SEARCH

SEARCH BY CITATION

Keywords:

  • HSPG;
  • Ndst;
  • sulfotransferase;
  • Fgf;
  • optic disc;
  • optic stalk

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental Procedures
  7. References

Background: Heparan sulfate proteoglycans (HSPG) are important for embryonic development by means of the regulation of gradient formation and signaling of multiple growth factors and morphogens. Previous studies have shown that Bmp/Shh/Fgf signaling are required for the regionalization of the optic vesicle (OV) and for the closure of the optic fissure (OF), the disturbance of which underlie ocular anomalies such as microphthalmia, coloboma, and optic nerve hypoplasia. Results:To study HSPG-dependent coordination of these signaling pathways during mammalian visual system development, we have generated a series of OV-specific mutations in the heparan sulfate (HS) N-sulfotransferase genes (Ndst1 and Ndst2) and HS O-sulfotransferase genes (Hs2st, Hs6st1, and Hs6st2) in mice. Of interest, the resulting HS undersulfation still allowed for normal retinal neurogenesis and optic fissure closure, but led to defective optic disc and stalk development. The adult mutant animals further developed optic nerve aplasia/hypoplasia and displayed retinal degeneration. We observed that MAPK/ERK signaling was down-regulated in Ndst mutants, and consistent with this, HS-related optic nerve morphogenesis defects in mutant mice could partially be rescued by constitutive Kras activation. Conclusions: These results suggest that HSPGs, depending on their HS sulfation pattern, regulate multiple signaling pathways in optic disc and stalk morphogenesis. Developmental Dynamics 243:1310–1316, 2014. © 2014 Wiley Periodicals, Inc.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental Procedures
  7. References

Optic nerve hypoplasia and coloboma are important causes of childhood visual disorders and blindness. Although it is clear that these congenital ocular deformations result from failures of optic fissure closure and optic nerve development, the genetic bases of these developmental defects are only poorly understood (Chang et al., 2006; Ghiasvand et al., 2011).

Multiple morphogenetic pathways, including Sonic hedgehog (Shh), bone morphogenetic protein (Bmp), retinoic acid (RA), fibroblast growth factor (Fgf), and Wnt signaling, are required for mammalian retinal development. During early retinal development, the proximal–distal patterning of optic vesicle is controlled by midline-derived Shh, while the dorsal–ventral patterning is regulated by Bmp and RA signaling (Chiang et al., 1996; Furuta and Hogan, 1998; Matt et al., 2008; Lupo et al., 2011). Homozygous Shh null mutant mice exhibit cyclopia and optic nerve aplasia (Chiang et al., 1996), and Bmp7 knockout mice display coloboma and optic nerve and optic disc aplasia (Morcillo et al., 2006). Once the optic cup is formed, Fgf and Wnt signaling are required for the regionalization of the presumptive neural retina (NR) and the retinal pigmented epithelium (RPE), respectively. Specific depletion of the Wnt signaling component β-catenin in the dorsal optic cups leads to the failure of RPE specification (Westenskow et al., 2009). In recent studies on Fgf signaling, we and others have shown that Fgf signaling is required not only for neural retina determination but also for optic disc and optic fissure development (Cai et al., 2010, 2013; Chen et al., 2013).

To study the coordination of the various signaling pathways during embryonic development, we have previously used mammalian lens and lacrimal gland as models to investigate the role and mechanisms of heparan sulfate proteoglycans (HSPGs) (Pan et al., 2006, 2008; Qu et al., 2011a,b, 2012). As a conserved and crucial part of the extracellular matrix, HSPGs are functionally involved in the regulation of multiple intercellular signaling molecules, including but not restricted to Shh, Wnt, and Fgf, during organ morphogenesis. HSPGs consist of glycoprotein cores covalently linked to heparan sulfate (HS) glycosaminoglycan chains. Nascent HS consisting of alternating residues of glucuronic acid and N-acetyglucosamine is initially modified by N-deacetylation of N-acetylglucosamine residues, followed by their subsequent N-sulfation. Both reactions are catalyzed by one or more of the four N-deacetylase-N-sulfotransferase (Ndst) family members. Importantly, HS N-sulfation serves as an essential prerequisite for subsequent HS modifications by heparan sulfate 2-O, 3-O, and 6-O sulfotransferases (Hs2st/Hs6st) and one epimerase, that together create ligand-specific sulfated binding sites on the proteoglycan-linked HS chain (Bishop et al., 2007). Consistent with this, we previously found that FGF-regulated early lens and lacrimal development was disrupted in Ndst and Hs2st/Hs6st mutants, whereas BMP and Wnt signaling appeared unaffected (Pan et al., 2006, 2008; Qu et al., 2011a,b, 2012).

