Identification of Chondroitin Sulfate Glucuronyltransferase as Chondroitin Synthase-3 Involved in Chondroitin Polymerization

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 α-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, GlcUAβ1–3Galβ1-O-naphthalenemethanol and GlcUAβ1–3Galβ1-O-C2H4NH-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).

Chondroitin sulfates (CSs) 3 are universally ubiquitous molecules distributed on cell surfaces and in extracellular matrices (1-4). CS is a linear, sulfated polysaccharide composed of repeating disaccharide units consisting of alternating uronic acid (GlcUA) and GalNAc residues and synthesized as a proteoglycan bound to specific Ser residues in the core protein (1-4). Compelling evidence has shown that CS-proteoglycans play crucial roles in a number of physiological phenomena, such as cell adhesion, morphogenesis, neural network formation, and cell division (5,6). Therefore, an understanding of CS synthesis and its regulatory mechanism underlying diverse CS functions is essential.
The biosynthesis of CS is initiated by the addition of Xyl to specific serine residues in the core protein, followed by the sequential addition of two Gal residues and a GlcUA residue, forming the tetrasaccharide linkage structure GlcUA␤1-3Gal␤1-3Gal␤1-4Xyl␤1-O-Ser. Each transferring reaction is catalyzed by the corresponding glycosyltransferase. Then chondroitin polymerization with alternating GalNAc and GlcUA takes place, forming the repeating disaccharide region. Then a number of sulfotransferases modify the chondroitin backbone with sulfate at specific positions, resulting in the structural diversity of CS (7).

Materials-UDP-[U-
Construction of a Soluble Form of CSGlcA-T (ChSy-3)-The cDNA fragment of a truncated form of CSGlcA-T (ChSy-3), lacking the first 57 N-terminal amino acids, was amplified by PCR with KIAA1402 cDNA obtained from the Kazusa DNA Research Institute (Chiba, Japan) as a template using a 5Јprimer (5Ј-GAAGATCTAGAGCTCGGCTAGACCAAAG-3Ј) containing an in-frame BglII site and a 3Ј-primer (5Ј-GAA-GATCTCCATCTTGCCTTGCCCTTCC-3Ј) containing a BglII site located 72 bp downstream of the stop codon. PCR was carried out with KOD-Plus DNA polymerase (TOYOBO, Tokyo) for 30 cycles at 94°C for 30 s, 58°C for 30 s, and 68°C for 150 s in 5% (v/v) dimethyl sulfoxide. The PCR fragment was subcloned into the BamHI site of pGIR201protA (20), resulting in the fusion of the insulin signal sequence and the protein A sequence present in the vector, as described previously (8,14,16). The nucleotide sequence of the amplified cDNA was determined in a 377 DNA sequencer (PE Applied Biosystems). Soluble forms of ChSy-1, ChSy-2, and ChPF were constructed previously (8,14,16).
Site-directed Mutagenesis-A two-stage PCR mutagenesis method was used to construct CSGlcA-T (ChSy-3) mutant. Two separate PCRs were performed to generate two overlapping gene fragments using the soluble form of CSGlcA-T (ChSy-3) cDNA as a template. In the first PCR, the sense 5Ј-primer described above and the antisense internal mutagenic primer listed below were used: D184A 5Ј-GCACATAT-GTGGCATCCTGCATGATG-3Ј (the mutated nucleotide is underlined). In the second round of PCR, the sense internal mutagenic primer (complementary to the antisense internal mutagenic primer) and the antisense 3Ј-primer described above were used. These two PCR products were gel-purified and then used as a template for a third PCR containing the sense 5Ј-primer and the antisense 3Ј-primer described above. The final PCR fragment was subcloned into the BamHI site of pGIR201protA (20). The nucleotide sequence of the amplified cDNA was determined in a 377 DNA sequencer (PE Applied Biosystems).
Characterization of the Enzyme Reaction Products-Products of polymerization reactions on ␣-TM were isolated by gel filtration on a Superdex peptide column with 0.2 M NH 4 HCO 3 as the eluent. The [ 3 H]GalNAc-labeled oligosaccharide chains were released from ␣-TM by alkaline reduction treatment using 1.0 M NaBH 4 , 0.05 M NaOH and then exhaustively digested with chondroitinase ABC using 50 mIU of the enzyme for 1 h, as described previously (23). An aliquot of the enzyme digest was subjected to gel filtration on a Superdex peptide column, as described above. To determine the size of reaction products, the remaining aliquot was subjected to gel filtration on a Superdex 200 column with 0.2 M NH 4 HCO 3 as the eluent. Calibration of the Superdex 200 column was performed using a series of commercial polysaccharides of known size.
