Molecular cloning and expression of a second chondroitin N-acetylgalactosaminyltransferase involved in the initiation and elongation of chondroitin/dermatan sulfate

We identified a novel human chondroitin N-acetylgalactosaminyltransferase, designated chondroitin GalNAcT-2 after a BLAST analysis of the GenBank(TM) data base using the sequence of a previously described human chondroitin N-acetylgalactosaminyltransferase (chondroitin GalNAcT-1) as a probe. The new cDNA sequence contained an open reading frame encoding a protein of 542 amino acids with a type II transmembrane protein topology. The amino acid sequence displayed 60% identity to that of human chondroitin GalNAcT-1. Like chondroitin GalNAcT-1, the expression of a soluble form of the protein in COS-1 cells produced an active enzyme, which not only transferred beta1,4-N-acetylgalactosamine (GalNAc) from UDP-[(3)H]GalNAc to a polymer chondroitin representing growing chondroitin chains (beta-GalNAc transferase II activity) but also to GlcUA beta 1-3Gal beta 1-O-C(2)H(4)NHCbz, a synthetic substrate for beta-GalNAc transferase I that transfers the first GalNAc to the core tetrasaccharide in the protein-linkage region of chondroitin sulfate. In contrast, the tetrasaccharide serine (GlcUA beta 1-3Gal beta 1-3Gal beta 1-4Xyl beta 1-O-Ser) derived from the linkage region, which is an inert acceptor substrate for chondroitin GalNAcT-1, served as an acceptor substrate. The coding region of this enzyme was divided into seven discrete exons, which is similar to the genomic organization of the chondroitin GalNAcT-1 gene, and was localized to chromosome 10q11.22. Northern blot analysis revealed that the chondroitin GalNAcT-2 gene exhibited a ubiquitous but differing expression in human tissues, and the expression pattern differed from that of chondroitin GalNAcT-1. Thus, we demonstrated redundancy in the chondroitin GalNAc transferases involved in the biosynthetic initiation and elongation of chondroitin sulfate, which is important for understanding the biosynthetic mechanisms leading to the selective chain assembly of chondroitin/dermatan sulfate on the linkage region tetrasaccharide common to various proteoglycans containing chondroitin/dermatan sulfate and heparin/heparan sulfate chains.     

The tumor suppressor EXT-like gene EXTL2 encodes an alpha1, 4-N-acetylhexosaminyltransferase that transfers N-acetylgalactosamine and N-acetylglucosamine to the common glycosaminoglycan-protein linkage region. The key enzyme for the chain initiation of heparan sulfate.

We previously demonstrated a unique alpha-N-acetylgalactosaminyltransferase that transferred N-acetylgalactosamine (GalNAc) to the tetrasaccharide-serine, GlcAbeta1-3Galbeta1-3Galbeta1-4Xylbeta1-O-Ser (GlcA represents glucuronic acid), derived from the common glycosaminoglycan-protein linkage region, through an alpha1,4-linkage. In this study, we purified the enzyme from the serum-free culture medium of a human sarcoma cell line. Peptide sequence analysis of the purified enzyme revealed 100% identity to the multiple exostoses-like gene EXTL2/EXTR2, a member of the hereditary multiple exostoses (EXT) gene family of tumor suppressors. The expression of a soluble recombinant form of the protein produced an active enzyme, which transferred alpha-GalNAc from UDP-[3H]GalNAc to various acceptor substrates including GlcAbeta1-3Galbeta1-3Galbeta1-4Xylbeta1-O-Ser. Interestingly, the enzyme also catalyzed the transfer of N-acetylglucosamine (GlcNAc) from UDP-[3H]GlcNAc to GlcAbeta1-3Galbeta1-O-naphthalenemethanol, which was the acceptor substrate for the previously described GlcNAc transferase I involved in the biosynthetic initiation of heparan sulfate. The GlcNAc transferase reaction product was sensitive to the action of heparitinase I, establishing the identity of the enzyme to be alpha1, 4-GlcNAc transferase. These results altogether indicate that EXTL2/EXTR2 encodes the alpha1,4-N-acetylhexosaminyltransferase that transfers GalNAc/GlcNAc to the tetrasaccharide representing the common glycosaminoglycan-protein linkage region and that is most likely the critical enzyme that determines and initiates the heparin/heparan sulfate synthesis, separating it from the chondroitin sulfate/dermatan sulfate synthesis.     

