SEARCH

SEARCH BY CITATION

Keywords:

  • heparan sulfate;
  • Ndst;
  • hedgehog;
  • Fgf;
  • Wnt

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Disruption of heparan sulfate (HS) synthesis in vertebrate development causes malformations that are composites of those caused by mutations of multiple HS binding growth factors and morphogens. We previously reported severe developmental defects of the forebrain and the skull in mutant mice bearing a targeted disruption of the heparan sulfate-generating enzyme GlcNAc N-deacetylase/GlcN N-sulfotransferase 1 (Ndst1). Here, we further characterize the molecular mechanisms leading to frontonasal dysplasia in Ndst1 mutant embryos and describe additional malformations, including impaired spinal and cranial neural tube fusion and skeletal abnormalities. Of the numerous proteins that bind HS, we show that impaired fibroblast growth factor, Hedgehog, and Wnt function may contribute to some of these phenotypes. Our findings, therefore, suggest that defects in HS synthesis may contribute to multifactor types of congenital developmental defects in humans, including neural tube defects. Developmental Dynamics 236:556–563, 2007. © 2006 Wiley-Liss, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Heparan sulfate (HS) is produced by most mammalian cells as part of membrane and extracellular matrix proteoglycans (Esko and Lindahl, 2001). The chain grows by exostosin (Ext) copolymerization of GlcAβ1,4 and GlcNAcα1,4 and is modified by one or more of the four N-deacetylase/GlcN N-sulfotransferase (Ndst) isozymes; the N-deacetylase activity of Ndsts removes acetyl groups from GlcNAc residues, which are then converted to GlcNS through the N-sulfotransferase activity. Subsequent modifications of the HS chain by O-sulfotransferases and a GlcA C5-epimerase depend on the presence of GlcNS residues, making the Ndsts responsible for the generation of sulfated ligand binding sites in HS (Lindahl et al., 1998). Ndst1 and Ndst2 mRNA are expressed in all embryonic and adult tissues examined, whereas Ndst3 and Ndst4 transcripts are predominantly expressed during embryonic development and in the brain (Aikawa et al., 2001).

Many growth factors and morphogens bind to HS. In some cases, HS proteoglycans are thought to act as coreceptors for these ligands. Studies in Drosophila melanogaster demonstrated that HS is crucial for embryonic development (Perrimon and Bernfield, 2000) and that the fly Ndst ortholog Sulfateless affects signaling mediated by Wingless (Wg), Hedgehog (Hh), and fibroblast growth factor (Fgf; Lin et al., 1999; Lin and Perrimon, 1999; The et al., 1999). As those factors also play critical roles in morphogenesis, growth regulation, and differentiation, defective HS synthesis affects multiple aspects of vertebrate development. Mice deficient in Ext1, Ndst1, 2-Ost, and GlcA C5-epimerase function have defective brain morphogenesis, axon guidance defects, craniofacial defects, renal agenesis, and eye defects due to the simultaneous inhibition of multiple HS binding factors (Bullock et al., 1998; Inatani et al., 2003; Li et al., 2003; McLaughlin et al., 2003; Grobe et al., 2005). Here, we report additional developmental defects of the skeleton and the developing head in Ndst1-deficient embryos resulting from impaired function of Hh, Fgf, and possibly Wnt. Thus, multiple developmental processes and signaling pathways depend on HS, underlining the crucial role of regulated HS synthesis during development.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Ndst1 Null Embryos Show Defective Skull Development

Most Ndst1 null homozygotes developed to term but then died perinatally, showing various developmental deficiencies. In 6 of 10 embryos investigated by coronal sectioning of the head, palatal shelves were found to be elevated but fusion of those shelves was impaired. A total of 3% of Ndst mutant embryos (n = 2/62) had a median cleft of the primary palate and the nasal primordium (Fig. 1b,d), which is caused by incomplete merging of the two medial nasal prominences in the midline, indicating a severe midline defect underlying the facial phenotype. Moreover, lack of lower incisors was observed in 50% of Ndst1−/− embryos (n = 3/6, not shown). Other malformations affecting the developing skull were more commonly observed; coronal sections in embryonic day (E) 14.5–E18.5 Ndst1 null homozygotes revealed hypoplastic tongue, jaw, and nasal epithelium in all embryos (n = 6, Fig. 1f). Notably, all embryos showed strongly impaired craniofacial development, resulting in smaller/flattened snouts and hypoplastic skulls.

