Mice deficient in Ext2 lack heparan sulfate and develop exostoses

Hereditary multiple exostoses (HME) is a genetically heterogeneous human disease characterized by the development of bony outgrowths near the ends of long bones. HME results from mutations in EXT1 and EXT2, genes that encode glycosyltransferases that synthesize heparan sulfate chains. To study the relationship of the disease to mutations in these genes, we generated Ext2-null mice by gene targeting. Homozygous mutant embryos developed normally until embryonic day 6.0, when they became growth arrested and failed to gastrulate, pointing to the early essential role for heparan sulfate in developing embryos. Heterozygotes had a normal lifespan and were fertile; however, analysis of their skeletons showed that about one-third of the animals formed one or more ectopic bone growths (exostoses). Significantly, all of the mice showed multiple abnormalities in cartilage differentiation, including disorganization of chondrocytes in long bones and premature hypertrophy in costochondral cartilage. These changes were not attributable to a defect in hedgehog signaling, suggesting that they arise from deficiencies in other heparan sulfate-dependent pathways. The finding that haploinsufficiency triggers abnormal cartilage differentiation gives insight into the complex molecular mechanisms underlying the development of exostoses.     

Ext1-dependent heparan sulfate regulates the range of Ihh signaling during endochondral ossification

Exostosin1 (Ext1) belongs to a family of glycosyltransferases necessary for the synthesis of the heparan sulfate (HS) chains of proteoglycans, which regulate signaling of several growth factors. Loss of tout velu (ttv), the homolog of Ext1 in Drosophila, inhibits Hedgehog movement. In contrast, we show that reduced HS synthesis in mice carrying a hypomorphic mutation in Ext1 results in an elevated range of Indian hedgehog (Ihh) signaling during embryonic chondrocyte differentiation. Our data suggest a dual function for HS: First, HS is necessary to bind Hedgehog in the extracellular space. Second, HS negatively regulates the range of Hedgehog signaling in a concentration-dependent manner. Additionally, our data indicate that Ihh acts as a long-range morphogen, directly activating the expression of parathyroid hormone-like hormone. Finally, we propose that the development of exostoses in the human Hereditary Multiple Exostoses syndrome can be attributed to activation of Ihh signaling.     

Proteoheparan sulfate from human skin fibroblasts. Evidence for self-interaction via the heparan sulfate side chains

We have studied the affinity between fibroblast proteoheparan sulfate (medium- and cell surface-derived species) and heparan sulfate-agaroses by affinity chromatography. The evidence for an interaction between the heparan sulfate side chains of the proteoglycans and the immobilized heparan sulfate are as follows: (a) the individual side chains released from the proteoglycan by papain bind to the affinity matrix, (b) the bound proteoglycans are desorbed by a solution of cognate heparan sulfate chains, and (c) the core protein obtained by heparan sulfate-lyase digestion of the proteoglycan does not bind to the affinity matrix. The proteoglycans interact only with one subtype of heparan sulfate. The binding of free heparan sulfate chains to the affinity matrix is completely abolished by heparan sulfate oligosaccharides provided they are composed of both iduronate- and glucuronate-containing disaccharide sequences.     

Selective removal of heparan sulfate chains from proteoheparan sulfate with a commercial preparation of heparitinase

Procedures employing the commercial preparation of heparitinase were developed for isolating a protein-enriched core molecule from proteoheparan sulfate by selective removal of the heparan sulfate chains. Treatment of proteoheparan sulfate with the enzyme preparation caused seriously extensive degradation owing to the presence of proteolytic activity in the enzyme preparation. This effect could be avoided by using a series of protease inhibitors which prevented proteolytic degradation with less significant effect on the heparitinase activity. Application of the procedures to a purified preparation from the Engelbreth-Holm-Swarm tumor yielded a single protein-enriched core fraction with a molecular weight of approximately 450,000, as ascertained by sodium dodceyl sulfate-gel electrophoresis.     

