Identification of a Novel Chondroitin Hydrolase in Caenorhabditis elegans*

Hyaluronidases have been postulated to be the enzyme acting at the initial step of chondroitin sulfate (CS) catabolism in vivo. Since chondroitin (Chn) but not hyaluronic acid (HA) has been detected in Caenorhabditis elegans, the nematode is a good model for elucidating the mechanism of the degradation of CS/Chn in vivo. Here we cloned the homolog of human hyaluronidase in C. elegans, T22C8.2. The Chn-degrading activity in vitro was first demonstrated when it was expressed in COS-7 cells. The enzyme cleaved preferentially Chn. CS-A and CS-C were also depolymerized but to lesser extents, and HA was hardly degraded. In order of preference, the substrates ranked Chn >> CS-A > CS-C >> HA. The products of the degradation of Chn by the enzyme were characterized by anion-exchange high performance liquid chromatography and delayed extraction matrix-assisted laser desorption ionization time-of-flight mass spectrometry. The structure of the major component in the digest was determined as GlcUAβ1-3GalNAcβ1-4GlcUAβ1-3GalNAc, where GlcUA and GalNAc represent d-glucuronic acid and N-acetyl-d-galactosamine, respectively, indicating that this enzyme is a Chn hydrolase, an endo-β-galactosaminidase specific for Chn. Investigation of the effects of pH on the activity revealed the optimum pH of Chn hydrolase to be 6.0. Since Chn in C. elegans has been demonstrated to play critical roles in cell division, Chn hydrolase possibly regulates the function of Chn in vivo. This is the first demonstration of a Chn hydrolase in an animal.

which Chn is degraded in vivo. Unexpectedly, we have identified a homolog of human hyaluronidase in the C. elegans genome. Chn and HA differ in structure only in the configuration at the C-4 position of the hexosamine residue. Thus, the gene may encode a Chn-degrading enzyme. In this study, we examined the activity of the recombinant protein to depolymerize Chn and demonstrated that the gene product hydrolyzes Chn, and to a much lesser extent, HA.
Preparation of Fluorescein 5(6)-Isothiocyanate (FITC)-labeled Chn-Chn (2 mg) was incubated with 1 mg of FITC dissolved in 400 l of 0.1 M sodium carbonate buffer, pH 9.5, and stirred at 4°C overnight in darkness (17). To remove excess FITC-labeling reagents, the reaction mixture was subjected to gel filtration on a PD-10 column with phosphate-buffered saline as the eluent.
Molecular Cloning of T22C8.2-A tBLASTn analysis of the GenBank TM data base, using the sequences of human hyaluronidases, identified a highly homologous clone, T22C8.2 (GenBank TM accession number NM_063429) in the C. elegans genome. The cDNA sequence was obtained using WormBase (accession number WBGene 00011923). Since Chn but not HA was found in C. elegans, this sequence was predicted to encode a Chn hydrolase (Chnase).
Cloning of C. elegans Chnase cDNA-The putative fulllength open reading frame encoding a presumed Chnase was amplified from C. elegans N2 cDNA by two rounds of PCR using specific primers corresponding to the sequences in the 5Јand 3Ј-noncoding regions. The first PCR was performed with the primers, 5Ј-CTC GAG TGG GGT GAA GTT TGG TAG   GA-3Ј and 5Ј-CTC GAG TCT CAA AGT CAA AAT GGC  AAA-3Ј. The second PCR was performed with the nested primers, 5Ј-CTC GAG GGC AAA TAA AGC TTG ATC CAA-3Ј  and 5Ј-CTC GAG AGA AGA TTG CAC TGC CAA CA-3Ј. Each PCR was carried out with KOD-Plus DNA polymerase (Toyobo, Tokyo, Japan) in the presence of 5% (v/v) dimethyl  sulfoxide for 30 cycles at 95°C for 30 s, 54°C for 45 s, and 68°C for 2 min. The amplified cDNA fragment of expected size (ϳ1.4 kb) was subcloned into a pGEM-T Easy vector (Promega, Tokyo) and sequenced in a CEQ 8000 DNA sequencer (Beckman Coulter).