thumbnail image

Figure 1. Midline fusion of the nasal prominences and secondary palate formation is impaired in N-deacetylase/GlcN N-sulfotransferase 1 (Ndst1) mutant embryos, and proliferation/apoptosis is affected in the frontonasal mesenchyme and olfactory epithelium of Ndst1 mutant embryos. a,b: Ndst1 mutant embryonic day (E) 18.5 embryo showing a median cleft (arrow, b), littermate control (a). c,d: Horizontal sections through a,b, respectively, hematoxylin and eosin (H&E) stained. Ndst1 mutant embryos show a split philtrum (arrow), and the olfactory epithelia are hypoplastic and separated from each other. e,f: Coronal sections through the skull of E15.5 embryos: e, wild-type littermate control; f, Ndst1 mutant embryo. Note impaired fusion of the secondary palate (f, arrow) and lack of tongue in the Ndst1 mutant embryo. t, tongue; ns, cartilage primordium of nasal septum; ob, olfactory bulb; oe, olfactory epithelium. g–j: In general, cell proliferation in the head mesenchyme and olfactory epidermis is reduced to ∼50% of wild-type levels in Ndst1−/− embryos (n = 4, 52.8% ± 7%, P ≤ 0.01). g,h: E17.5 wild-type embryo, i,j: Ndst1−/− littermate control. Arrows show comparable regions. g,l: Nasal cavity. h,j: Mesenchyme in lateral area of the developing face. Cp, cartilage primordium of nasal septum; op, ossification in outer table of cartilage primordium of nasal septum; oe, olfactory epithelium. k–n: Increased levels of apoptotic nuclei (green) in E10.5 Ndst1 mutant embryos (m) and mutant E12.5 embryos (n) if compared with wild-type littermate controls (k,l). o,p: Loss of monoclonal anti-heparan sulfate (HS) antibody 10E4 reactivity (green) in Ndst1 mutant facial mesenchyme (p). Note the lack of proliferation in 10E4-negative structures such as the hair follicles (arrowheads, h,o) in contrast to high proliferation levels in 10E4-positive areas (arrows, h,o). Scale bar = 3 mm in a,b, 400 μm in c–f, 250 μm in h,j, 100 μm in k,m, 50 μm in l,n–p.

Download figure to PowerPoint

Ndst1-Deficient Embryos Show Reduced Cell Proliferation, Increased Apoptosis, and Impaired Fgf Binding to HS in the Developing Facial Mesenchyme

Our previous studies showed that mitogen-activated protein kinase (MAPK) signaling was strongly impaired in fibroblasts derived from Ndst1−/− embryos (Grobe et al., 2005), suggesting impaired Fgf-dependent proliferation as an underlying reason for the observed craniofacial defects. Indeed, the number of bromodeoxyuridine (BrdU) -labeled nuclei relative to total nuclei was found to be reduced to 53% ± 7% (P < 0.001; n = 5) in E17.5 mutant nasal epithelium and skin, confirming reduced cell proliferation in dysmorphic mutant heads (arrows, Fig. 1g–j). Of interest, the distribution of HS as detected by the anti-HS antibody 10E4 closely resembled the proliferation pattern (compare Fig. 1h with 1o). In addition to those findings, Terminal deoxynucleotidyl transferase–mediated deoxyuridinetriphosphate nick end-labeling (TUNEL) staining revealed increased levels of programmed cell death in E10.5, E12.5, and E15.5 maxillary primordia that give rise to the face (Fig. 1k–n and not shown). In E15.5 facial mesenchyme, the relative number of apoptotic cells was increased by 66% (P ≤ 0.001; n = 3).

To investigate whether reduced binding of Fgf to mutant HS underlies this finding, we next performed ligand blotting assays. Embryo sections (E10.5) were incubated with biotin-tagged Fgf-2, and Ffg binding was detected by biotin immunohistochemistry. As shown in Figure 2c, Fgf binding was specifically localized to the basement membranes and closely resembled the distribution of HS as detected by the anti-HS antibody 10E4 (Fig. 2a). In contrast, corresponding sections derived from Ndst1 mutant embryos had little detectable Fgf binding and 10E4 staining (Fig. 2b,d). Impaired Fgf/HS interaction in Ndst1-deficient E17.5 embryos was also shown by affinity chromatography. HS was isolated from the heads of wild-type and mutant embryos, coupled to Affi-Gel columns, and incubated with biotinylated Fgf-2. The elution profile demonstrated a significantly reduced capacity of mutant HS to bind Fgf-2, indicating a reduction of Fgf-2 high affinity binding sites (Fig. 2e).

thumbnail image

Figure 2. Impaired fibroblast growth factor-2 (Fgf-2) binding to HS present in N-deacetylase/GlcN N-sulfotransferase 1 (Ndst1) -deficient embryos. a–d: Sections of embryonic day (E) 10.5 nasal placode were incubated with anti-heparan sulfate (HS) antibody 10E4 (red, a,b) and with biotin-tagged Fgf-2 (brown, c,d). Fgf binding was specifically localized to the basement membranes and closely resembled the distribution of HS as detected by 10E4 (a,c). In contrast, corresponding sections derived from Ndst1 mutant embryos had little detectable Fgf binding and 10E4 staining (b,d). Heparitinase treatment of wild-type sections demonstrated specificity of the 10E4 antibody (not shown). e: More biotinylated Fgf-2 binds to E17.5 wild-type GAGs (red line) than to GAGs isolated from mutant embryos (blue line). Equal amounts of GAGs isolated from these embryos were coupled to Affi-Gel, loaded with biotinylated Fgf-2, and incubated with streptavidin–horseradish peroxidase (HRP). Fgf-2/streptavidin–HRP complexes were then eluted with increasing NaCl concentrations, ranging from 0 to 1 M in 50 mM increments.