In the present study, we examined the role of HSPGs in retinal development by specifically disrupting Ndst1/Ndst2 and Hs2st/Hs6st in early optic vesicle development. The resulting HS sulfation mutants exhibited normal retinal neurogenesis and optic fissure closure, but defective optic disc and stalk development, resulting in optic nerve hypoplasia and aplasia. Whereas we have previously shown that constitutive Kras activity fully rescued optic nerve dysgenesis caused by the loss of Fgf-Frs2-Shp2 signaling (Cai et al., 2013), similar optic nerve defects in the HS sulfation mutants were only partially ameliorated by activated Kras signaling. Therefore, our results show that HS is required for other signaling pathways in addition to the established FGF-MAPK signaling in optic disc and stalk morphogenesis.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental Procedures
  7. References

Combined Ablation of Ndst1/Ndst2 or Hs2st/Hs6st1/Hs6st2 HS Biosynthetic Genes in the Optic Vesicle Caused Optic Nerve Dysgenesis and Retinal Degeneration

We have previously shown that two of the four Ndst genes, Ndst1 and Ndst2, are expressed in E12.5 mouse retinae (Pan et al., 2006). The systemic deletion of Ndst1 function in mice is lethal at birth, however, Ndst2 knockout animals are grossly normal except for a mast cell specific defect (Forsberg et al., 1999; Grobe et al., 2005). Ndst1 systemic knockout embryos exhibit ocular phenotypes ranging from small eyes to a complete lack of eyes. However, the remaining retinae in either Ndst1 or Ndst2 deficient animals are normal, suggesting that Ndst genes play redundant roles in retina development (Pan et al., 2006). To test this idea, we first generated Ndst1 conditional mutants by using Six3-Cre, which is an optic vesicle-specific deleter active as early as at embryonic day (E)9.5 (Cai et al., 2010, 2013). Six3-Cre;Ndst1flox/flox were viable and fertile without any obvious ocular phenotypes (data not shown). The compound Six3-Cre;Ndst1flox/flox;Ndst2KO/KO mutants also showed normal eye size and a fully fused ventral ocular structure (Fig. 1A–D). Closer examination of adult eyes, however, revealed hypoplasia or aplasia of optic nerves (100% penetrant) (Fig. 1E–H; Table 1). The thickness of retinae and the number of retinal ganglion cells at 4 weeks were reduced by 39% and 50%, respectively (P < 0.001; n = 10). These results indicate that the two HSPG N-deacetylase/N-sulfotransferase enzymes are required for optic nerve development and retinal homeostasis.

image

Figure 1. Comparable ocular defects in Six3-Cre; Ndst1flox/flox; Ndst2KO/KO and Six3-Cre; Hs6st1flox/flox; Hs6st2KO/KO; Hs2stflox/flox mice. A–H: Conditional Ndst1/Ndst2 mutants displayed normal closure of the optic cup in the front (A–D), but optic nerve hypoplasia or aplasia at the back side of the eye (E,F, arrow). Retinal thinning was evident at adult stages (G,H). I–P: Hs2st/Hs6st1/Hs6st2 conditional mutants also showed hypoplastic optic nerves and degenerating retinae.

Download figure to PowerPoint

Table 1. Ocular Phenotypes in Ndst and Hs6st/Hs2st Mutants
  Ocular Phenotypes 
 ColobomaON aplasiaON hypoplasian
Control0%0%0%20
Six3-Cre;Ndst1flox/flox0%0%0%20
Six3-Cre;Ndst1flox/flox;Ndst2KO/KO0%90%10%40
Six3-Cre;Hs2stflox/flox0%0%0%20
Six3-Cre;Hs6st1flox/flox;Hs6st2KO/KO0%0%0%20
Six3-Cre;Hs2stflox/flox;Hs6st1flox/flox;Hs6st2KO/KO0%25%75%20
Rx-Cre;Ndst1flox/flox;Ndst2KO/KO100%100%0%10
Six3-Cre;Ndst1flox/flox;Ndst2KO/KO;LSL-KrasG12D0%5%95%40

To investigate the function of HSPG O-sulfation enzymes in retinal development, we also generated Hs2st/Hs6st compound mutants, lacking the single 2-O sulfotransferase gene and up to two of the three 6-O sulfotransferase genes. Although single or double ablation of Hs2st, Hs6st1, and Hs6st2 by Six3-Cre did not affect ocular development, possibly due to compensatory up-regulation of 6-O sulfation in response to abolished 2-O sulfation (data not shown) (Merry et al., 2001; Merry and Wilson, 2002), we observed significant optic nerve defects and retinal degeneration in Six3-Cre;Hs6st1flox/flox;Hs6st2KO/KO;Hs2stflox/flox animal (also 100% penetrant) (Fig. 1I–P; Table 1). These results support that HS O-sulfation is a critical determinant for retinal development. Because Ndst1/Ndst2 and Hs2st/Hs6st1/Hs6st2 mutants share similar ocular defects, we chose Six3-Cre;Ndst1flox/flox; Ndst2KO/KO for further analysis.