Heterogeneity of heparan sulfate chains in a proteoglycan from bovine lung

A heparan sulfate proteoglycan from bovine lung gas-exchange tissue was isolated by extraction of the tissue with 4.0 M guanidine HCl in the presence of multiple protein inhibitors. The proteoglycan was purified by precipitation with cetylpyridinium chloride in 0.5 M KCl followed by CsCl isopycnic centrifugation (po = 1.45) in 4.0 M guanidine/HCl. Further purification was achieved by gel filtration on Sepharose CL-2B and by chromatography in DEAE-Sepharose CL-6B column. The proteoglycan had 14.9% protein and 22.4% uronate. Heparan sulfate chains from the proteoglycan were isolated after beta-elimination. Fractionation of heparan sulfate chains was achieved on Dowex-1 Cl- column, eluting with a stepwise increase in the concentration of NaCl, 1.0 to 2.0 M with 0.2 M increments. Of the total heparan sulfate recovered from the column, about 10% eluted by 1.2 M NaCl, 68% by 1.4 M NaCl, 18% by 1.6 M NaCl and 4% by 1.8 M NaCl. The fractions varied in their total and N-sulfate ester contents and iduronic acid to glucuronic acid ratios. The fraction that eluted from the Dowex-1 Cl- column at 1.6 M NaCl had the highest molecular weight, 37000, and the fraction that eluted at 1.8 M NaCl had the lowest molecular weight, 12000, as determined by gel filtration method, and the greatest sulfate content. The core protein, obtained by digestion of proteoglycan by heparan sulfate lyase, showed mostly a single band in SDS-polyacrylamide gel electrophoresis. The observations indicate a heterogeneity of the composition of heparan sulfate chains in the proteoglycan. This heterogeneity likely contributes to variations in biologic properties of different heparan sulfate proteoglycan preparations.     

Synovial joint formation requires local Ext1 expression and heparan sulfate production in developing mouse embryo limbs and spine

Heparan sulfate proteoglycans (HSPGs) regulate a number of major developmental processes, but their roles in synovial joint formation remain unknown. Here we created conditional mouse embryo mutants lacking Ext1 in developing joints by mating Ext1^f/f and Gdf5-Cre mice. Ext1 encodes a subunit of the Ext1/Ext2 Golgi-associated protein complex responsible for heparan sulfate (HS) synthesis. The proximal limb joints did form in the Gdf5-Cre;Ext1^f/f mutants, but contained an uneven articulating superficial zone that expressed very low lubricin levels. The underlying cartilaginous epiphysis was deranged as well and displayed random patterns of cell proliferation and matrillin-1 and collagen IIA expression, indicative of an aberrant phenotypic definition of the epiphysis itself. Digit joints were even more affected, lacked a distinct mesenchymal interzone and were often fused likely as a result of local abnormal BMP and hedgehog activity and signaling. Interestingly, overall growth and lengthening of long bones were also delayed in the mutants. To test whether Ext1 function is needed for joint formation at other sites, we examined the spine. Indeed, entire intervertebral discs, normally composed by nucleus pulposus surrounded by the annulus fibrosus, were often missing in Gdf5-Cre;Ext1^f/f mice. When disc remnants were present, they displayed aberrant organization and defective joint marker expression. Similar intervertebral joint defects and fusions occurred in Col2-Cre;β-catenin^f/f mutants. The study provides novel evidence that local Ext1 expression and HS production are needed to maintain the phenotype and function of joint-forming cells and coordinate local signaling by BMP, hedgehog and Wnt/β-catenin pathways. The data indicate also that defects in joint formation reverberate on, and delay, overall long bone growth.     

Multiple distinct membrane heparan sulfate proteoglycans in human lung fibroblasts