Molecular cloning and expression of a human chondroitin synthase

We have identified a human chondroitin synthase from the HUGE (human unidentified gene-encoded large proteins) protein data base by screening with two keywords: "one transmembrane domain" and "galactosyltransferase family." The identified protein consists of 802 amino acids with a type II transmembrane protein topology. The protein showed weak homology to the beta1,3-galactosyltransferase family on the amino-terminal side and to the beta1,4-galactosyltransferase family on the carboxyl-terminal side. The expression of a soluble recombinant form of the protein in COS-1 cells produced an active enzyme, which transferred not only the glucuronic acid (GlcUA) from UDP-[(14)C]GlcUA but also N-acetylgalactosamine (GalNAc) from UDP-[(3)H]GalNAc to the polymer chondroitin. Identification of the reaction products demonstrated that the enzyme was chondroitin synthase, with both beta1,3-GlcUA transferase and beta1,4-GalNAc transferase activities. The coding region of the chondroitin synthase was divided into three discrete exons and localized to chromosome 15. Northern blot analysis revealed that the chondroitin synthase gene exhibited ubiquitous but markedly differential expression in the human tissues examined. Thus, we demonstrated that analogous to human heparan sulfate polymerases, the single polypeptide chondroitin synthase possesses two glycosyltransferase activities required for chain polymerization.     

Chondroitin sulfate N-acetylgalactosaminyltransferase-1 (CSGalNAcT-1) involved in chondroitin sulfate initiation: Impact of sulfation on activity and specificity

Glycosaminoglycan (GAG) assembly initiates through the formation of a linkage tetrasaccharide region serving as a primer for both chondroitin sulfate (CS) and heparan sulfate (HS) chain polymerization. A possible role for sulfation of the linkage structure and of the constitutive disaccharide unit of CS chains in the regulation of CS-GAG chain synthesis has been suggested. To investigate this, we determined whether sulfate substitution of galactose (Gal) residues of the linkage region or of N-acetylgalactosamine (GalNAc) of the disaccharide unit influences activity and specificity of chondroitin sulfate N-acetylgalactosaminyltransferase-1 (CSGalNAcT-1), a key glycosyltransferase of CS biosynthesis. We synthesized a series of sulfated and unsulfated analogs of the linkage oligosaccharide and of the constitutive unit of CS and tested these molecules as potential acceptor substrates for the recombinant human CSGalNAcT-1. We show here that sulfation at C4 or C6 of the Gal residues markedly influences CSGalNAcT-1 initiation activity and catalytic efficiency. Kinetic analysis indicates that CSGalNAcT-1 exhibited 3.6-, 1.6-, and 2.2-fold higher enzymatic efficiency due to lower K(m) values toward monosulfated trisaccharides substituted at C4 or C6 position of Gal1, and at C6 of Gal2, respectively, compared with the unsulfated oligosaccharide. This highlights the critical influence of Gal substitution on both CSGalNAcT-1 activity and specifity. No GalNAcT activity was detected toward sulfated and unsulfated analogs of the CS constitutive disaccharide (GlcA-β1,3-GalNAc), indicating that CSGalNAcT-1 was involved in initiation but not in elongation of CS chains. Our results strongly suggest that sulfation of the linkage region acts as a regulatory signal in CS chain initiation.     