Normal Neurogenesis in Ndst1/Ndst2 Mutants

Optic nerve development is dependent on proper differentiation of retinal ganglion cells, which project axons through the optic disc to connect to the brain. To determine the molecular mechanisms of the observed optic nerve defects in HS-sulfation mutants, we analyzed the expression of key transcriptional factors required for retinal development. At E11.5, the RPE marker Mitf is properly restricted to the outer layer of the optic vesicle and the ventral retina, while the NR marker Chx10 was expressed correctly in the inner layer of the optic vesicle (Fig. 2A–D). Similarly, normal expressions of Otx1 was detected in the distal retinae and the RPE differentiation marker Dct1 was confined to the RPE (Fig. 2E–J). These results suggested that the regionalization of optic vesicles was unaffected in Ndst1/Ndst2 mutants. Sagittal sections of E13.5 embryos showed that mutant eyes were fused at the Pax2-positive ventral retina, consistent with a lack of coloboma defects in these mice (Fig. 2K–N, arrows). In transverse section, however, Pax2 was already down-regulated in the presumptive optic disc region (Fig. 2O,P, arrows). Sox2 is a transcriptional factor for neural retinal progenitors, while Math5 and Brn3b are specifically required for ganglion cell development. We found that these transcriptional factors are expressed in expected spatial patterns in Ndst1/Ndst2 mutants (Fig. 2Q–T and data not shown). Additionally, the cell proliferation markers Cyclin D1 and Ki-67 were also expressed indistinguishably between control and mutant retinae (Fig. 2U–X). These results suggest that the optic nerve dysgenesis in Ndst1/Ndst2 mutants is not due to defects in the optic vesicle patterning or ganglion cell differentiation.

image

Figure 2. Molecular markers for earlier neural retina development in Ndst mutants. A–J: The NR marker Chx10, RPE markers Mitf and Dct1, and ciliary margin marker Otx1 are expressed normally in Ndst1/Ndst2 conditional mutants. K–N: Sagittal sections of Ndst1/Ndst2 mutants showed that the optic cup is fused at the ventral side (arrows). O,P: Transverse sections showed that Pax2 was reduced in the optic disc region (arrows). Q–T: Sox2 and Brn3b expression indicates unaffected differentiation of retinal ganglion cells in conditional Ndst1/Ndst2 mutants. U–X: No significant differences in the expression of cell proliferation markers Cyclin D1 and Ki-67 were observed in eyes of E13 Ndst1/Ndst2 conditional mutants.

Download figure to PowerPoint

Failure of Optic Disc and Stalk Morphogenesis in HS-Sulfation Mutants

Because Ndst1/Ndst2 mutants showed normal neural retina determination and ganglion cell genesis, we next analyzed the growth of ganglion cell axons and the development of the optic discs. The optic discs are composed of Pax2-positive astrocytes, which express Netrin-1 to guide the projection of retinal ganglion cell axons. Double staining of Netrin-1 by RNA in situ hybridization and Pax2 by immuno-fluorescence demonstrated that Ndst1/Ndst2 mutants had reduced numbers of optic disc cells in central retinae, which were also grossly disorganized (Fig. 3A–D″ and Table 2).

image

Figure 3. Disrupted optic disc and stalk development in Ndst and Hs6st/2st conditional mutant mice. A–D″: The optic disc markers Netrin-1 and Pax2 were down regulated in E14 Ndst1/Ndst2 mutants. Arrows indicate Netrin-1 and Pax2-expressing optic discs. E–N″: Abnormal development of optic stalks and nerves was observed in both E14 Ndst and Hs6st/2st mutants, as indicated by altered expression of NF-165 and Pax2. Arrowheads denote the misrouting of retinal ganglion cell axons. Optic stalks were marked by circling the cells surrounding the NF-165-expressing neural fibers in dashed white lines (L,N).