Heparitinase digestion of the hydrophobic membrane-associated heparan sulfate proteoglycans (HSPG) of fetal human lung fibroblasts yields core proteins of various sizes: i.e. monomeric core proteins of 125, 90, 64, 48, and 35 kDa and a disulfide-linked dimeric core protein composed of approximately 35-kDa subunits. By immunizing BALB/c mice with liposome-incorporated HSPG, we have obtained a total of five anti-HSPG monoclonal antibodies (Mabs, i.e. Mabs S1, 1C7, 2E9, 6G12, and 10H4) with different specificities. Polyacrylamide gel electrophoresis of 125I-labeled membrane HSPG immunoprecipitated with these Mabs revealed that Mabs 1C7 and 2E9 bind only membrane HSPG which yield a 125-kDa core protein after heparitinase digestion, whereas Mab S1-bound HSPG yield a 64-kDa core protein, and Mabs 6G12 and 10H4 retain membrane HSPG with a 48-kDa core protein. Western blotting of the heparitinase-digested proteoglycans and immunostaining with the Mabs confirmed this pattern of reactivity. However, in this assay, Mabs 6G12 and 10H4 also detected a minor approximately 90-kDa core protein in addition to the 48-kDa core protein. Except perhaps for the 10H4 epitope, the epitopes recognized by these Mabs appear to be part of the peptide moieties as they resisted complete deglycosylation of the HSPG with trifluoromethanesulfonic acid. Since these data were inconsistent with a direct relationship between the major core proteins, the 48-, 64-, and 125-kDa core proteins were immunopurified and further compared by peptide mapping with Staphylococcus aureus protease V8, trypsin, and CNBr cleavage. Clearly distinct peptide patterns were obtained for the three different core proteins. These results imply that the 48-, 64-, and the 125-kDa membrane HSPG core proteins of human lung fibroblasts are derived from distinct proteoglycans.     

Defective nitric oxide-dependent, deaminative cleavage of glypican-1 heparan sulfate in Niemann-Pick C1 fibroblasts

Exit of recycling cholesterol from late endosomes is defective in Niemann-Pick C1 (NPC1) and Niemann-Pick C2 (NPC2) diseases. The traffic route of the recycling proteoglycan glypican-1 (Gpc-1) may also involve late endosomes and could thus be affected in these diseases. During recycling through intracellular compartments, the heparan sulfate (HS) side chains of Gpc-1 are deaminatively degraded by nitric oxide (NO) derived from preformed S-nitroso groups in the core protein. We have now investigated whether this NO-dependent Gpc-1 autoprocessing is active in fibroblasts from NPC1 disease. The results showed that Gpc-1 autoprocessing was defective in these cells and, furthermore, greatly depressed in normal fibroblasts treated with U18666A (3-beta-[2-(diethylamino)ethoxy]androst-5-en-17-one), a compound widely used to induce cholesterol accumulation. In both cases, autoprocessing was partially restored by treatment with ascorbate which induced NO release, resulting in deaminative cleavage of HS. However, when NO-dependent Gpc-1 autoprocessing is depressed and heparanase-catalyzed degradation of HS remains active, a truncated Gpc-1 with shorter HS chains would prevail, resulting in fewer NO-sensitive sites/proteoglycan. Therefore, addition of ascorbate to cells with depressed autoprocessing resulted in nitration of tyrosines. Nitration was diminished when heparanase was inhibited with suramin or when Gpc-1 expression was silenced by RNAi. Gpc-1 misprocessing in NPC1 cells could thus contribute to neurodegeneration mediated by reactive nitrogen species.     

Human follicular fluid heparan sulfate contains abundant 3-O-sulfated chains with anticoagulant activity

Anticoagulant heparan sulfate proteoglycans bind and activate antithrombin by virtue of a specific 3-O-sulfated pentasaccharide. They not only occur in the vascular wall but also in extravascular tissues, such as the ovary, where their functions remain unknown. The rupture of the ovarian follicle at ovulation is one of the most striking examples of tissue remodeling in adult mammals. It involves tightly controlled inflammation, proteolysis, and fibrin deposition. We hypothesized that ovarian heparan sulfates may modulate these processes through interactions with effector proteins. Our previous work has shown that anticoagulant heparan sulfates are synthesized by rodent ovarian granulosa cells, and we now have set out to characterize heparan sulfates from human follicular fluid. Here we report the first anticoagulant heparan sulfate purified from a natural human extravascular source. Heparan sulfate chains were fractionated according to their affinity for antithrombin, and their structure was analyzed by 1H NMR and MS/MS. We find that human follicular fluid is a rich source of anticoagulant heparan sulfate, comprising 50.4% of total heparan sulfate. These antithrombin-binding chains contain more than 6% 3-O-sulfated glucosamine residues, convey an anticoagulant activity of 2.5 IU/ml to human follicular fluid, and have an anti-Factor Xa specific activity of 167 IU/mg. The heparan sulfate chains that do not bind antithrombin surprisingly exhibit an extremely high content in 3-O-sulfated glucosamine residues, which suggest that they may exhibit biological activities through interactions with other proteins.     