Pull-down Assays-The cDNA fragment of a truncated form of CSGlcA-T (ChSy-3), lacking the first 57 N-terminal amino acids of putative CSGlcA-T (ChSy-3), was amplified using a 5Ј-primer (5Ј-CGGAATTCAGAGCTCGGCTAGACCAAAG-3Ј) containing an in-frame EcoRI site and a 3Ј-primer (5Ј-CGGA-ATTCCCATCTTGCCTTGCCCTTCC-3Ј) containing an EcoRI site. The cDNA fragment of a truncated form of ChSy-2, lacking the first 129 N-terminal amino acids of ChSy-2, was also amplified using a 5Ј-primer (5Ј-GCTCTAGAGGCTGCCGGTCCG-GGCAG-3Ј) containing an in-frame XbaI site and a 3Ј-primer (5Ј-GCTCTAGACAATCTTAAAGGAGTCCTATGTA-3Ј) containing an XbaI site. Each DNA fragment was inserted into a pcDNA3Ins-His expression vector, resulting in the fusion of the protein with the insulin signal sequence and His 6 sequence present in the vector. Combinations of these constructs and the protein A-tagged expression vectors were transfected into COS-1 cells on 100-mm plates using FuGENE TM 6 (Roche Applied Science) according to the manufacturer's instructions. Two days after transfection, 1 ml of the culture medium was collected and incubated with 10 l of Ni 2ϩ -NTA-agarose (Qiagen) overnight at 4°C. The beads recovered by centrifugation were washed with TBS buffer containing Tween 20 three times and subjected to SDS-PAGE (7% gel), and proteins were transferred to a polyvinylidene difluoride membrane. The membrane, after blocking in PBS containing 2% skim milk and 0.1% Tween 20, was incubated with IgG antibody and then treated with anti-mouse IgG conjugated with horseradish peroxidase (Amersham Biosciences). Proteins bound to the antibody were visualized with an ECL advance kit (Amersham Biosciences).

Establishment of an Expression Vector for CSGlcA-T (ChSy-3) and Preparation of Cells That Stably Overexpress CSGlcA-T (ChSy-3)-
The cDNA fragment encoding CSGlcA-T (ChSy-3) was amplified from KIAA1402 cDNA as a template using a 5Ј-primer (5Ј-CGGAATTCCTGGCAGGGCCTACCACC-3Ј) containing an EcoRI site and a 3Ј-primer (5Ј-CGGAATTCCC-ATCTTGCCTTGCCCTTCC-3Ј) containing an EcoRI site. PCR was carried out with KOD-Plus DNA polymerase (TOYOBO) for 30 cycles at 94°C for 30 s, 53°C for 42 s, and 68°C for 180 s in 5% (v/v) dimethyl sulfoxide. The PCR fragments were subcloned into the EcoRI site of the pCMV expression vector (Invitrogen). The nucleotide sequence of the amplified cDNA was determined in a 377 DNA sequencer (PE Applied Biosystems). The expression plasmid (6.7 g) was transfected into HeLa cells on 100-mm plates using FuGENE 6 (Roche Applied Science) according to the manufacturer's instructions. Transfectants were cultured in the presence of 1,000 g/ml G418. Then resultant colonies were picked up and propagated for experiments.
Derivatization of Glycosaminoglycans from HeLa Cells Using a Fluorophore, 2-Aminobenzamide-Cells were homogenized in acetone and air-dried. The dried materials were digested with heat-pretreated (60°C for 30 min) actinase E in 200 l of 0.1 M borate-sodium, pH 8.0, containing 10 mM calcium acetate at 60°C for 24 h. Following incubation, each sample was treated with trichloroacetic acid, and the resultant precipitate was removed by centrifugation. The soluble fraction was extracted with ether. The aqueous phase was neutralized with 1.0 M sodium carbonate and adjusted to contain 80% ethanol. The resultant precipitate was dissolved in 50 mM pyridine acetate and subjected to gel filtration on a PD-10 column using 50 mM pyridine acetate as an eluent. The flow-through fractions were collected and evaporated to dryness. The dried sample was subsequently dissolved in water. Digestion with chondroitinase ABC (5 mIU) was conducted as described previously at 37°C for 1 h in a total volume of 10 l (28). Reactions were terminated by boiling for 1 min. Each digest was derivatized with 2-aminobenzamide and then analyzed by HPLC, as reported previously (29).