Download figure to PowerPoint

Table 2. Quantification of the Optic Disc Defectsa
 pERK (average pixel intensity/retina)Pax2 (cells/section)NF-165 (area /section)Netrin-1 (cells/section)Brn3b (cells/section)
  1. a

    The number of Pax2, Netrin-1, and Brn3b positive cells were counted on immunostained sections. ImageJ software was used to measure the area of NF-165 positive region at the optic stalk and the average pixel intensity of phospho-ERK (pERK) immunofluorescence in the retina. Value are mean ± SEM and statistical significance determined by one-way ANOVA analysis. *P < 0.01, compared to wild-type control. **P < 0.01, compared to Ndst1/Ndst2 mutants. N.D., not determined. N.S., not statistically significant.

Control24 ± 2129 ± 7205 ± 7113 ± 4527 ± 14
Six3-Cre; Ndst1flox/flox; Ndst2ko/ko15 ± 1*22 ± 4*13 ± 2*21 ± 4*515 ± 23N.S.
Six3-Cre; Ndst1flox/flox; Ndst2ko/ko; LSL-KrasG12D28 ± 2**49 ± 5**45 ± 4**N.D.N.D.
n51010105

We examined the pathway-finding of axons within retinae and the optic stalks using NF-165, an neurofilament protein marker that traces each optic nerve fiber. We found the retinal ganglion cell axons in Ndst1/Ndst2 mutants were misrouted to the sub-retinal space, apparently because they were unable to properly project into the optic disc and stalk (Fig. 3E–H″). Similar defects were observed in Hs2st/Hs6st1/Hs6st2 mutants (Fig. 3I–J″), suggesting that the specification of the optic disc and guidance of retinal ganglion axons both require HS O-sulfations, consistent with the described strong reduction of HS 2-O- and 6-O-sulfation in Ndst1/Ndst2 compound mutant cells and tissues (Holmborn et al., 2004; Grobe et al., 2005; Raman et al., 2011; Sheng et al., 2011). Although analysis of Pax2 and NF-165 expression on the sagittal sections demonstrated that optic fissures were closed properly in Ndst1/Ndst2 mutants, only a few axons were detected in optic stalks in these mice (Fig. 3K–N″). These results suggest that optic disc and stalk morphogenesis were disrupted during embryogenesis of Ndst1/Ndst2 mutants.

Disrupted Fgf-Fgfr Association in HS-Sulfation Mutants

The ocular phenotypes in Ndst1/Ndst2 mutants are highly similar to what we have reported in mouse mutants that have lost either Fgfr1 and Fgfr2 or their downstream mediators Frs2a and Shp2 (Cai et al., 2013). To determine whether Ndst1/Ndst2 mutations disrupted FGF signaling, we next analyzed the Fgf-Fgfr binding activity by the ligand and carbohydrate engagement (LACE) assay. In this assay, recombinant Fgf10 was incubated with an Fgfr2-Ig chimera protein on brain cryosections, which presented the endogenous HS as the required co-receptor for functional trimeric Fgf10/Fgfr2/HS assembly (Pan et al., 2006). The Fgfr2-Ig chimera protein is unable to bind HS in the absence of Fgf ligand under the experimental conditions used. In situ formation of Fgf10/Fgfr2/HS complex was detected using a secondary antibody that recognizes the Ig domain fused to Fgfr2. Consistent with the timing of the Six3-Cre deleter, we observed that the Fgf-Fgfr binding activity in Ndst1/Ndst2 mutants gradually diminished from E10.5 to E12.5 (Fig. 4A–F). At E13.5, LACE signals were mostly lost in the central retina and the optic stalk, where the Six3-Cre deleter is known be the most active (Fig. 4G–J). Therefore, HS-dependent Fgf-Fgfr assembly into signaling-competent trimeric complexes was impaired in Ndst1/Ndst2 mutants, indicating defective FGF signaling in the affected tissues.

image

Figure 4. Ligand and carbohydrate engagement (LACE) analysis of disrupted HS-dependent Fgf-Fgfr assembly in Ndst mutants. A–H: LACE assay on the frontal sections of Ndst mutants showed a gradual loss of HS-dependent Fgf-Fgfr association in central retinae isolated from Ndst mutant E10 to E13 mice (arrows). I–L: LACE signals were also diminished in Pax2-positive Ndst mutant optic stalks on sagittal sections.