Molecular cloning of a phosphatidylinositol-anchored membrane heparan sulfate proteoglycan from human lung fibroblasts

Two mAbs raised against the 64-kD core protein of a membrane heparan sulfate proteoglycan from human lung fibroblasts also recognize a nonhydrophobic proteoglycan which accumulates in the culture medium of the cells. Pulse-chase studies suggest that the hydrophobic cell- associated forms act as precursors for the nonhydrophobic medium- released species. The core proteins of the medium-released proteoglycans are slightly smaller than those of the hydrophobic cell- associated species, but the NH2-terminal amino acid sequences of both forms are identical. The characterization of human lung fibroblast cDNAs that encode the message for these core proteins and the effect of bacterial phosphatidylinositol-specific phospholipase C suggest that the hydrophobic proteoglycan is membrane-anchored through a phospholipid tail. These data identify a novel membrane proteoglycan in human lung fibroblasts and imply that the shedding of this proteoglycan may be related to the presence of the phospholipid anchor.     

Basement-membrane heparan sulfate proteoglycan binds to laminin by its heparan sulfate chains and to nidogen by sites in the protein core.

A large, low-density form of heparan sulfate proteoglycan was isolated from the Engelbreth-Holm-Swarm (EHS) tumor and demonstrated to bind in immobilized-ligand assays to laminin fragment E3, collagen type IV, fibronectin and nidogen. The first three ligands mainly recognize the heparan sulfate chains, as shown by inhibition with heparin and heparan sulfate and by the failure to bind to the proteoglycan protein core. Nidogen, obtained from the EHS tumor or in recombinant form, binds exclusively to the protein core in a heparin-insensitive manner. Studies with other laminin fragments indicate that the fragment E3 possesses a unique binding site of laminin for the proteoglycan. A major binding site of nidogen was localized to its central globular domain G2 by using overlapping fragments. This allows for the formation of ternary complexes between laminin, nidogen and proteoglycan, suggesting a key role for nidogen in basement-membrane assembly. Evidence is provided for a second proteoglycan-binding site in the C-terminal globule G3 of nidogen, but this interaction prevents the formation of such ternary complexes. Therefore, the G3-mediated nidogen binding to laminin and proteoglycan are mutually exclusive.     

A conformational change in heparan sulfate 3-O-sulfotransferase-1 is induced by binding to heparan sulfate

The 3-O-sulfation of glucosamine by heparan sulfate 3-O-sulfotransferase-1 (3-OST-1) is a key modification step during the biosynthesis of anticoagulant heparan sulfate (HS). In this paper, we present evidence of a conformational change that occurs in 3-OST-1 upon binding to heparan sulfate. The intrinsic fluorescence of 3-OST-1 was increased in the presence of HS, suggesting a conformational change. This apparent conformational change was further investigated using differential chemical modification of 3-OST-1 to measure the solvent accessibility of the lysine residues. 3-OST-1 was treated with acetic anhydride in either the presence or absence of HS using both acetic anhydride and hexadeuterioacetic anhydride under nondenaturing and denaturing conditions, respectively. The relative reactivity of the lysine residues to acetylation and [2H] acetylation in the presence or absence of HS was analyzed by measuring the ratio of acetylated and deuterioacetylated peptides using matrix-assisted laser desorption ionization mass spectrometry. The solvent accessibilities of the lysine residues were altered differentially depending on their location. In particular, we observed a group of lysine residues in the C-terminus of 3-OST-1 that become more solvent accessible when 3-OST-1 binds to HS. This observation indicates that a conformational change could be occurring during substrate binding. A truncated mutant of 3-OST-1 that lacked this C-terminal region was expressed and found to exhibit a 200-fold reduction in sulfotransferase activity. The results from this study will contribute to our understanding of the interactions between 3-OSTs and HS.     