Construction of an Expression Vector Containing a cDNA Fragment Encoding a Soluble Form of Chnase-The DNA fragment, which encodes the putative Chnase protein lacking the first N-terminal 26 amino acids, the predicted transmembrane region, was amplified by PCR with the pGEM-T Easy vector containing the full-length form of Chnase as a template, using a 5Ј primer containing an in-frame BamHI site (5Ј-GCG GAT CCG GTT CGG GAG CCT CC-3Ј) and a 3Ј primer containing a BamHI site located 19 bp downstream from the stop codon (5Ј-GCG GAT CCT TTT GTT CAC TAT TAT-3Ј). PCR was carried out with KOD-Plus DNA polymerase for 30 cycles at 95°C for 30 s, 55°C for 45 s, and 68°C for 2 min. The amplified PCR fragment was subcloned into the BamHI site of the expression vector p3XFLAG-CMV-8 (Sigma), resulting in the fusion of Chnase to the preprotrypsin leader sequence and the 3XFLAG tag sequence present in the vector.
Expression of a Soluble Form of Chnase-The expression plasmid (6.6 g) was introduced into COS-7 cells using FuGENE TM 6 (Roche Diagnostics, Basel, Switzerland) according to the manufacturer's instructions. After 3 days of culture at 37°C, 10 ml of the culture medium was collected and incubated with 1 ml of anti-FLAG M2 affinity gel (Sigma) at 4°C overnight. The resin was washed with 25 mM Tris-buffered saline containing 0.05% Tween 20 (TBST), and bound proteins were eluted with 5 ml of 100 g/ml of the FLAG-peptide (Sigma). The eluates were incubated with the SDS sample buffer and dithiothreitol solution (New England Biolabs, Ipswich, MA) in a total volume of 30 l at 100°C for 5 min and subjected to SDS-PAGE. Proteins were resolved on 7.5% SDS-polyacrylamide gels and visualized by silver staining or transferred to a polyvinylidene difluoride membrane. The membrane was incubated with anti-FLAG M2 monoclonal antibody (Sigma) diluted 1:1,000 with TBST overnight and then with the ECL TM horseradish peroxidase-labeled anti-mouse IgG antibody (GE Healthcare) diluted 1:10,000 with TBST. The bound antibody was detected using an ECL Advance western blotting detection kit (GE Healthcare).
Measurement of the Enzyme Activity-The cells transfected with T22C8.2 were cultured for 2 days, and 3 ml of the medium was purified with 15 l of anti-FLAG M2 affinity gel at 4°C overnight. The resin was washed with TBST and subsequently with 50 mM phosphate, pH 6.0, containing 150 mM NaCl, and then resuspended in 4 l of the same buffer containing ϳ6 g of the FITC-labeled Chn. The mixture was incubated at 28°C for 12 h. The resin was removed by filtration using Ultrafree-MC (Durapore polyvinylidene difluoride 0.45 m) (Millipore), and the sample solution was subjected to gel filtration chromatog-raphy on a Superdex peptide column equilibrated with phosphate-buffered saline. Eluates were monitored by fluorescence with excitation and emission wavelengths of 490 and 520 nm, respectively.
The purified enzyme-bound resin (5 l) was also incubated with FITC-Chn (ϳ18 g) in 50 mM phosphate buffer, pH 6.0, at 15, 20, or 28°C for 2 h to determine its optimal temperature. The degrading activity of Chnase was assessed based on the proportion (in percentage) of low molecular weight fragments of FITC-Chn in the digests, which was determined by gel filtration HPLC on a Superdex peptide column by measuring the peak areas. The total fluorescence intensity was calculated based on the areas from 19 to 51 min on each chromatogram and regarded as the total amount of FITC-Chn. With this value taken as 100%, the proportion (in percentage) of the low molecular weight fragments (the area from 34 to 51 min) was calculated.