Chondroitin 4-O-sulfotransferase-2 regulates the number of chondroitin sulfate chains initiated by chondroitin N-acetylgalactosaminyltransferase-1

Recently, it has been shown that a deficiency in ChGn-1 (chondroitin N-acetylgalactosaminyltransferase-1) reduced the numbers of CS (chondroitin sulfate) chains, leading to skeletal dysplasias in mice. Although these results indicate that ChGn-1 regulates the number of CS chains, the mechanism mediating this regulation is not clear. ChGn-1 is thought to initiate CS biosynthesis by transferring the first GalNAc (N-acetylgalactosamine) to the tetrasaccharide in the protein linkage region of CS. However, in vitro chondroitin polymerization does not occur on the non-reducing terminal GalNAc-linkage pentasaccharide structure. In the present study we show that several different heteromeric enzyme complexes composed of different combinations of four chondroitin synthase family members synthesized more CS chains when a GalNAc-linkage pentasaccharide structure with a non-reducing terminal 4-O-sulfation was the CS acceptor. In addition, C4ST-2 (chondroitin 4-O-sulfotransferase-2) efficiently transferred sulfate from 3'-phosphoadenosine 5'-phosphosulfate to position 4 of non-reducing terminal GalNAc-linkage residues, and the number of CS chains was regulated by the expression levels of C4ST-2 and of ChGn-1. Taken together, the results of the present study indicate that C4ST-2 plays a key role in regulating levels of CS synthesized via ChGn-1.     

Biosynthesis of chondroitin/dermatan sulfate

Chondroitin sulfate and dermatan sulfate are synthesized as galactosaminoglycan polymers containing N-acetylgalactosmine alternating with glucuronic acid. The sugar residues are sulfated to varying degrees and positions depending upon the tissue sources and varying conditions of formation. Epimerization of any of the glucuronic acid residues to iduronic acid at the polymer level constitutes the formation of dermatan sulfate. Chondroitin/dermatan glycosaminoglycans are covalently attached by a common tetrasaccharide sequence to the serine residues of core proteins while they are adherent to the inner surface of endoplasmic reticulum/Golgi vesicles. Addition of the first sugar residue, xylose, to core proteins begins in the endoplasmic reticulum, followed by the addition of two galactose residues by two distinct glycosyl transferases in the early cis/medial regions of the Golgi. The linkage tetrasaccharide is completed in the medial/trans Golgi by the addition of the first glucuronic acid residue, followed by transfer of N-acetylgalactosamine to initiate the formation of a galactosaminoglycan rather than a glucosaminoglycan. This specific N-acetylgalactosaminyl transferase is different from the chondroitin synthase involved in generation of the repeating disaccharide units to form the chondroitin polymer. Sulfation of the chondroitin polymer by specific sulfotransferases occurs as the polymer is being formed. All the enzymes in the pathway for synthesis have been cloned, with the exception of the glucuronyl to iduronyl epimerase involved in the formation of dermatan residues.     

Molecular cloning and characterization of chondroitin polymerase from Escherichia coli strain K4

Escherichia coli strain K4 produces the K4 antigen, a capsule polysaccharide consisting of a chondroitin backbone (GlcUA beta(1-3)-GalNAc beta(1-4))(n) to which beta-fructose is linked at position C-3 of the GlcUA residue. We molecularly cloned region 2 of the K4 capsular gene cluster essential for biosynthesis of the polysaccharide, and we further identified a gene encoding a bifunctional glycosyltransferase that polymerizes the chondroitin backbone. The enzyme, containing two conserved glycosyltransferase sites, showed 59 and 61% identity at the amino acid level to class 2 hyaluronan synthase and chondroitin synthase from Pasteurella multocida, respectively. The soluble enzyme expressed in a bacterial expression system transferred GalNAc and GlcUA residues alternately, and polymerized the chondroitin chain up to a molecular mass of 20 kDa when chondroitin sulfate hexasaccharide was used as an acceptor. The enzyme exhibited apparent K(m) values for UDP-GlcUA and UDP-GalNAc of 3.44 and 31.6 microm, respectively, and absolutely required acceptors of chondroitin sulfate polymers and oligosaccharides at least longer than a tetrasaccharide. In addition, chondroitin polymers and oligosaccharides and hyaluronan polymers and oligosaccharides served as acceptors for chondroitin polymerization, but dermatan sulfate and heparin did not. These results may lead to elucidation of the mechanism for chondroitin chain synthesis in both microorganisms and mammals.     