Download figure to PowerPoint

Another group of proteins known to contribute to the formation of facial primordia belong to the Wnt family of signaling molecules (Yanfeng et al., 2003). Wnt proteins bind to receptors of the Frizzled and LRP families on the cell surface. The signal is transduced to stimulate β-catenin, which then enters the nucleus and forms a complex with TCF to activate transcription of Wnt target genes. To investigate the role of Ndst1 for Wnt signaling in the developing facial mesenchyme, a Wnt reporter based on a multimerized TCF binding site (TOPGAL; DasGupta and Fuchs, 1999) driving expression of LacZ was used (not shown). In wild-type (n = 2), heterozygous (n = 6), and Ndst1−/− (n = 6) embryos, similar TOPGAL transgene expression was observed in the developing face, indicating that the observed facial phenotype was unlikely to result from perturbed canonical Wnt signaling.

Taken together, these findings demonstrate that Ndst1 function is necessary for Fgf binding, Fgf-dependent signaling and proliferation during craniofacial development in the embryo. Also, as we and others (Ledin et al., 2004; Grobe et al., 2005) found that (undersulfated) HS was still being made in Ndst1 mutant embryos, we conclude that Ndst1 is responsible for the generation of specific HS structures recognized by the monoclonal antibody 10E4 that are not synthesized by other Ndst isoforms.

Defects of the Skeleton and Reduced Expression of the Osteogenic Marker Alkaline Phosphatase in HS-Deficient Osteoblast Precursor Cells

Development of the axial and appendicular skeleton was also found to be disrupted in Ndst1 mutant embryos. Alizarin red staining was used to detect mineralized tissues and revealed that all Ndst1 mutant mice investigated (n = 4) had delayed mineralization of vertebrae (Fig. 3b,d) and the long bones, especially metatarsal/metacarpal II–IV (Fig. 3i,j, inset). Closer examination also revealed severe hypoplasia of intervertebral discs in all mutant mice (Fig. 3d), which was confirmed by histological examinations of hematoxylin and eosin (H&E) -stained sagittal sections of the vertebral column. Here, the notochord-derived nucleus pulposus was found to be hypoplastic or completely missing (Fig. 3f, arrowhead). Also, all Ndst1 mutant mice investigated displayed a partially split sternum (n = 3, not shown).

thumbnail image

Figure 3. Defects of the axial skeleton in embryonic day (E) 18.5 N-deacetylase/GlcN N-sulfotransferase 1 (Ndst1) -deficient embryos and impaired heparan sulfate (HS) synthesis in C3H10T1/2 cells leads to reduced hedgehog-induced differentiation. Alcian blue and Alizarin red staining in whole embryos. a,b: Ossification is strongly reduced in vertebrae of Ndst1-deficient embryos (b). c,d: Closer examination revealed fusion of vertebrae in Ndst1−/− embryos (d, arrowhead) and strongly hypoplastic intervertebral discs (arrow, d). e,f: Fusion of vertebrae and strongly hypoplastic or missing nucleus pulposus as shown in sagittal sections through the spine of E18.5 Ndst1 mutant embryos (arrowhead, f). g: If maintained in Shh-conditioned medium for >5 days, C3H10T1/2 mouse mesenchymal precursor cells express the osteogenic marker alkaline phosphatase (AP, Shh-d5), which is not detectable before induction (Shh-d0), after 3 days (Shh-d3), or after culture in unconditioned media (c-d0, c-d3 and c-d5; P ≤ 0.01). Treatment with 25mM sodium chlorate (Shh-Ch), resulting in the production of undersulfated HS, or 3 mg/ml heparin (Shh-H) leads to a reduction of hedgehog-induced AP expression after 5 days (P ≤ 0.01 and P ≤ 0.1, respectively). Similarly, Ndst1 knockdown using shRNAi (Shh-i) leads to a complete inhibition of AP expression in the presence of Shh after 5 days, indicating an important role of Ndst1 in hedgehog-induced osteogenic differentiation (P ≤ 0.01). c–i: Empty vector control. Inset: Reverse transcriptase-polymerase chain reaction (RT-PCR) analysis demonstrated efficient Ndst1 knockdown in two siRNA expressing C3H10T1/2 cell lines (2,4) if compared with control cells (1,3). h: Semiquantitative RT-PCR analysis of Ndst1-4 expression in C3H10T1/2 cells (1–4) shows strongest expression of Ndst1 in undifferentiated and differentiated cells. A, actin control. One representative result from a total of five experiments is shown. Quantification of signal strength using ImageJ software revealed that Ndst2 product levels were 28% ± 8%, Ndst3 product levels were 13% ± 9%, and Ndst4 product levels were 10% ± 5% of Ndst1 product levels (P ≤ 0.001). i,j: HepSS-1 antibody detects sulfated HS in wild-type E18.5 metatarsal periosteum (arrows, i and inset; cortical bone of periosteal collar, asterisk) but not in the metatarsal periosteum of Ndst1 mutant littermates (j, arrowhead; inset shows lack of mineralized deposits in periosteal collar). k,l: Ptc expression is also strongly reduced in mutant periosteum (l, arrowhead), indicating impaired Indian hedgehog (Ihh) signaling when compared with wild-type periosteum (arrow, k). Scale bar = 1 mm in a,b, 500 μm in c–f, 25 μm in i,j, 50 μm in k,l.