Download figure to PowerPoint

Fgf-MAPK Signaling was Disrupted in Ndst Mutants, and Constitutive Kras Activation Partially Rescued the Ocular Defects

We have previously shown that genetic ablation of Fgfr1/Fgfr2 or Frs2a/Shp2 abrogated MAPK/ERK signaling in the central retina and the optic stalk (Cai et al., 2013). To further determine if Fgf signaling was impaired in Ndst1/Ndst2 mutants, we developed two approaches: one is based on phospho-ERK immunostaining; another is to generate compound mutants with a gain-of-function of Ras signaling. We detected that ERK phosphorylation was down-regulated in both the central retina region and the optic stalk in Ndst1/Ndst2 mutants (Fig. 5A–H). In contrast, when Six3-Cre;Ndst1flox/flox; Ndst2KO/KO mutants were crossed with LSL-KrasG12D, an oncogenic allele of Kras that can be conditionally activated by Cre deletion, the resulting activation of Ras signaling restored ERK phosphorylation in mutant retinae (Fig. 5I; Table 2). This led to a partial recovery of optic disc development, as indicated by Pax2 expression (Fig. 5J; Table 2). However, NF-165 staining showed that many axons of retinal ganglion cells were still misrouted to the subretinal space (Fig. 5J, arrow). As a result, optic nerves in Six3-Cre;Ndst1flox/flox; Ndst2KO/KO;LSL-KrasG12D mutants remained thinner than those of wild-type controls (Fig. 5K,L; Table 1). Therefore, restoration of FGF-Ras signaling did not fully compensate for the loss of HS sulfation during the optic disc and stalk development, suggesting that other factors required for these processes depend on the specifically regulated HS sulfation.

image

Figure 5. MAPK signaling defects in Ndst mutant are partially rescued by Kras. A–H: In Ndst1/Ndst2 mutants, phospho-ERK was down-regulated in central retina, and Pax2 expression was also found to be strongly reduced in the putative optic disc region (E and F, arrows indicate the missing optic discs). As a result, the NF-165-expressing axons projected to the sub-retinal space, failing to reach the optic stalk (F,G), and the optic nerve was missing in adult animal (H). I–L: Partial rescue of developmental defects by constitutive Kras signaling in Ndst mutant mice. Notice that phospho-ERK was recovered to the wild type level (I, arrow) and Pax2-positive cells were detected in the optic disc and stalk regions (J,K, arrows). Partially restored formation of the optic nerve could therefore be observed in Ndst/Kras compound mutants (L).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental Procedures
  7. References

In this study, we have generated a series of mouse mutants with deficient HS sulfation in the optic vesicle and observed resulting hypoplasia or aplasia of the optic disc, stalk and nerve. Mice and cells lacking Ndst1 and Ndst2 isoenzyme function are known to lack HS 2-O and 6-O sulfates in addition to their reduced relative levels of N-sulfation, consistent with the established Ndst gateway function during HS biosynthesis (Holmborn et al., 2004; Grobe et al., 2005; Raman et al., 2011; Sheng et al., 2011). In contrast, HS from cells and mice with targeted deletions in Hs2st and Hs6st function completely lacks 2-O sulfates and is expected to show strongly reduced 6-O sulfation with increased N-sulfation levels (Merry et al., 2001; Habuchi and Kimata, 2010; Qu et al., 2011a). The phenotypic similarities between Ndst1/Ndst2 and Hs2st/Hs6st mutants may thus be best explained by a predominant role of HS O-sulfation in optic disc, stalk, and nerve development. These findings are consistent with impaired Fgf function and retinal axon guidance abnormalities observed at the optic chiasm in Hs2st, Hs6st1, and Hs2st/Hs6st1 compound mutant mice (Pratt et al., 2006; Habuchi and Kimata, 2010). By genetic manipulation of the endogenous Ras signaling activity, we demonstrated that retinal development in Ndst1/Ndst2 mutants could indeed be partially rescued by constitutive MAPK/ERK signaling, supporting our previous study in lens and lacrimal gland development, where activated Ras-ERK signaling also subverted HS sulfation deficiency to reverse FGF signaling defects (Qu et al., 2011b). Together, these results strongly suggest that FGF-Ras-ERK signaling is one of the key downstream pathways regulated by the extracellular heparan sulfate proteoglycans in retinal development, consistent with the established role of HS 6-O sulfation and Ndst activity in specific Fgf functions in other developmental processes (Habuchi and Kimata, 2010; Qu et al., 2011a,b, 2012).