A fingerprinting method for chondroitin/dermatan sulfate and hyaluronan oligosaccharides

A previously published method for the analysis of glycosaminoglycan disaccharides by high pH anion exchange chromatography (Midura,R.J., Salustri,A., Calabro,A., Yanagishita,M. and Hascall,V.C. (1994), Glycobiology,4, 333-342) has been modified and calibrated for chondroitin and dermatan sulfate oligosaccharides up to hexasaccharide in size and hyaluronan oligosaccharides up to hexadecasaccharide. For hyaluronan oligosaccharides chain length controls elution position; however, for chondroitin and dermatan sulfate oligosaccharides elution times primarily depend upon the level of sulfation, although chain length and hence charge density plays a role. The sulfation position of GalNAc residues within an oligosaccharide is also important in determining its elution position. Compared to 4-sulfation a reducing terminal 6-sulfate retards elution; however, when present on an internal GalNAc residue it is the 4-sulfate containing oligosaccharide which elutes later. These effects allow discrimination between oligosaccharides differing only in the position of GalNAc sulfation. Using this simple methodology, a Dionex CarboPac PA-1 column with NaOH/NaCl eluents and detection by absorbance at 232 nm, a quantitative analytical fingerprint of a chondroitin/dermatan sulfate chain may be obtained, allowing a determination of the abundance of chondroitin sulfate, dermatan sulfate, and hyaluronan along with an analysis of structural features with a linear response to approximately 0.1 nmol. The method may readily be calibrated using either commercial disaccharides or the di- and tetrasaccharide products of a limit digest of commercial chondroitin sulfate by chondroitin ABC endolyase. Commercially available and freshly prepared shark, whale, bovine, and human cartilage chondroitin sulfates have been examined by this methodology and we have confirmed that freshly isolated shark cartilage CS contains significant amounts of the biologically important GlcA2Sbeta(1-3)GalNAc6S structure.     

A chondroitin synthase-1 (ChSy-1) missense mutation in a patient with neuropathy impairs the elongation of chondroitin sulfate chains initiated by chondroitin N-acetylgalactosaminyltransferase-1

Background

Previously, we identified two missense mutations in the chondroitin N-acetylgalactosaminyltransferase-1 gene in patients with neuropathy. These mutations are associated with a profound decrease in chondroitin N-acetylgalactosaminyltransferase-1 enzyme activity. Here, we describe a patient with neuropathy who is heterozygous for a chondroitin synthase-1 mutation. Chondroitin synthase-1 has two glycosyltransferase activities: it acts as a GlcUA and a GalNAc transferase and is responsible for adding repeated disaccharide units to growing chondroitin sulfate chains.

Methods

Recombinant wild-type chondroitin synthase-1 enzyme and the F362S mutant were expressed. These enzymes and cells expressing them were then characterized.

Results

The mutant chondroitin synthase-1 protein retained approximately 50% of each glycosyltransferase activity relative to the wild-type chondroitin synthase-1 protein. Furthermore, unlike chondroitin polymerase comprised of wild-type chondroitin synthase-1 protein, the non-reducing terminal 4-O-sulfation of GalNAc residues synthesized by chondroitin N-acetylgalactosaminyltransferase-1 did not facilitate the elongation of chondroitin sulfate chains when chondroitin polymerase that consists of the mutant chondroitin synthase-1 protein was used as the enzyme source.