Download figure to PowerPoint

To analyze the putative role of HS for bone mineralization in more detail, we investigated the ability of multipotent C3H10T1/2 precursor cells to differentiate into alkaline phosphatase (AP) -producing osteoblasts (mineralizing cells) after stimulation with soluble Shh protein (ShhN, amino acid residues 1–198, Fig. 3g; Kinto et al., 1997). C3H10T1/2 cells can serve as a model for early stages of endochondral bone formation, in which alkaline phosphatase-expressing osteoblastic cells derived from the periosteum establish and calcify an osteoid matrix. First, HS dependency of Hh-induced differentiation was assessed by sodium chlorate treatment of C3H10T1/2 cells, which led to significantly impaired AP production, as did the addition of heparin (Fig. 3g). Next, the role of Ndst1 in C3H10T1/2 cells was assessed by Ndst1 knockdown assays. Ndst1 shRNAi led to a complete inhibition of AP expression (Fig. 3g, Shh-i vs. Shh-d5), indicating an essential role of Ndst1 in osteogenic differentiation. To explain this severe effect, semiquantitative reverse transcriptase-polymerase chain reaction (RT-PCR) analysis of undifferentiated and differentiated C3H10T1/2 cells was conducted, revealing that Ndst1 was most strongly expressed, and only weak expression of Ndst2, Ndst3, and Ndst4 was seen (in five independent assays, Fig. 3h and not shown). Thus, we conclude that, in C3H10T1/2 cells, Ndst1 function is essential for osteogenic differentiation in response to Hh stimulation.

Immunohistochemical analysis of mutant and wild-type embryonic day (E) 18.5 metatarsal periosteum (a site of primary ossification) also indicated that loss of Ndst1 may result in impaired Hh function in vivo. Ndst1 mutants lacked a metacarpal/metatarsal periosteal collar (Fig. 3i,j, inset, asterisk) and had no HS immunoreactivity (Fig. 3j). Immunoreactivity against the hedgehog receptor Ptc was also markedly decreased in the mutant periosteal collar (Fig. 3l), indicating impaired Hh signaling in that tissue. We, therefore, conclude that loss of Ndst1 function in the periosteum may impair hedgehog-dependent osteogenic differentiation of osteoblasts, resulting in reduced primary ossification and cortical bone formation.

Neural Tube Closure Defects, Eye Defects, and Scoliosis in Ndst1-Deficient Embryos

Neural tube closure defects (NTDs) were also detected at low penetrance in these mutants: incomplete fusion of the neural tube in the thoracic/lumbar region was detected in three embryos (5%, n = 62, Fig. 4a) and exencephaly was detected in four embryos (6%, n = 62, Fig. 4b,e). Scoliosis (curved spine) was also observed in two embryos (3%, n = 62, Fig. 4e, arrowhead). Additionally, 12 Ndst1 mutant embryos (20%) showed unilateral eye loss, whereas the remaining 50 embryos showed highly variable defects, ranging from bilateral, complete lack of eyes (in 30% of mutant embryos) and micro-ophthalmia (20%) to bilateral lack of the eye lenses (50%). Of interest, unilateral eye loss/hypoplasticity was restricted to the right eye in all cases (P ≤ 0.001). In those embryos, histological analysis of the right half of the developing skull revealed the presence of malformed retinal tissue in various locations inside the head (Fig. 4c,d, arrow), indicating a selective inhibition of right eye development over left eye development. The malformed right eye consisted only of retina in all cases, and no lens could be detected (aphakia).