Glycosyltransferase Activity of CSGlcA-T (ChSy-3)-Recent
studies revealed that co-expression of any two of ChSy-1, ChSy-2 (CSS3), and ChPF augmented glycosyltransferase activities when compared with ChSy-1 or ChSy-2 expressed alone (14,16). These findings prompted us to investigate whether the co-expression of an additional ChSy family member, CSGlcA-T (ChSy-3), despite having only GlcAT-II activity responsible for the elongation of CS chains (13), with ChSy-1, ChSy-2 (CSS3), or ChPF might augment the glycosyltransferase activities. Hence, co-expression of CSGlcA-T (ChSy-3) with ChSy-1, ChSy-2 (CSS3), or ChPF was carried out. To facilitate the functional analysis of CSGlcA-T (ChSy-3), a soluble form of CSGlcA-T (ChSy-3) was generated by replacing the first 57 amino acids of the protein with a cleavable insulin signal sequence and a protein A IgG-binding domain, as described under "Experimental Procedures." Then the soluble protein was expressed in COS-1 cells as a recombinant protein fused with the protein A IgG-binding domain. The fusion protein secreted into the medium was adsorbed onto IgG-Sepharose beads for purification to eliminate endogenous glycosyltransferases, and then the protein-bound beads were used as an enzyme source. When CSGlcA-T (ChSy-3) bound to beads was evaluated for glycosyltransferase activities using chondroitin as an acceptor and either UDP-GalNAc or UDP-GlcUA as a donor substrate, weak GlcAT-II and GalNAcT-II activities were detected (Table 1). These results were distinct from the findings of Gotoh et al. (13), who reported only weak activities of GlcAT-II for CSGlcA-T (ChSy-3). Although it is unclear why they could not detect any GalNAcT-II activities for CSGlcA-T, it may be due to the difference in the epitope tag system used. (They used FLAG-tagged CSGlcA-T.) In addition, the discrepancy in enzyme specificity was not due to contamination with endogenous ChSy family members, because the recombinant enzymes used in this study were purified to apparent homogeneity, as assessed by SDS-PAGE (supplemental Fig. 1). Moreover, when soluble CSGlcA-T (ChSy-3) was co-expressed with soluble ChSy-1, ChSy-2 (CSS3), or ChPF, significant glycosyltransferase activities were detected (Table 1). Notably, co-expression of soluble CSGlcA-T (ChSy-3) with soluble ChSy-1 markedly augmented both the GalNAcT-II and GlcAT-II activities (Table 1).
Polymerase Activity of CSGlcA-T (ChSy-3)-Previously, coexpression of any two of soluble ChSy-1, ChSy-2 (CSS3), and ChPF showed not only the dramatic augmentation of glycosyltransferase activities of ChSy-1 or ChSy-2 (CSS3) but also polymerizing activities for disaccharide-repeating units of CS onto ␣-TM (14,16), which bears a truncated linkage region tetrasaccharide, GlcUA␤1-3Gal␤1-3Gal␤1-4Xyl (18). Hence, we measured the polymerizing activity upon co-expression of soluble CSGlcA-T (ChSy-3) with soluble ChSy-1,  ChSy-2 (CSS3), or ChPF. When ␣-TM was used as an acceptor substrate for the polymerization assay in the presence of UDP-[ 3 H]GalNAc and UDP-GlcUA, incubation of both of the coexpressed putative enzyme complexes yielded radiolabeled saccharide chains on ␣-TM. We then examined whether the length of the chondroitin chains formed by the co-expressed proteins in various combinations could be different. For this analysis, an equal amount of the co-expressed proteins in various combinations was used as each of the enzyme sources. Each reaction product obtained with ␣-TM was subjected to reductive ␤-elimination using NaBH 4 /NaOH, and the released radiolabeled saccharides were analyzed by gel filtration chromatography using a Superdex 200 column, as shown in Fig. 1. The lengths of chondroitin chains formed by co-expressed CSGlcA-T (ChSy-3) and ChPF were comparable with those of commercial chondroitin chains (a chemically desulfated derivative of whale cartilage chondroitin sulfate A), although chondroitin chains polymerized by the coexpressed CSGlcA-T (ChSy-3) and ChSy-1 were longer than those polymerized by the co-expressed CSGlcA-T (ChSy-3) and ChPF and shorter than those formed by the co-expressed ChSy-1 and ChPF. In contrast, co-expressed CSGlcA-T (ChSy-3) and ChSy-2 (CSS3) formed shorter chains compared with commercial chondroitin chains. It should be noted that no polymerization was induced on ␣-TM either through using one of the soluble forms of ChSy-1, ChSy-2 (CSS3), CSGlcA-T (ChSy-3), or ChPF or via a mixture of separately expressed soluble forms of any two of ChSy-1, ChSy-2 (CSS3), CSGlcA-T (ChSy-3), and ChPF (data not shown). These results clearly suggest the critical requirement of the co-expressed proteins in any two combinations of ChSy-1, ChSy-2 (CSS3), CSGlcA-T (ChSy-3), and ChPF for chondroitin polymerization.