Conclusions

The chondroitin synthase-1 F362S mutation in a patient with neuropathy resulted in a decrease in chondroitin polymerization activity and the mutant protein was defective in regulating the number of chondroitin sulfate chains via chondroitin N-acetylgalactosaminyltransferase-1. Thus, the progression of peripheral neuropathies may result from defects in these regulatory systems.

General significance

The elongation of chondroitin sulfate chains may be tightly regulated by the cooperative expression of chondroitin synthase-1 and chondroitin N-acetylgalactosaminyltransferase-1 in peripheral neurons and peripheral neuropathies may result from synthesis of abnormally truncated chondroitin sulfate chains.     

Purification and characterization of N-acetylgalactosamine 4-sulfate 6-O-sulfotransferase from the squid cartilage

N-Acetylgalactosamine 4-sulfate 6-O-sulfotransferase (GalNAc4S-6ST), which transfers sulfate from 3'-phosphoadenosine 5'-phosphosulfate (PAPS) to position 6 of N-acetylgalactosamine 4-sulfate in chondroitin sulfate and dermatan sulfate, was purified 19,600-fold to apparent homogeneity from the squid cartilage. SDS-polyacrylamide gel electrophoresis of the purified enzyme showed a broad protein band with a molecular mass of 63 kDa. The protein band coeluted with GalNAc4S-6ST activity from Toyopearl HW-55 around the position of 66 kDa, indicating that the active form of GalNAc4S-6ST may be a monomer. The purified enzyme transferred sulfate from PAPS to chondroitin sulfate A, chondroitin sulfate C, and dermatan sulfate. The transfer of sulfate to chondroitin sulfate A and dermatan sulfate occurred mainly at position 6 of the internal N-acetylgalactosamine 4-sulfate residues. Chondroitin sulfate E, keratan sulfate, heparan sulfate, and completely desulfated N-resulfated heparin were not efficient acceptors of the sulfotransferase. When a trisaccharide or a pentasaccharide having sulfate groups at position 4 of N-acetylgalactosamine was used as acceptor, efficient sulfation of position 6 at the nonreducing terminal N-acetylgalactosamine 4-sulfate residue was observed.     

Identification of chondroitin sulfate glucuronyltransferase as chondroitin synthase-3 involved in chondroitin polymerization: chondroitin polymerization is achieved by multiple enzyme complexes consisting of chondroitin synthase family members

Recently, we demonstrated that chondroitin polymerization is achieved by any two combinations of human chondroitin synthase-1 (ChSy-1), ChSy-2 (chondroitin sulfate synthase 3, CSS3), and chondroitin-polymerizing factor (ChPF). Although an additional ChSy family member, called chondroitin sulfate glucuronyltransferase (CSGlcA-T), has been identified, its involvement in chondroitin polymerization remains unclear because it possesses only glucuronyltransferase II activity responsible for the elongation of chondroitin sulfate (CS) chains. Herein, we report that CSGlcA-T exhibits polymerization activity on alpha-thrombomodulin bearing the truncated linkage region tetrasaccharide through its interaction with ChSy-1, ChSy-2 (CSS3), or ChPF, and the chain length of chondroitin formed by the co-expressed proteins in various combinations is different. In addition, ChSy family members co-expressed in various combinations exhibited distinct but overlapping acceptor substrate specificities toward the two synthetic acceptor substrates, GlcUAbeta1-3Galbeta1-O-naphthalenemethanol and GlcUAbeta1-3Galbeta1-O-C(2)H(4)NH-benzyloxycarbonyl, both of which share the disaccharide sequence with the glycosaminoglycan-protein linkage region tetrasaccharide. Moreover, overexpression of CSGlcA-T increased the amount of CS in HeLa cells, whereas the RNA interference of CSGlcA-T resulted in a reduction of the amount of CS in the cells. Furthermore, the analysis using the CSGlcA-T mutant that lacks any glycosyltransferase activity but interacts with other ChSy family members showed that the glycosyltransferase activity of CSGlcA-T plays an important role in chondroitin polymerization. Overall, these results suggest that chondroitin polymerization is achieved by multiple combinations of ChSy-1, ChSy-2, CSGlcA-T, and ChPF and that each combination may play a unique role in the biosynthesis of CS. Based on these results, we renamed CSGlcA-T chondroitin synthase-3 (ChSy-3).     