thumbnail image

Figure 4. Neural tube closure defects and craniofacial abnormalities in N-deacetylase/GlcN N-sulfotransferase 1 (Ndst1) mutant embryos. a: Impaired fusion of the neural tube in embryonic day (E) 16.5 Ndst1−/− embryos (right) compared with a wild-type littermate control (left). b: Exencephaly in an E18.5 Ndst1−/− embryo. c: Coronal section through b, hematoxylin and eosin (H&E) stained. Note the exposed neural tissue (arrow) and additional developmental defects, such as cleft palate and aberrant development of the right eye (arrow). d: Magnification of c. Note the presence of pigmented retina epithelium (arrowhead) and absence of the eye lens in the malformed right eye (arrow). e: Exencephaly (arrow) and scoliosis (arrowhead) in an E15.5 Ndst1−/− embryo. Exposed hindbrain and midbrain caused by failure of neural tube closure in the cephalic region. f: Sagittal section through e, H&E stained. Midbrain tissue and hindbrain tissue are exposed (arrows). g,h: Canonical Wnt signaling as indicated by TOPGAL reporter activity was reduced in the midbrain and hindbrain (arrowheads) of severely affected E13.5 Ndst1 mutant embryos (g, 33% of mutant embryos, arrowhead, n = 6). Notably, canonical Wnt signaling was unchanged in the face and forebrain (asterisk) of Ndst1 mutant embryos. Scale bar = 2 mm in a,b,d, 1 mm in c,f, 500 μm in d.

Download figure to PowerPoint

In contrast to the developing face, analysis of TOPGAL reporter staining revealed reduced canonical Wnt signaling in the presumptive midbrain, hindbrain, and spinal cord of 30% of E13.5 Ndst1−/− embryos (Fig. 4g,h, in two of six mutant embryos). Although staining was variable among mutant embryos, a comparable loss of staining was never observed in wild-type or heterozygous littermates (n = 8). Notably, β-Gal expression was unchanged in the (hypoplastic) forebrain and face. Thus, we conclude that, within a subgroup of Ndst1−/− embryos, reduced canonical Wnt signaling is observed in restricted areas, indicating a direct or indirect role of Ndst1 modified HS for Wnt signaling.

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

HS is known to bind numerous growth factors and morphogens in different tissues during development. Of interest, Ndst1 mutant mice display developmental defects that mainly resemble those found in embryos made deficient for two families of growth factors: the Fgfs and Hhs (Table 1). Previously, we showed genetic interaction between Shh and Ndst1 and reduced Ptc expression in facial Ndst1−/− mesenchyme as well as strongly reduced MAPK activity after Fgf-2 stimulation of Ndst1−/− mesenchymal fibroblasts (Grobe et al., 2005). In this work, we confirmed impaired Fgf function in the developing mutant skull in agreement with the established role of Fgf-8 in skull and facial development (Meyers et al., 1998; Trumpp et al., 1999; Abu-Issa et al., 2002; Frank et al., 2002). Notably, Fgf-8 mutants also share neural and limb defects similar to those found in Ndst1 mutant embryos. Moreover, impaired function of various Fgf proteins during eye development was recently found (Pan et al., 2006), confirming that Ndst1 regulates Fgf signaling. However, the striking finding of unilateral eye loss in Ndst1 mutant embryos cannot be explained by the impaired function of single HS binding factors but instead are likely to result from impaired function of numerous factors involved in the development of the vertebrate skull and eyes.

Table 1. Developmental Defects Displayeda
Tissue or organPhenotype in Ndst1−/− embryosSimilarity to known mutants
  • a

    Many of the deficiencies found in Ndst1 mutant embryos have also been described in mouse mutants of other HS-synthesizing proteins or HSPG core proteins, as well as HS-binding soluble factors, their receptors or signal transduction molecules. Notably, many of the phenotypes found in Ndst1 mutant mice are also prominent in mice made deficient in Hh/Gli and Fgf/FgfR signaling. HS, heparan sulfate; HSPG, heparan sulfate proteoglycan; Fgf, fibroblast growth factor; NTD, neural tube closure defect; Shh, Sonic hedgehog; n.f.: not found.