To determine the optimal pH, the purified enzyme-bound resin (5 l) was incubated with nonlabeled Chn (ϳ10 g) in 50 mM phosphate buffers, pH 4.0 -8.0, at 28°C for 30 min. After incubation, the resin was removed by filtration using Ultrafree-MC, each sample was labeled with 2AB (15), and excess 2ABderivatizing reagents were removed by chloroform extraction (18). Aliquots of the 2AB-derivatives (ϳ100 pmol) were analyzed by anion-exchange HPLC on an amine-bound silica PA03 column (4.6 ϫ 250 mm, YMC Co., Kyoto, Japan) using a linear gradient of NaH 2 PO 4 from 16 to 800 mM over 60 min at a flow rate of 1 ml/min. Eluates were monitored by fluorescence with excitation and emission wavelengths of 330 and 420 nm, respectively. The degrading activity of Chnase was assessed based on the proportion (in percentage) of 2AB-labeled oligosaccharides formed, which was determined by measuring the peak areas. Taking the total amount of 2AB-oligosaccharide generated by exhaustive digestion with Chnase as 100%, the proportion (in percentage) of 2AB-labeled oligosaccharides was calculated.
Nonlabeled Chn, CS-A, CS-C, heparan sulfate, dermatan, and HA preparations from human umbilical cord as well as S. pyogenes were incubated individually with the Chnase-bound resin at 28°C. After incubation, the resin was removed by filtration using Ultrafree-MC, and each sample was labeled with 2AB and analyzed by anion-exchange HPLC or by gel filtration HPLC on a Superdex peptide column equilibrated with 0.2 M NH 4 HCO 3 . The reaction rate was measured as moles of product formed/min, and apparent Michaelis-Menten constants were determined by fitting the data to the Michaelis-Menten . The major peaks detected by gel filtration HPLC (ϳ10 pmol) were collected and characterized by mass spectrometry (MS) and/or anion-exchange HPLC.
Transglycosylation activity of Chnase was also examined. A typical transglycosylation reaction was carried out as follows (19). 1.5 g of Chn as a donor, 60 pmol of 2AB-labeled Chnhexasaccharide as an acceptor, and the purified Chnase-bound resin (10 l) were incubated in a 50 mM phosphate buffer containing 150 mM NaCl, pH 5.0, or 150 mM Tris-HCl buffer, pH 7.0, at 37°C for 1 h. As positive control experiments, 1.5 g of HA as a donor, 60 pmol of 2AB-labeled HA-hexasaccharide as an acceptor, and 10 national formulary units of testicular hyaluronidase were incubated (1 national formulary unit corresponds to the amount of the enzyme which hydrolyzes 74 g of hyaluronate/min).
Characterization of the Reducing Terminal Sugar Residue of the Degradation Products-The 2AB-derivative of the degradation products of Chn was digested with CSase AC-II (5 mIU) in 50 mM sodium acetate buffer, pH 6.0 (20). The digest was analyzed by anion-exchange HPLC on an amine-bound silica PA03 column as described above. Delayed

RESULTS
Molecular Cloning of T22C8.2-Screening of the nonredundant data base at the NCBI, National Institutes of Health (Bethesda, MD), using the deduced amino acid sequences of human hyaluronidases, identified a homolog designated T22C8.2 (WormBase accession T22C8.2) in C. elegans (Fig. 1). It contained a 5Ј-untranslated region of 151 bp, a single open reading frame of 1,377 bp coding for a protein of 458 amino acids with four potential N-glycosylation sites, and a 3Ј-untranslated region of 84 bp. The SOSUI system, a web site for predicting the secondary structure of membrane proteins from protein sequences, revealed a prominent hydrophobic segment of 23 amino acid residues in the N-terminal region of T22C8.2, predicting that the protein has a type II transmembrane topology. Data base searches suggested that the amino acid sequence displayed 28,27,27,27, and 28% identity with human hyaluronidases 1, 2, 3, and 4 and PH-20, respectively (Fig. 1). Thus, the structural features of the identified protein sequence suggest that the gene product of T22C8.2 is involved in the degradation of HA. However, C. elegans produces Chn but not HA (13). Hence, the protein encoded by T22C8.2 may have Chndegrading activity and play a crucial role in the metabolism of Chn in C. elegans. Thus, the putative catalytic activity of a recombinant form of T22C8.2 was investigated.