Expression of human perlecan domain I as a recombinant heparan sulfate proteoglycan with 20-kDa glycosaminoglycan chains

Recombinant forms of human perlecan domain I were secreted as proteoglycans by stably transfected human 293 cells. A recombinant domain I-only proteoglycan spanned the 95- to 265-kDa region in SDS-PAGE and appeared to be 160 kDa in denaturing gel filtration. Its glycosaminoglycan (GAG) content was approximately 67% heparan sulfate, and its average GAG chain size of 20 kDa suggested that the true molecular mass of the proteoglycan was 90 kDa. Domain I with enhanced green fluorescent protein fused to its C-terminus had an apparent molecular mass of 210-220 kDa and contained approximately 100% heparan sulfate. Its average GAG chain size (also 20 kDa) suggested a true molecular mass of 117 kDa for this proteoglycan. Its sulfate content of 53-77 mol SO2-4 per mole of protein indicated the presence of one sulfate group per 4-7 GAG sugar residues.     

Undersulfated heparan sulfate in a Chinese hamster ovary cell mutant defective in heparan sulfate N-sulfotransferase

Heparan sulfate N-sulfotransferase catalyzes the transfer of sulfate groups from adenosine 3'-phosphate, 5'-phosphosulfate to the free amino groups of glucosamine residues in heparan sulfate. We have identified a Chinese hamster ovary cell mutant, designated pgsE-606, which is 3-5-fold defective in N-sulfotransferase activity. The residual enzyme activity is indistinguishable from the wild-type enzyme with respect to Km values for adenosine 3'-phosphate,5'-phosphosulfate and N-desulfoheparin, pH dependence, Arrhenius activation energy, and thermal lability. The mutation is recessive, and mixing experiments indicate that the mutant does not produce soluble antagonists of N-sulfotransferase. Inspection of the heparan sulfate chains from the mutant showed that the extent of N-sulfation is reduced about 2-3-fold. The addition of sulfate to hydroxyl groups on the chain is reduced to a similar extent, suggesting that N-sulfation and O-sulfation are normally coupled. Nitrous acid fragmentation of the chains showed that N-sulfated glucosamine residues are spaced much less frequently than in heparan sulfate from wild-type cells. The close correlation of enzyme activity to the number and position of N-sulfate groups indicates that N-sulfotransferase plays a pivotal role in determining the extent of sulfation of heparan sulfate.     

Altered heparan sulfate structure in mice with deleted NDST3 gene function

We report the generation and analysis of mutant mice bearing a targeted disruption of the heparan sulfate (HS)-modifying enzyme GlcNAc N-deacetylase/N-sulfotransferase 3 (NDST3). NDST3(-/-) mice develop normally, are fertile, and show only subtle hematological and behavioral abnormalities in agreement with only moderate HS undersulfation. Compound mutant mice made deficient in NDST2;NDST3 activities also develop normally, showing that both isoforms are not essential for development. In contrast, NDST1(-/-);NDST3(-/-) compound mutant embryos display developmental defects caused by severe HS undersulfation, demonstrating NDST3 contribution to HS synthesis in the absence of NDST1. Moreover, analysis of HS composition in dissected NDST3 mutant adult brain revealed regional changes in HS sulfation, indicating restricted NDST3 activity on nascent HS in defined wild-type tissues. Taken together, we show that NDST3 function is not essential for development or adult homeostasis despite contributing to HS synthesis in a region-specific manner and that the loss of NDST3 function is compensated for by the other NDST isoforms to a varying degree.     