Neuroectoderm derivativesHypoplastic forebrain (100%)2-Ost (McLaughlin et al., 2003), Ext (Mitchell et al., 2001; Inatani et al., 2003), Fgf-2 (Dono et al., 1998; Raballo et al., 2000)
Brain  
 Lack of commissures (100%)Ext (Inatani et al., 2003)
 Eye defects (100%)2-Ost (Bullock et al., 1998), Epimerase (Li et al., 2003)
 Infrequent exencephalyPerlecan (Arikawa-Hirasawa et al., 1999)
 Infrequent NTDn.f.
Spinal cord  
Neural crest derivatives  
 Tongue frequently missingShh (Jeong et al., 2004)
Craniofacial bonesHypoplasia of NC derived bones (14%)Shh (Chiang et al., 1996; Jeong et al., 2004)
Craniofacial prominencesHypoplastic maxillae/mandible (100%)Shh (Jeong et al., 2004), Fgf-8 (Trumpp et al., 1999), Gli2 (Mo et al., 1997)
1st arch derivativesFrequent secondary cleft palate (75%)Fgf-8 (Trumpp et al., 1999; Abu-Issa et al., 2002; Frank et al., 2002; Macatee et al., 2003; Rice et al., 2004), Perlecan (Arikawa-Hirasawa et al., 1999), Fgf-10/FgfR2b (Rice et al., 2004), Gli2/Gli3 (Mo et al., 1997), TGFβ3 (Taya et al., 1999), Shh (Rice et al., 2004)
Mesoderm derivatives  
Lateral plate mesodermPartially split sternum (100%)n.f.
SclerotomeVertebrae fusion and delayed ossification (100%)Perlecan (Arikawa-Hirasawa et al., 1999), Ext (Koziel et al., 2004), Gli2 (Mo et al., 1997)
NotochordHypoplastic nucleus pulposus (100%)Gli2 (Mo et al., 1997), Ext (Koziel et al., 2004), Shh (Chiang et al., 1996)
LungNot functional (100%)Perlecan (Arikawa-Hirasawa et al., 1999)
Limb bud mesenchymeInfrequent syndactylyExt (Koziel et al., 2004)
 Delayed ossification (100%)Gli2 (Mo et al., 1997), Ext (Koziel et al., 2004)

In contrast to Fgf and Hh function, canonical Wnt signaling was not found to be significantly changed during facial development in all Ndst1−/− embryos investigated. In total, these results indicate that Ndst1 deficiency impairs numerous soluble, HS-binding protein factors during the formation of the skull, among those, Fgfs and Hhs.

Impaired Hh function may also underlie defects in skeletal development. We hypothesized that undersulfated HS in Ndst1 mutant embryos, by providing limited Ihh binding sites, may influence Ihh signaling in the embryonic spine and limb digits, resulting in delayed mineralization in the osteoblastic layer. This hypothesis was supported by four findings: First, mouse Gli2 mutants (a transcription factor that mediates intracellular hedgehog effects) have cleft palate, delayed ossification of the digits, no or little ossification of vertebrae, and lack of intervertebral discs (Mo et al., 1997), all of which are also commonly found in Ndst1 mutant embryos (Table 1). Second, our previous work showed that binding of a recombinant Shh fusion protein to HS derived from Ndst1 mutant embryos is reduced (Grobe et al., 2005), indicating direct hedgehog–HS interaction. Third, mouse embryos made deficient in Ndst2 or Ndst3 function did not show delayed ossification (not shown). Lastly, Ndst1 is the predominant Ndst expressed in multipotent C3H10T1/2 mouse mesenchymal precursor cells, which, under the induction of Hhs, differentiate into AP-producing osteoblasts (Kinto et al., 1997). The importance of Ndst1 in Hh-induced C3H10T1/2 differentiation was shown by siRNAi, and the general importance of HS in that system was confirmed by chlorate treatment or addition of heparin, which resulted in undersulfated cell surface HS (Safaiyan et al., 1999) or competing Hh binding sites in the medium, respectively, resulting in reduced availability of soluble Hh to cell surface HS. Because Ndst1 mutant embryos display some but not all defects seen in Shh mutant embryos (Grobe et al., 2005), Ndst1 does not seem to be always necessary for Hh signaling, but becomes crucial in the absence of other, possibly compensating Ndsts. In contrast, strong expression of all Ndst isoforms in other systems may allow for sufficient HS sulfation in the absence of Ndst1. Consistent with this possibility, Ndst1;Ndst2 and Ndst1;Ndst3 double mutants exhibited much more severe phenotypes than either single mutant (Grobe et al., 2002; and unpublished observations). However, lack of unique HS epitopes generated by Ndst1 (as shown by general loss of 10E4 reactivity in the Ndst1 mutant embryo at all stages) indicates that compensation for Ndst1 activity may only be partial.

Ndst1 mutant mice also have several informative low-frequency malformations, including cleft formation of the primary and secondary palate and median clefts. Cleft lip and cleft palate are common multifactor birth defects. Notably, among those HS binding factors, their receptors and transcription factors, deficient function of members of the Fgf- and Hh families often results in cleft formation that can also be observed in HS-deficient embryos (Frank et al., 2002; Rice et al., 2004; reviewed in Jiang et al., 2006). Mild posterior cleft deformities could be detected in 75% of all Ndst1 mutant mice investigated and 3% of Ndst1 mutant mice also display median cleft lip, a rare abnormality in humans related to defective Shh signaling. A second group of low-frequency midline defects found in Ndst1 mutants are neural tube closure defects (NTDs). Although NTDs are the second most common human birth defect, with an incidence of approximately 1/1,000 live births, the underlying genetic basis of NTDs is poorly understood. More than 90 mutations in a variety of genes have been identified and linked to rodent NTDs, mostly demonstrating variable low penetrance and complex inheritance patterns. NTDs originate in impaired rising of elevation zones that depends on the highly conserved Wnt/frizzled signal transduction pathway (for a review, see Copp et al., 2003). HS is known to bind to Wnt/Wg (Ai et al., 2003), and notably, reduced Wnt signaling was observed in the presumptive midbrain, hindbrain, and spinal cord of 30% of the Ndst1−/− embryos. This finding indicates that NTDs in Ndst1-deficient embryos may (partially) be based on impaired Wnt signaling. Additionally, Shh mutations as well as mutations for the hedgehog-dependent transcription factor Gli3 have been described to be associated with the NTD exencephaly (anencephaly; Harris and Juriloff, 1997). However, due to their low penetrance, molecular mechanisms leading to clefting and NTDs were not assessed in more detail.