Molecular cloning and characterization of a human uronyl 2-sulfotransferase that sulfates iduronyl and glucuronyl residues in dermatan/chondroitin sulfate

A partial-length human cDNA with a predicted amino acid sequence homologous to a previously described heparan sulfate iduronyl 2-sulfotransferase (Kobayashi, M., Habuchi, H., Yoneda, M., Habuchi, O., and Kimata, K. (1997) J. Biol. Chem. 272, 13980-13985) was obtained by searching the expressed sequence-tagged data bank. Northern blot analysis was performed using this homologous cDNA as a probe, which demonstrated ubiquitous expression of messages of 5.1 and 2.0 kilobases in a number of human tissues and in several human cancer cell lines. Since the human lymphoma Raji cell line had the highest level of expression, it was used to isolate a full-length cDNA clone. The full-length cDNA was found to contain an open reading frame that predicted a type II transmembrane protein composed of 406 amino acid residues. The cDNA in a baculovirus expression vector was expressed in Sf9 insect cells, and cell extracts were then incubated together with 3'-phosphoadenosine 5'-phospho[35S]sulfate and potential glycosaminoglycan acceptors. This demonstrated substantial sulfotransferase activity with dermatan sulfate, a small degree of activity with chondroitin sulfate, but no sulfotransferase activity with desulfated N-resulfated heparin. Analysis of [35S]sulfate-labeled disaccharide products of chondroitin ABC, chondroitin AC, and chondroitin B lyase treatment demonstrated that the enzyme only transferred sulfate to the 2-position of uronyl residues, which were preponderantly iduronyl residues in dermatan sulfate, but some lesser transfer to glucuronyl residues of chondroitin sulfate.     

Purification and characterization of fetal bovine serum beta-N-acetyl-D-galactosaminyltransferase and beta-D-glucuronyltransferase involved in chondroitin sulfate biosynthesis.

beta-N-Acetylgalactosaminyltransferase II and beta-glucuronyltransferase II, involved in chondroitin sulfate biosynthesis, transfer an N-acetylgalactosamine (GalNAc) and glucuronic acid (GlcA) residue, respectively, through beta-linkages to an acceptor chondroitin oligosaccharide derived from the repeating disaccharide region of chondroitin sulfate. They were copurified from fetal bovine serum approximately 2500-fold and 850-fold, respectively, by sequential chromatographies on Red A-agarose, phenyl-Sepharose, S-Sepharose and wheat germ agglutinin-agarose. Identical and inseparable chromatographic profiles of both glycosyltransferase activities obtained through the above chromatographic steps and gel filtration suggest that the purified enzyme activities are tightly coupled, which could imply a single enzyme with dual transferase activities; beta-N-acetylgalactosaminyltransferase and beta-glucuronyltransferase, reminiscent of the heparan sulfate polymerase reaction. However, when a polymerization reaction was performed in vitro with the purified serum enzyme preparation under the polymerization conditions recently developed for the chondroitin-synthesizing system, derived from human melanoma cells, each monosaccharide transfer took place, but no polymerization occurred. These results may suggest that the purified serum enzyme preparation contains both beta-N-acetylgalactosaminyltransferase II and beta-glucuronyltransferase II activities on a single polypeptide or on the respective polypeptides forming an enzyme complex, but is different from that obtained from melanoma cells in that it transfers a single GalNAc or GlcA residue but does not polymerize chondroitin.     