One of our recent studies on Six3-Cre induced FGFR/Frs2/Shp2 conditional mutants provided direct evidence that FGF/MAPK signaling is also required for optic disc and nerve development (Cai et al., 2013). However, whereas both Fgfr1/Fgfr2 and Frs2/Shp2 mutant eyes displayed coloboma as their optic fissures failed to close, we failed to observe similar defects in either Ndst1/Ndst2 or Hs2st/Hs6st mutants presented in this work. This lack of coloboma defects in HS mutants can be explained by residual FGF binding and Fgf/Fgfr formation by undersulfated HS, or by the slow turnover rate of HS in retinal cells, resulting in the persistence of extracellular HS after the Cre-mediated deletion of HS-sulfation genes. Consistent with the latter scenario, we observed that the Fgf/Fgfr binding as indicated by LACE signal in Ndst1/Ndst2 mutation slowly diminished from E10.5 to E12.5 until the optic fissure was already closed, thus precluding any coloboma defects. Indeed, when we deleted Ndst genes used Rx-Cre, another optic vesicle specific Cre line that acts earlier than Six3-Cre (Cai et al., 2010), we observed optic coloboma (Table 1). Therefore, the timing of HS depletion plays an important role in the severity of ocular phenotype.

We have presented evidence that depletion of HS disrupted the assembly of FGF/FGFR and phosphorylation of ERK in developing retina, which can be rescued by constitutively active Kras. It is notable, however, that Kras signaling only partially restored the formation of the optic disc, suggesting that there are additional HSPG-dependent signaling pathways not rescued by Kras signaling. This is in agreement with previous studies showing that HSPGs regulated multiple signaling pathways, including Wnt, Shh, BMP, all of which have been implicated in optic disc development (Kirkpatrick and Selleck, 2007). Moreover, Kras signaling also failed to rescue the misrouting of retinal ganglion cell axons in Ndst1/Ndst2 mutants, consistent with reports that retinal-specific ablation of HS co-polymerase gene Ext1 or systemic knockouts of Hs2st and Hs6st1 genes resulted in defective axon guidance due to impaired Netrin-1 and Slit-1 signaling (Pratt et al., 2006; Ogata-Iwao et al., 2011). Therefore, in line with these previous studies, our work demonstrates that HSPGs also play essential roles in regulating multiple signaling pathways in the optic disc and nerve development.

Experimental Procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental Procedures
  7. References

Mice

Ndst1flox mice have been previously reported (Grobe et al., 2005). Hs6st1flox is a kind gift from Dr. Wellington V. Cardoso (Columbia University, New York, NY) (Izvolsky et al., 2008). Hs2stflox is a kind gift from Dr. Jeffrey D. Esko (University of California San Diego, La Jolla, CA) (Stanford et al., 2010). Ndst2KO is a kind gift from Dr. Lena Kjellén (University of Uppsala, Uppsala, Sweden) (Forsberg et al., 1999). Six3-Cre mice were kindly provided by Dr. Yasuhide Furuta (M.D. Anderson Cancer Center, Houston, TX) (Furuta et al., 2000). Hs6st2KO and LSL-KrasG12D mice were obtained from Mutant Mouse Regional Resource Centers (MMRRC) and the Mouse Models of Human Cancers Consortium (MMHCC) Repository at National Cancer Institute, respectively (Tuveson et al., 2004). All mice were maintained in mixed genetic background. The floxed animals which do not carry Six3-Cre transgene were used as controls. All experiments were performed in accordance with institutional guidelines.

Immunohistochemistry

Immunohistochemistry was performed as previously described with the following antibodies were used: anti-Pax2 (PRB-276P) (from Covance, Berkeley, CA), anti-Sox2 (#Ab5603, Chemicon, Temecula, CA), anti-phospho-ERK1/2 (#9101), and anti-Cyclin D1 (#2926) (from Cell Signaling Technology, Beverly, MA), anti-Ki67 (#550609, BD Pharmingen San Diego, CA) anti-NF165 (#2H3, from Developmental Studies Hybridoma Bank, University of Iowa, Iowa City, IA), anti-Mitf (#MS-771-P0, Thermo Scientific, Fremont, CA) (Pan et al., 2006; Pan et al., 2008). At least three embryos of each genotype were fully sectioned and stained for each marker. The central retina sections across the optic disc region were used in cell counting. The thickness of the adult retina was measured from the ganglion cell layer to the outer nuclear layer based on the DAPI nuclear staining.

RNA In Situ Hybridization

RNA in situ hybridization on cryosections were carried out as previously described (Pan et al., 2006; Cai et al., 2010). The following probes were used: Netrin-1 (from Valerie Wallace, Ottawa Health Research Institute, Ottawa, Ontario, Canada), Math5 (from Dr. Tom Glaser, University of Michigan, Ann Arbor, MI), Brn3b (from Dr. Lin Gan, University of Rochester, Rochester, NY), Chx10, Mitf, and Dct1 (from Dr. Roderick R. McInnes, Hospital for Sick Children, Toronto, Ontario, Canada), Otx1 (from Naoki Takahashi, Nara Institute of Science and Technology, Nara, Japan). At least three independent and fully sectioned eye balls of each genotype were analyzed for each RNA probe.