Demonstration of the Chn-degrading Activity-To facilitate the functional analysis of Chnase, a soluble form of the protein was generated by replacing the putative signal sequence with a cleavable preprotrypsin leader sequence and a 3XFLAG tag fusion protein as described under "Experimental Procedures." The soluble protein was expressed in COS-7 cells at 37°C as a recombinant protein fused with the 3XFLAG tag. The fusion protein secreted in the medium was adsorbed onto an anti-FLAG M2 affinity gel for the elimination of endogenous glycosidases, and then the protein-bound resin was used as an enzyme source. The purity of the enzyme was examined by western blotting as well as silver staining. Upon the silver staining, in contrast to the sample from mock-transfected cells, almost a single induced band was discernible at a molecular   mass of 80 kDa (supplemental Fig. 1A), although the expression level was low. The band with the same molecular size was clearly detected by western blotting under reducing and nonreducing conditions (supplemental Fig. 1, B and C). Since the expected molecular mass of this polypeptide is 53 kDa, the recombinant protein detected by western blotting seems to be posttranslationally modified, presumably glycosylated. The bound fusion protein was assayed for Chn-degrading enzyme activity at 28°C for 12 h using FITC-Chn as a substrate, and the digest was analyzed by gel filtration HPLC on a Superdex peptide column (Fig. 2). FITC-Chn was efficiently degraded into oligosaccharides (Fig. 2B). No detectable catalytic activity was observed for the control sample prepared from a conditioned medium of mock-transfected cells. Thus, Chn-degrading activity was demonstrated for T22C8.2. Note that the oligosaccharides detected were labeled with FITC at the carboxyl groups of some GlcUA residues and that elution of FITC-labeled oligosaccharides was retarded by hydrophobic interactions of FITC with the resin. Therefore, the positions of FITClabeled Chn oligosaccharides could not be calibrated.
The effect of temperature on the Chnase activity was examined by incubating at 15, 20, or 28°C. With the activity at 28°C taken as 1.0, the relative activity detected at 15 or 20°C was 0.5 or 0.7, respectively. The following experiments were conducted by incubating at 28°C.
Mechanism of Degradation of Chn-To investigate the Chndegrading activity of T22C8.2 in more detail, the enzyme digest of Chn was labeled with 2AB (see "Experimental Procedures") and subjected to gel filtration HPLC on a Superdex peptide column (Fig. 3A).  /z 1,277, 1,299, 1,315, and  1,274, respectively (Fig. 4B, Table 1), suggesting that the component in this fraction was HexUA 3 HexNAc 3 -2AB. Thus, the major components in fractions O-1 and O-2 were assumed to contain saturated tetra-and hexasaccharides, respectively, indicating that the enzyme is not an eliminase but a hydrolase. Hence, the enzyme was identified as Chn hydrolase.
To investigate whether the enzyme is a hexosaminidase or glucuronidase, the terminal sugar residue of the oligosaccharides in the enzyme digest was analyzed. The digest was labeled with 2AB and analyzed by anion-exchange HPLC before and after digestion with CSase AC-II (Fig. 5). Before the digestion, peaks were eluted at the positions corresponding to those of the  2AB-derivatives of the authentic Chn tetra-(GlcUA-GalNAc-GlcUA-GalNAc), hexa-(GlcUA-GalNAc-GlcUA-GalNAc-Glc-UA-GalNAc), and octasaccharide (GlcUA-GalNAc-GlcUA-Gal-NAc-GlcUA-GalNAc-GlcUA-GalNAc), respectively, which were produced by digestion with testicular hyaluronidase. After the digestion with CSase AC-II, peaks were eluted at the position of the unsaturated disaccharide, ⌬HexUA-GalNAc-2AB, indicating that the reducing terminal sugar residue of the products of digestion by Chnase is GalNAc and that Chnase is a hexosaminidase, not a glucuronidase.