Molecular cloning of a chondroitin polymerizing factor that cooperates with chondroitin synthase for chondroitin polymerization

We recently cloned human chondroitin synthase (ChSy) exhibiting the glucuronyltransferase-II (GlcATII) and N-acetylgalactosaminyltransferase-II (GalNAcTII) activities responsible for the biosynthesis of repeating disaccharide units of chondroitin sulfate, but chondroitin polymerization was not demonstrated in vitro using the recombinant ChSy. We report here that the chondroitin polymerizing activity requires concomitant expression of a novel protein designated chondroitin polymerizing factor (ChPF) with ChSy. The human ChPF consists of 775 amino acids with a type II transmembrane protein topology. The amino acid sequence displayed 23% identity to that of human ChSy. The expression of a soluble recombinant form of the protein in COS-1 cells produced a protein with little GlcAT-II or GalNAcT-II activity. In contrast, coexpression of the ChPF and ChSy yielded markedly augmented glycosyltransferase activities, whereas simple mixing of the two separately expressed proteins did not. Moreover, using both UDP-glucuronic acid (GlcUA) and UDP-N-acetylgalactosamine (GalNAc) as sugar donors, chondroitin polymerization was demonstrated on the so-called glycosaminoglycan-protein linkage region tetrasaccharide sequence of alpha-thrombomodulin. These results suggested that the ChPF acts as a specific activating factor for ChSy in chondroitin polymerization. The coding region of the ChPF was divided into four discrete exons and localized to chromosome 2q35-q36. Northern blot analysis revealed that the ChPF gene exhibited a markedly different expression pattern among various human tissues, which was similar to that of ChSy. Thus, the ChPF is required for chondroitin polymerizing activity of mammalian ChSy.     

The phosphorylated and/or sulfated structure of the carbohydrate-protein-linkage region isolated from chondroitin sulfate in the hybrid proteoglycans of Engelbreth-Holm-Swarm mouse tumor.

The structure of the linkage region of chondroitin sulfate chains attached to the hybrid proteoglycans of the Engelbreth-Holm-Swarm mouse tumor was investigated. The peptidoglycan fraction which contains oversulfated chondroitin sulfate rich in the GlcA beta 1-3GalNAc-4,6-diO-sulfate unit and undersulfated heparan sulfate rich in GlcA beta 1-4GlcNAc and GlcA beta 1-4GlcN-2N-sulfate units was isolated after exhaustive protease digestion of the acetone powder of the tumor tissue, (GlcA, glucuronic acid; GalNAc, 2-deoxy-2-N-acetylamino-D-galactose). Glycosaminoglycans were released by beta-elimination using NaB3H4 and digested with chondroitinase ABC. The linkage region fraction was separated from heparan sulfate by gel filtration and fractionated by HPLC on an amine-bound silica column. Six radiolabeled compounds (L1-L6) were obtained and structurally analyzed by cochromatography with authentic hexasaccharide alditols recently isolated by us from the linkage region, and by digestion using chondroitinase ACII, alkaline phosphatase and beta-galactosidase in conjugation with HPLC. These compounds shared the conventional hexasaccharide backbone structure: delta GlcA beta 1-3GalNAc beta 1-4GlcA beta 1-3Gal beta 1-3Gal beta 1-4Xyl-ol, (delta GlcA, delta 4.5-GlcA or D-gluco-4-enepyranosyluronic acid). L1 was not sulfated or phosphorylated. L2 and L4 were monosulfated at C-6 and C-4 of the GalNAc residue, respectively. Upon alkaline phosphatase digestion, L3, L5 and L6 were converted to L1, L2 and L4, respectively. Analysis of the periodate oxidation products indicated that the phosphate group in L3, L5 and L6 is located at C-2 of Xyl-ol. These results suggest that Xyl-2-O-phosphate is associated with both 4-O-sulfated and 6-O-sulfated GalNAc units and does not directly determine the sulfation pattern of chondroitin sulfate.