Ligand and Carbohydrate Engagement (LACE) Assay

The LACE assay was used to probe the in situ binding affinity of Fgf-Fgfr complexes to heparan sulfates on retina sections as previously described (Pan et al., 2006). Recombinant Fgf10 and Fgfr2b were obtained from R&D Systems (Minneapolis, MN).

Acknowledgments

The authors thank Drs. Wellington V. Cardoso, Jeffrey D. Esko, Yasuhide Furuta, Lena Kjellén for mice, Valerie Wallace, Tom Glaser, Lin Gan, Roderick R. McInnes for in situ probes, and members of the Zhang lab for discussions. X.Z. was funded by grants from the NIH and is supported by Jules and Doris Stein Research to Prevent Blindness Professorship.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental Procedures
  7. References
  • Bishop JR, Schuksz M, Esko JD. 2007. Heparan sulphate proteoglycans fine-tune mammalian physiology. Nature 446:10301037.
  • Cai Z, Feng GS, Zhang X. 2010. Temporal requirement of the protein tyrosine phosphatase Shp2 in establishing the neuronal fate in early retinal development. J Neurosci 30:41104119.
  • Cai Z, Tao C, Li H, Ladher R, Gotoh N, Feng GS, Wang F, Zhang X. 2013. Deficient FGF signaling causes optic nerve dysgenesis and ocular coloboma. Development 140:27112723.
  • Chang L, Blain D, Bertuzzi S, Brooks BP. 2006. Uveal coloboma: clinical and basic science update. Curr Opin Ophthalmol 17:447470.
  • Chen S, Li H, Gaudenz K, Paulson A, Guo F, Trimble R, Peak A, Seidel C, Deng C, Furuta Y, Xie T. 2013. Defective FGF signaling causes coloboma formation and disrupts retinal neurogenesis. Cell Res 23:254273.
  • Chiang C, Litingtung Y, Lee E, Young KE, Corden JL, Westphal H, Beachy PA. 1996. Cyclopia and defective axial patterning in mice lacking Sonic hedgehog gene function. Nature 383:407413.
  • Forsberg E, Pejler G, Ringvall M, Lunderius C, Tomasini-Johansson B, Kusche-Gullberg M, Eriksson I, Ledin J, Hellman L, Kjellen L. 1999. Abnormal mast cells in mice deficient in a heparin-synthesizing enzyme. Nature 400:773776.
  • Furuta Y, Hogan BLM. 1998. BMP4 is essential for lens induction in the mouse embryo. Genes Dev 12:37643775.
  • Furuta Y, Lagutin O, Hogan BLM, Oliver GC. 2000. Retina- and ventral forebrain-specific Cre recombinase activity in transgenic mice. Genesis 26:130132.
  • Ghiasvand NM, Rudolph DD, Mashayekhi M, Brzezinski JA, Goldman D, Glaser T. 2011. Deletion of a remote enhancer near ATOH7 disrupts retinal neurogenesis, causing NCRNA disease. Nat Neurosci 14:578586.
  • Grobe K, Inatani M, Pallerla SR, Castagnola J, Yamaguchi Y, Esko JD. 2005. Cerebral hypoplasia and craniofacial defects in mice lacking heparan sulfate Ndst1 gene function. Development 132:37773786.
  • Habuchi H, Kimata K. 2010. Mice deficient in heparan sulfate 6-O-sulfotransferase-1. Prog Mol Biol Transl Sci 93:79111.
  • Holmborn K, Ledin J, Smeds E, Eriksson I, Kusche-Gullberg M, Kjellen L. 2004. Heparan sulfate synthesized by mouse embryonic stem cells deficient in NDST1 and NDST2 is 6-O-sulfated but contains no N-sulfate groups. J Biol Chem 279:4235542358.
  • Izvolsky KI, Lu J, Martin G, Albrecht KH, Cardoso WV. 2008. Systemic inactivation of Hs6st1 in mice is associated with late postnatal mortality without major defects in organogenesis. Genesis 46:818.
  • Kirkpatrick CA, Selleck SB. 2007. Heparan sulfate proteoglycans at a glance. J Cell Sci 120:18291832.