Cerebral hypoplasia and craniofacial defects in mice lacking heparan sulfate Ndst1 gene function

Mutant mice bearing a targeted disruption of the heparan sulfate (HS) modifying enzyme GlcNAc N-deacetylase/N-sulfotransferase 1 (Ndst1) exhibit severe developmental defects of the forebrain and forebrain-derived structures, including cerebral hypoplasia, lack of olfactory bulbs, eye defects and axon guidance errors. Neural crest-derived facial structures are also severely affected. We show that properly synthesized heparan sulfate is required for the normal development of the brain and face, and that Ndst1 is a modifier of heparan sulfate-dependent growth factor/morphogen signalling in those tissues. Among the multiple heparan sulfate-binding factors potentially affected in Ndst1 mutant embryos, the facial phenotypes are consistent with impaired sonic hedgehog (Shh) and fibroblast growth factor (Fgf) interaction with mutant heparan sulfate. Most importantly, the data suggest the possibility that defects in heparan sulfate synthesis could give rise to or contribute to a number of developmental brain and facial defects in humans.     

Multiple forms of heparan sulfate proteoglycans in the Engelbreth-Holm-Swarm mouse tumor. The occurrence of high density forms bearing both heparan sulfate and chondroitin sulfate side chains

Heparan sulfate proteoglycan from the Engelbreth-Holm-Swarm mouse tumor was previously separated into two forms: a high density form (Form HD) and low density form (Form LD). In this study, the two forms were radiolabeled either metabolically with [35S]sulfate or [3H]serine or chemically with 125I. Pulse-chase experiments with [35S]sulfate showed no clear precursor-product relationship between the two forms. Analyses of the labeled proteoglycan samples with heparitinase and chondroitinase ABC indicated that Form LD is a large proteoglycan containing heparan sulfate chains attached to a single core molecule (Mr = 450,000), whereas Form HD is a mixture of small proteoglycans with four different size core molecules (Mr = 34,000, 29,000, 27,000, and 21,000), most, if not all, of which bear both heparan sulfate (Mr = 60,000) and chondroitin sulfate (Mr = 17,000) chains. Glycosaminoglycan-enriched fragments obtained from Form HD by V8 protease digestion were also shown to contain both heparitinase-susceptible chains and chondroitinase ABC-susceptible chains. Tryptic peptide maps of 125I-labeled Form HD and the glycosaminoglycan-enriched fragments derived therefrom were quite different from the corresponding maps for Form LD.     

Heparan sulfate structure in mice with genetically modified heparan sulfate production

Using a high throughput heparan sulfate (HS) isolation and characterization protocol, we have analyzed HS structure in several tissues from mice/mouse embryos deficient in HS biosynthesis enzymes (N-deacetylase/N-sulfotransferase (NDST)-1, NDST-2, and C5-epimerase, respectively) and in mice lacking syndecan-1. The results have given us new information regarding HS biosynthesis with implications on the role of HS in embryonic development. Our main conclusions are as follows. 1) The HS content, disaccharide composition, and the overall degree of N- and O-sulfation as well as domain organization are characteristic for each individual mouse tissue. 2) Removal of a key biosynthesis enzyme (NDST-1 or C5-epimerase) results in similar structural alterations in all of the tissues analyzed. 3) Essentially no variation in HS tissue structure is detected when individuals of the same genotype are compared. 4) NDST-2, although generally expressed, does not contribute significantly to tissue-specific HS structures. 5) No change in HS structure could be detected in syndecan-1-deficient mice.     

Cell surface heparan sulfate chains regulate local reception of FGF signaling in the mouse embryo

Heparan sulfate (HS) proteoglycans modulate the activity of multiple growth factors on the cell surface and extracellular matrix. However, it remains unclear how the HS chains control the movement and reception of growth factors into targeted receiving cells during mammalian morphogenetic processes. Here, we found that HS-deficient Ext2 null mutant mouse embryos fail to respond to fibroblast growth factor (FGF) signaling. Marker expression analyses revealed that cell surface-tethered HS chains are crucial for local retention of FGF4 and FGF8 ligands in the extraembryonic ectoderm. Fine chimeric studies with single-cell resolution and expression studies with specific inhibitors for HS movement demonstrated that proteolytic cleavage of HS chains can spread FGF signaling to adjacent cells within a short distance. Together, the results show that spatiotemporal expression of cell surface-tethered HS chains regulate the local reception of FGF-signaling activity during mammalian embryogenesis.