To further determine substrate specificities of the co-expressed proteins in various combinations, chondroitin polymerizing activities were tested using two authentic synthetic substrates, GlcUA␤1-3Gal␤1-O-C 2 H 4 NH-Cbz and GlcUA␤1-3Gal␤1-O-NM, both of which share the disaccharide sequence with the glycosaminoglycan-protein linkage region tetrasaccharide. Each reaction product obtained with GlcUA␤1-3Gal␤1-O-C 2 H 4 NH-Cbz or GlcUA␤1-3Gal␤1-O-NM was analyzed by gel filtration chromatography using a Superdex 75 column, as shown in Fig. 2, A and B, respectively. When co-expression of ChPF with ChSy-1 was used as an enzyme source, polymerization activity was detected using GlcUA␤1-3Gal␤1-O-C 2 H 4 NH-Cbz as an acceptor substrate ( Fig. 2A). On the other hand, cotransfection of ChSy-1 and CSGlcA-T (ChSy-3), ChSy-1 and ChPF, or CSGlcA-T (ChSy-3) and ChPF showed polymerization activities using GlcUA␤1-3Gal␤1-O-NM as an acceptor substrate (Fig. 2B). Interestingly, co-expression of ChPF with CSGlcA-T (ChSy-3) formed the longest chondroitin chains when GlcUA␤1-3Gal␤1-O-NM was used as an acceptor (Fig. 2B). These results suggest that the various, co-expressed proteins that share polymerization activity might be critical for the differential assembly of CS chains on different core proteins by discriminating the amino acid sequences. In addition, to examine whether the formation of enzyme complexes might be affected by the pH, chondroitin polymerizing activities were also tested at different pH from the above one (pH 6.5). Interestingly, at pH 5.8, in sharp contrast to the results obtained at pH 6.5 (Fig. 2B), co-expression of ChPF with ChSy-1 formed longer chondroitin chains on GlcUA␤1-3Gal␤1-O-NM as an acceptor compared with the chain length synthesized on the same acceptor using the co-expression of CSGlcA-T (ChSy-3) with ChSy-1 (Fig. 2C).
Prompted by these observations, we next examined whether knockdown of CSGlcA-T (ChSy-3) expression by RNA interference decreases the amount of CS, as described under "Experimental Procedures." The efficiency of gene silencing was deter-  Table 2, transfection of the CSGlcA-T (ChSy-3) siRNA (ChSy-3 siRNA cells) resulted in a 50% knockdown of the CSGlcA-T (ChSy-3) mRNA and a 46% reduction of the amount of CS when compared with that of the control siRNA.
Contribution of Glycosyltransferase Activities of CSGlcA-T (ChSy-3) in Chondroitin Polymerization-When any two of ChSy-1, ChSy-2 (CSS3), and CSGlcA-T (ChSy-3) are coexpressed, the enzyme complex contains two sets of glycosyltransferase domains. Then, to clarify whether two glycosyltransferase domains of the complex equally contribute to polymerase activity or one totally contributes and the other serves as a chaperone-like molecule, such as ChPF, we tried to construct the CSGlcA-T (ChSy-3) mutant, which is expected to lack any glycosyltransferase activity. Based on the sequence alignment of ChSy family members, CSGlcA-T (ChSy-3) has only one putative DXD motif, QDD (Gln 182 -Asp 184 ), which is most likely responsible for UDP-sugar binding in many glycosyltransferases (32). It was therefore expected that the CSGlcA-T (ChSy-3) D184A mutant would not possess any glycosyltransferase activity. To confirm the expression and activity of the mutant protein, the soluble mutant was expressed in COS-1 cells, and the culture medium was purified with IgG-Sepharose. The purified mutant proteins were used for Western blotting analysis and evaluated for glycosyltransferase activities using chondroitin as an acceptor. As expected, neither GalNAcT-II nor GlcAT-II activity was detected (Table 1), although the mutant proteins were expressed (Fig. 6, lane 1). Nevertheless, co-expression of the soluble mutant with other wild-type ChSy family members resulted in an augmentation of glycosyltransferase activities, as in the case of that of wild-type CSGlcA-T (ChSy-3) (Table 1). However, when CSGlcA-T (ChSy-3) D184A was coexpressed with ChSy-1, the complex showed a 20% decrease in GalNAcT-II and a 47% decrease in GlcAT-II activities when compared with that of wild-type CSGlcA-T (ChSy-3) with ChSy-1 (Table 1). In addition, although the mutant was co-expressed with ChSy-2 (CSS3) or ChPF, each complex showed no GlcAT-II activity (Table 1).
One explanation for the existence of multiple enzyme complexes consisting of four ChSy family members that share chondroitin polymerization activity is that they initiate and polymerize chondroitin chains on different core proteins by discriminating the amino acid sequences. In fact, using GlcUA␤1-3Gal␤1-O-C 2 H 4 NH-Cbz as an acceptor substrate, chondroitin polymerization activity was only detected by coexpression of ChPF with ChSy-1 (Fig. 3A), whereas chondroitin