  • Lupo G, Gestri G, O'Brien M, Denton RM, Chandraratna RAS, Ley SV, Harris WA, Wilson SW. 2011. Retinoic acid receptor signaling regulates choroid fissure closure through independent mechanisms in the ventral optic cup and periocular mesenchyme. Proc Natl Acad Sci U S A 108:86988703.
  • Matt N, Ghyselinck NB, Pellerin I, Dupe V. 2008. Impairing retinoic acid signalling in the neural crest cells is sufficient to alter entire eye morphogenesis. Dev Biol 320:140148.
  • Merry CL, Bullock SL, Swan DC, Backen AC, Lyon M, Beddington RS, Wilson VA, Gallagher JT. 2001. The molecular phenotype of heparan sulfate in the Hs2st-/- mutant mouse. J Biol Chem 276:3542935434.
  • Merry CL, Wilson VA. 2002. Role of heparan sulfate-2-O-sulfotransferase in the mouse. Biochim Biophys Acta 1573:319327.
  • Morcillo J, Martínez-Morales JR, Trousse F, Fermin Y, Sowden JC, Bovolenta P. 2006. Proper patterning of the optic fissure requires the sequential activity of BMP7 and SHH. Development 133:31793190.
  • Ogata-Iwao M, Inatani M, Iwao K, Takihara Y, Nakaishi-Fukuchi Y, Irie F, Sato S, Furukawa T, Yamaguchi Y, Tanihara H. 2011. Heparan sulfate regulates intraretinal axon pathfinding by retinal ganglion cells. Invest Ophthalmol Vis Sci 52:66716679.
  • Pan Y, Carbe C, Powers A, Zhang EE, Esko JD, Grobe K, Feng GS, Zhang X. 2008. Bud specific N-sulfation of heparan sulfate regulates Shp2-dependent FGF signaling during lacrimal gland induction. Development 135:301310.
  • Pan Y, Woodbury A, Esko JD, Grobe K, Zhang X. 2006. Heparan sulfate biosynthetic gene Ndst1 is required for FGF signaling in early lens development. Development 133:49334944.
  • Pratt T, Conway CD, Tian NM, Price DJ, Mason JO. 2006. Heparan sulphation patterns generated by specific heparan sulfotransferase enzymes direct distinct aspects of retinal axon guidance at the optic chiasm. J Neurosci 26:69116923.
  • Qu X, Carbe C, Tao C, Powers A, Lawrence R, van Kuppevelt TH, Cardoso WV, Grobe K, Esko JD, Zhang X. 2011a. Lacrimal Gland Development and Fgf10-Fgfr2b Signaling Are Controlled by 2-O- and 6-O-sulfated Heparan Sulfate. J Biol Chem 286:1443514444.
  • Qu X, Hertzler K, Pan Y, Grobe K, Robinson ML, Zhang X. 2011b. Genetic epistasis between heparan sulfate and FGF-Ras signaling controls lens development. Dev Biol 355:1220.
  • Qu X, Pan Y, Carbe C, Powers A, Grobe K, Zhang X. 2012. Glycosaminoglycan-dependent restriction of FGF diffusion is necessary for lacrimal gland development. Development 139:27302739.
  • Raman K, Nguyen TK, Kuberan B. 2011. Is N-sulfation just a gateway modification during heparan sulfate biosynthesis? FEBS Lett 585:34203423.
  • Sheng J, Liu R, Xu Y, Liu J. 2011. The dominating role of N-deacetylase/N-sulfotransferase 1 in forming domain structures in heparan sulfate. J Biol Chem 286:1976819776.
  • Stanford KI, Wang L, Castagnola J, Song D, Bishop JR, Brown JR, Lawrence R, Bai X, Habuchi H, Tanaka M, Cardoso WV, Kimata K, Esko JD. 2010. Heparan sulfate 2-O-sulfotransferase is required for triglyceride-rich lipoprotein clearance. J Biol Chem 285:286294.
  • Tuveson DA, Shaw AT, Willis NA, Silver DP, Jackson EL, Chang S, Mercer KL, Grochow R, Hock H, Crowley D, Hingorani SR, Zaks T, King C, Jacobetz MA, Wang L, Bronson RT, Orkin SH, DePinho RA, Jacks T. 2004. Endogenous oncogenic K-ras(G12D) stimulates proliferation and widespread neoplastic and developmental defects. Cancer Cell 5:375387.
  • Westenskow P, Piccolo S, Fuhrmann S. 2009. β-catenin controls differentiation of the retinal pigment epithelium in the mouse optic cup by regulating Mitf and Otx2 expression. Development 136:25052510.