Substrate Specificity of Chnase-To characterize the specificity of Chnase, CS-A, CS-C, dermatan, heparan sulfate, HA from human umbilical cord, and HA preparations from S. pyogenes (an average molecular mass of 980, 132, and 35 kDa) were used as substrates. The incubation mixtures were analyzed by gel filtration HPLC on a Superdex peptide column after labeling with 2AB to detect the newly formed reducing ends (Fig. 3). The chromatograms showed that Chnase had activity to degrade CS-A (Fig. 3B), CS-C (Fig. 3C), and HA (Fig. 3D), but dermatan and heparan sulfate were not depolymerized (data not shown). No difference was observed in the amount of oligosaccharides formed among the HA preparations used, suggesting that the chain length of HA does not affect the susceptibility to the enzyme. Although the same amount of each substrate was used under the same conditions, the amounts of the products generated by the digestion were different. Chn was almost completely broken down into tetra-and hexasaccharides (Fig. 3A), whereas CS-A, CS-C, and HA were degraded to a lesser extent. The relative rates of the degradation of these polysaccharides suggested that in order of preference by Chnase, the substrates ranked as follows: Chn Ͼ Ͼ CS-A Ͼ CS-C Ͼ Ͼ HA.
To clarify whether Chnase recognizes and acts on the nonsulfated or sulfated structure in the CS chains, the peaks A-1 and C-1 (Fig. 3, B and C)  The Optimum pH for the Chnase Activity-The effect of pH on the Chnase activity was examined by incubating the Chnasebound resin with FITC-labeled Chn over a range of pH values from 4.5 to 8.0 (Fig. 7). The results indicated the optimum pH of Chnase to be 6.0.
Kinetic Analysis of Chnase-The initial reaction rates and substrate concentrations (as disaccharide) were used for kinetic analysis by Lineweaver-Burk plot (supplemental Fig. 2). The apparent Michaelis-Menten constants as well as V max for Chn, CS-A, and HA from S. pyogenes (35 kDa) were determined as 0.12, 0.48, and 49.3 mM as well as 3.8, 1.2, and 0.4 pmol/min, respectively (Table 2).
Transglycosylation Activity-Since testicular hyaluronidase catalyzes transglycosylation as well as hydrolysis on HA (19), transglycosylation activity of nematode Chnase was examined by incubating with Chn as a donor and 2AB-labeled Chn-hexasaccharide as an acceptor. However, 2AB-labeled oligosaccharides longer than hexasaccharide were not detected under the conditions used (results not shown), although the transglycosylation activity by sheep testicular hyaluronidase was confirmed. Thus, Chnase has no or little, if any, transglycosylation activity.

DISCUSSION
In the present study, we demonstrated the Chn hydrolase activity for the C. elegans T22C8.2 protein and characterized its specificity. Although the clone T22C8.2 was found as the C. elegans ortholog of human hyaluronidase, it was not a hyaluronidase but a Chn hydrolase; the enzyme much prefers Chn to HA, and only Chn, no HA, is detected in C. elegans (13). This is the first report of a Chn-degrading enzyme in a metazoan or an animal. Although several CSases have been isolated from bacteria, they are not hydrolases but eliminases (23)(24)(25). These bacterial enzymes digest CS as a source of nutrients. Since typical CS-PGs are not synthesized in protozoa, bacterial enzymes are not involved in the in vivo metabolism of structural components of protozoa. Hyaluronidases from vertebrate tissues and the venom of snakes, bees, spiders, and stonefish can depolymerize both HA and Chn/CS and are known as spreading factors (26). These hyaluronidases digest Chn/CS more slowly than HA (27,28), and their genuine substrate is HA rather than Chn/CS. Thus, these hyaluronidases should not be termed Chnases. The enzyme reported here is the first Chn hydrolase to be identified.
Although Csoka et al. (27) described hyaluronidase-4 as a CS-specific enzyme based on the preliminary evidence, the actual data and conditions for the activity detection were not