression of T-brain gene in sea urchin, Hemicentrotus pulcherrimus. Mech. Dev. 125, 2–17. Okazaki, K., 1975. Normal development to metamorphosis. In: Czihak, G. (Ed.), The Sea Urchin Embryo: Biochemistry and Morphogenesis. Springer-Verlag, New York, pp. 177–232. Rast, J.P., 2000. Transgenic manipulation of the sea urchin embryo. Methods Mol. Biol. 136, 365–373. Ratzka, A., Kalus, I., Moser, M., Dierks, T., Mundlos, S., Vortkamp, A., 2008. Redundant function of the heparan sulfate 6-Oendosulfatases Sulf1 and Sulf2 during skeletal development. Dev. Dyn. 237, 339–353. Rong, J., Habuchi, H., Kimata, K., Lindahl, U., KuscheGullberg, M., 2001. Substrate specificity of the heparan sulfate hexuronic acid 2-O-sulfotransferase. Biochemistry 40, 5548–5555. Ro¨ttinger, E., Saudemont, A., Duboc, V., Besnardeau, L., McClay, D., Lepage, T., 2008. FGF signals guide migration of mesenchymal cells, control skeletal morphogenesis and regulate gastrulation during sea urchin development. Development 135, 353–365. Schmidt, B., Selmer, T., Ingendoh, A., von Figura, K., 1995. A novel amino acid modification in sulfatases that is defective in multiple sulfatase deficiency. Cell 82, 271–278. Sethi, A.J., Angerer, R.C., Angerer, L.M., 2009. Gene regulatory network interactions in sea urchin endomesoderm induction. PLoS Biol. 7, 0248–0264.

4 2 7 ( 2 0 1 0 ) 2 3 5 –2 4 5

245

Shworak, N.W., Liu, J., Fritze, L.M.S., Schwartz, J.J., Zhang, L., Logeart, D., Rosenberg, R.D., 1997. Molecular cloning and expression of mouse and human cDNAs encoding heparan sulfate D-glucosaminyl 3-O-sulfotransferase. J. Biol. Chem. 272, 28008–28019. Smeds, E., Habuchi, H., Do, A.T., Hjertson, E., Grundberg, H., Kimata, K., Lindahl, U., Kusche-Gullberg, M., 2003. Substrate specificities of mouse heparan sulphate glucosaminyl 6-Osulphotransferases. Biochem. J. 372, 371–380. 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. Spicer, A.P., Kaback, L.A., Smith, T.J., Seldin, M.F., 1998. Molecular cloning and characterization of the human and mouse UDP-glucose dehydrogenase genes. J. Biol. Chem. 273, 25117–25124. Tang, R., Rosen, S.D., 2009. Functional consequences of the subdomain organization of the sulfs. J. Biol. Chem. 284, 21505– 21514. Uchimura, K., Morimoto-Tomita, M., Bistrup, A., Li, J., Lyon, M., Gallagher, J., Werb, Z., Rosen, S.D., 2006. HSulf-2, an extracellular endoglucosamine-6-sulfatase, selectively mobilizes heparin-bound growth factors and chemokines: effects on VEGF, FGF-1, and SDF-1. BMC Biochem. 7, 2. Viviano, B.L., Paine-Saunders, S., Gasiunas, N., Gallagher, J., Saunders, S., 2004. Domain-specific modification of heparan sulfate by Qsulf1 modulates the binding of the bone morphogenetic protein antagonist Noggin. J. Biol. Chem. 279, 5604–5611. Wang, S., Ai, X., Freeman, S.D., Pownall, M.E., Lu, Q., Kessler, D.S., Emerson Jr., C.P., 2004. QSulf1, a heparan sulfate 6-Oendosulfatase, inhibits fibroblast growth factor signaling in mesoderm induction and angiogenesis. Proc. Natl. Acad. Sci. USA 101, 4833–4838. Wikramanayake, A.H., Peterson, R., Chen, J., Huang, L., Bince, J.M., McClay, D.R., Klein, W.H., 2004. Nuclear bcatenin-dependent Wnt8 signaling in vegetal cells of the early sea urchin embryo regulates gastrulation and differentiation of endoderm and mesodermal cell lineages. Genesis 39, 194–205. Yip, G.W., Ferretti, P., Copp, A.J., 2002. Heparan sulphate proteoglycans and spinal neurulation in the mouse embryo. Development 129, 2109–2119.