Taken together, we conclude that Ndst1 plays a significant role in the synthesis of HS necessary for neural tube closure, fusion of the primary and secondary palate, bone formation, and development of the face/skull. However, incomplete penetrance and strong variability for some of these developmental defects is observed, possibly because developmental processes in the Ndst1 mutant mouse may be (partially) rescued by the expression of compensatory Ndst isoforms. Among the multiple heparan sulfate-binding factors potentially affected, impaired Fgf signaling could be demonstrated in the developing face, impaired Hh signaling in osteoblast differentiation and, in approximately 30% of Ndst1 mutant embryos, also reduced Wnt signaling in the presumptive mid/hindbrain area.

EXPERIMENTAL PROCEDURES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Histology and Detection of mRNA Expression

The generation of the Ndst1 allele has been described previously (Grobe et al., 2005). TOPGAL mice were obtained from Jackson Laboratory (DasGupta and Fuchs, 1999). Embryos were fixed in 4% paraformaldehyde overnight, dehydrated, embedded in paraffin, and sectioned at 8 μm. Alternatively, frozen sections were prepared after embryos were fixed in 4% paraformaldehyde, cryoprotected in 30% sucrose, embedded in OCT compound, and sectioned on a Leica1900 cryostat. Sections were stained with H&E for histological analysis. Cartilage and bone were stained with Alcian blue and Alizarin red in whole embryos.

For HS detection, mouse monoclonal antibody 10E4 and HepSS1 (Seikagaku, Tokyo, Japan) was used at a 1:100 dilution in phosphate buffered saline (PBS)/5% milk powder at 4°C overnight. Before antibody incubation, epitope retrieval was performed by boiling slides in 0.1 M sodium citrate, pH 6.0, for 20 min, followed by three washing steps in PBS. Control slides were heparinase I–III digested (Sigma, St. Louis, MO; 50 mM Hepes, 100 mM NaCl, 1 mM CaCl2, 5 μg bovine serum albumin/ml, pH 7.0) for 1 hr at 37°C to prove antibody specificity. Alternatively, HS fractions derived from matching embryonic stages were added to the antibody incubation. After antibody incubation, slides were washed three times for 10 min each in PBS before incubation with secondary Alexa- or fluorescein isothiocyanate–labeled α-IgG H+L antibody (1:200, Molecular Probes, Eugene, OR). The α-glial fibrillary acidic protein antibodies (1:100, Sigma) were used in Tris buffered saline, 0.05% Tween, 5% milk powder, after deparaffinization, epitope retrieval, and washing as described above. BrdU-labeled nuclei were quantified using α-BrdU antibodies (Zymed) on two mutant and wild-type E17.5 embryos. BrdU (70 μg/g mouse) was injected intraperitoneally, and mothers were killed after 1 hr. The number of BrdU-labeled cells relative to nonlabeled cells in the olfactory epithelium and skin was determined. Two different horizontal levels of the embryonic face, 250 μm apart, were assessed in each embryo. TUNEL assays were performed using the In Situ Cell Death Detection Kit (Roche, Mannheim, Germany), according to the manufacturer's instructions.

Fgf Ligand Binding Assay

Frozen sections were incubated in 0.5 mg/ml NaBH4 for 10 min and 0.1 M glycine for 30 min. Embryo sections were quenched with 2% H2O2 and blocked using 0.05% TSA blocking reagent. Biotinylated Fgf-2 was incubated with sections at 4°C overnight. The bound Fgf-2 was detected using a Vectastain ABC Kit (Vector Labs, Burlingame, CA) and stained in diaminobenzidine solution (Sigma).

AP Activity Assay of C3H10T1/2 Cells

C3H10T1/2 cells were maintained in DME medium (Invitrogen, Carlsbad, CA), pH 7.6, containing 2 mM glutamine, 10% fetal calf serum, and 100 U/ml penicillin, and 100 μg/ml streptomycin sulfate. For ShhN expression, nucleotides 1–594 (corresponding to amino acid residues 1–198) were produced by PCR using primers Shh-F (5′-aaaagcttatggggcccggcagggggtttg-3′) and Shh-R (5′-ttggatccccgccgccggatttggccgcc-3′) using a murine Shh cDNA template. PCR products were gel-purified (QiaQuick Gel Purification Kit, Qiagen, Hilden, Germany), cloned in pGEM (Promega Corporation, Madison, WI), and sequenced. After restriction digest using BamHI and HindIII, the fragment coding for Shh was ligated in pcDNA3.1myc/hisC (Invitrogen) and B16-F1 mouse melanoma cells were transiently transfected using PolyFect (Qiagen). Conditioned, ShhN-containing medium and control medium derived from mock transfected B16-F1 cells were harvested after 36 hr, sterile filtered, and applied to 8 × 105 C3H10T1/2 cells in 60-mm plates. A total of 25 mM sodium chlorate (Sigma) and 3 mg/ml heparin were added to determine the role of HS for C3H10T1/2 differentiation. Cells were lysed after 0 days, 3 days, and 5 days after induction (20 mM Hepes, 150 mM NaCl, 0.5% Triton X-100, pH 7.4), and AP activity was measured at 405 nm by addition of 120 mM p-nitrophenol (Sigma) in 0.1 M glycine buffer, pH 10.4.

RT-PCR Analysis of mRNA Expression

For RT-PCR analysis of mRNA derived from undifferentiated (AP-negative) and differentiated (AP-positive) C3H10T1/2 cells, RNA was isolated using TriZol reagent (Invitrogen) according to the manufacturer's instructions. cDNA was produced using the First-Strand cDNA Synthesis Kit (Fermentas, Burlington, Canada) according to the manufacturer's instructions and used as a PCR template under the following conditions: 1/10 of cDNA (derived from 1 μg total RNA), hot start for 5 min, 94°C for 20 sec, 59°C for 30 sec, 72°C for 2 min, 35 cycles. For the specific amplification of mNdst1-4, eight specific, intron-spanning primers (Tm = 62°C) were used: mNDST1-F(5′-cttgagccctcggcagatgc- 3′)and -R(5′-ccagggtactcgttgtagaag-3′), mNDST2-F (5′-aggaacccttgcctctgccc-3′) and -R (5′-gatcgtgtgggtgaagaggc-3′), mNDST3-F (5′-gaaagtgaagtctctgggcgg-3′) and -R (5′-tccgtgaatactcttgtccag-3′), mNDST4-F (5′-aacaggaaatgacacttattgaaacg-3′) and -R (5′-aggtgtataagccgaggcgg-3′).

shRNAi for Ndst1

RNA interference (RNAi) was used to specifically knock down Ndst1 mRNA expression in C3H10T1/2 cells. Small interfering RNAs (siRNAs) were expressed from short hairpin RNAs (shRNAs) using the MISSION shRNA plasmid TRCN0000008646 (Sigma), targeting nucleotides 2186–2206 of murine Ndst1 (NM_008306). Transfected cells were enriched in the presence of 5 μg/ml puromycin for 2 days before induction of differentiation. C3H10T1/2 cells transfected with empty shRNA vector pSHAG-MAGIC2 served as a control. Efficacy of Ndst1 mRNA knockdown was assessed by RT-PCR in three independent experiments, followed by quantification of PCR products using ImageJ (http://rsb.info.nih.gov/ij/), resulting in >70% reduction in all cases.

Preparation of HS

The heads of two mutant and wild-type E17.5 embryos were pooled, digested with 2 mg/ml pronase in 320 mM NaCl, 100 mM sodium acetate (pH 5.5) overnight at 40°C, diluted 1:3 in water, and applied to a 2.5-ml column of DEAE Sephacel. After washing the column with 0.3 M NaCl, the glycosaminoglycans were eluted with 1 M NaCl. The glycosaminoglycan pool was applied to a PD-10 (Sephadex G25) column (Pharmacia, Uppsala, Sweden). Glycosaminoglycans eluting in the void volume were lyophilized, purified on DEAE as described above, again applied to a PD-10 column and lyophilized.

Fgf-HS Binding Assay

Approximately 200 μg of purified lyophilized glycosaminoglycans were covalently coupled to Affi-Gel 10 (Bio-Rad, Hercules, CA) according to the manufacturer's instructions and the extent of coupling was determined by Carbazole reaction. Biotinylated Fgf-2 was applied to the columns, followed by the application of streptavidin–horseradish peroxidase (HRP) after a washing step to remove unbound Fgf-2. HS-bound Fgf–biotin–streptavidin–HRP complexes were then eluted using a NaCl gradient ranging from 0–1 M in 0.1 M sodium acetate buffer (pH 6.0). HRP activity was measured at 490 nm in 0.5 mg/ml diaminobenzidine, 0.02% hydrogen peroxide in PBS, pH 7.4.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

The authors thank Dr. Andrew McMahon (Harvard University, Cambridge, MA) for Shh cDNA. K.G. was funded by the DFG. J.D.E. was supported by National Institutes of Health grant GM 33063. The authors state that they have no competing interests.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES