Separation and characterization of chondroitin 6-sulfotransferase and chondroitin 4-sulfotransferase from chick embryo cartilage

Two distinct sulfotransferases (chondroitin 6-sulfotransferase and chondroitin 4-sulfotransferase), which catalyzed transfer of sulfate to position 6 and position 4 of acetylgalactosamine residues of chondroitin, were extracted from epiphyseal cartilage of 14-day-old chick embryos and separated by gel chromatography on Sephacryl S-200 in the presence of 3 M guanidine-HCl. When the enzyme solutions containing 3 M guanidine-HCl were dialyzed against 0.02 M Tris-HCl, pH 7.2, containing 10% glycerol, chondroitin 4-sulfotransferase became almost insoluble, whereas chondroitin 6-sulfotransferase remained soluble. Endogenous acceptors for sulfate transfer were completely removed from both enzyme preparations. Addition of basic proteins and polyamines as well as Mn²⁺ to the incubation medium caused a stimulation of both sulfotransferases; the stimulation of chondroitin 6-sulfotransferase with these cations was higher than that of chondroitin 4-sulfotransferase. The K_m values for 3′-phosphoadenylyl sulfate of both enzymes were much smaller in the presence of protamine or spermine than in the presence of Mn²⁺. The two sulfotransferases differed in the requirement for sulfhydryl compounds; in the absence of sulfhydryl compounds, the activity of chondroitin 4-sulfotransferase was very low, whereas the activity of chondroitin 6-sulfotransferase was essentially unaffected. These observations indicate that at least two sulfotransferases are involved in the biosynthesis of chondroitin sulfate, and suggest that the production of the isomers of chondroitin sulfate in chondrocytes is affected by various factors such as the intracellular concentration of sulfhydryl compounds and basic substances.     

Secretion of chondroitin SO4 by monolayer cultures of chick embryo chondrocytes

Monolayer cultures of chick embryo tibial chondrocytes incorporate 35SO42- into chondroitin SO4 which is rapidly secreted from the cells into two extracellular pools. Part of the extracellular chondroitin SO4 is recovered in a soluble form in the culture medium, and the remainder is associated with the cell matrix from which it is released by isotonic trypsinization. At 38 degrees C labeled chondroitin SO4 appears in the cell matrix fraction within 5 min after addition of 35SO42- and in the culture medium fraction 15 min after 35SO42- is added. The intracellular pool of labeled chondroitin SO4 reaches a steady state level of 150 to 200 pmol of bound SO4 per 10(6) cells in 60 min, while the cell matrix and medium fractions increase at rates of 3 and 1 nmol of bound SO4 per h per 10(6) cells, respectively. After 4 h of labeling, less than 20% of the newly synthesized cell-associated chondroitin SO4 is in the intracellular fraction. By labeling cells for 15 min at 25 degrees C 80% of the cell-associated chondroitin 35SO4 is obtained in the intracellular fraction. This material is chased without lag into both the cell matrix fraction and the medium fraction. A mixture of NaF and NaCN, both at 30 mM, lowers the cellular ATP level to 15% of normal and blocks secretion of the intracellular chondroitin SO4 into both extracellular fractions. Colchicine at 10(-6) M gives a partial inhibition of both synthesis and secretion of chondroitinSO4. Sucrose density gradient sedimentation analysis of the intracellular chondroitin SO4 and the two extracellular fractions shows that all three fractions contain both a heavy and light proteoglycan fraction. The intracellular light proteoglycan fraction is secreted preferentially into the culture medium where it represents 30% of the total culture medium pool. The ratio of 6-sulfated GalNAc to 4-sulfated GalNAc in the heavy proteochondroitin SO4 fraction is approximately twice that found for the light fraction.     

Molecular cloning and expression of chondroitin 4-sulfotransferase

Chondroitin 4-sulfotransferase (C4ST) catalyzes the transfer of sulfate from 3'-phosphoadenosine 5'-phosphosulfate to position 4 of N-acetylgalactosamine residue of chondroitin. The enzyme has been previously purified to apparent homogeneity from the serum-free culture medium of rat chondrosarcoma cells (Yamauchi, A., Hirahara, Y., Usui, H., Takeda, Y., Hoshino, M., Fukuta, M., Kimura, J. H., and Habuchi, O. (1999) J. Biol. Chem. 274, 2456-2463). The purified enzyme also catalyzed the sulfation of partially desulfated dermatan sulfate. We have now cloned the cDNA of the mouse C4ST on the basis of the amino acid sequences of peptides obtained from the purified enzyme by protease digestion. This cDNA contains a single open reading frame that predicts a protein composed of 352 amino acid residues. The protein predicts a Type II transmembrane topology. The predicted sequence of the protein contains all of the known amino acid sequence and four potential sites for N-glycosylation, which corresponds to the observation that the purified C4ST is an N-linked glycoprotein. The amino acid sequence of mouse C4ST showed significant sequence homology to HNK-1 sulfotransferase. Comparison of the sequence of mouse C4ST with human HNK-1 sulfotransferase revealed approximately 29% identity and approximately 48% similarity at the amino acid level. When the cDNA was introduced in a eukaryotic expression vector and transfected in COS-7 cells, the sulfotransferase activity that catalyzes the transfer of sulfate to position 4 of GalNAc residue of both chondroitin and desulfated dermatan sulfate was overexpressed. Northern blot analysis showed that, among various mouse adult tissues, 5.7-kilobase message of C4ST was mainly expressed in the brain and kidney.     

Molecular cloning, expression, and chromosomal mapping of human chondroitin 4-sulfotransferase, whose expression pattern in human tissues is different from that of chondroitin 6-sulfotransferase

Chondroitin 4-sulfotransferase (C4ST) catalyzes the transfer of sulfate from 3'-phosphoadenosine 5'-phosphosulfate to position 4 of the N-acetylgalactosamine residues of chondroitin. We previously reported the cloning of C4ST cDNA from mouse brain. We here report the cloning and expression of human C4ST cDNA. The cDNA was isolated from a human fetal brain cDNA library by hybridization with a DNA probe prepared from rat poly(A)(+) RNA used for the cloning of mouse C4ST cDNA. The cDNA comprises a single open reading frame that predicts a Type II transmembrane protein composed of 352 amino acids. The protein has an amino acid sequence homology of 96% with mouse C4ST. When the cDNA was introduced into a eukaryotic expression vector and transfected in COS-7 cells, the sulfotransferase activity that transfers sulfate to both chondroitin and desulfated dermatan sulfate was overexpressed. Northern blot analysis indicated that human C4ST mRNAs (6.0 and 1.9 kb) are expressed ubiquitously in various adult human tissues. Dot blot analysis has shown that human C4ST is strongly expressed in colorectal adenocarcinoma and peripheral blood leukocytes, whereas strong expression of human chondroitin 6-sulfotransferase (C6ST) is observed in aorta and testis. These observations suggest that the expression of C4ST and C6ST may be controlled differently in human tissues. The C4ST gene was localized to chromosome 12q23.2-q23.3 by fluorescence in situ hybridization.     

Chondroitin SO4 catabolism in chick embryo chondrocytes

An enzyme preparation from cultured chick embryo vertebral chondrocytes attacks chondroitin SO4 oligosaccharides from the nonreducing terminal in a recycling pathway involving the sequential action of a beta-glucuronidase, a 4- or a 6-sulfatase, and a beta-N-acetylgalactosaminidase. The sequence is blocked by saccharo-1,4-lactone, an inhibitor of the beta-glucuronidase, or by 2-acetamido-2-deoxy-D-galactonolactone, an inhibitor of the beta-N-acetylgalactosaminidase. The level of 4-sulfatase activity is low relative to the other activities and limits the rate of catabolism of hybrid oligosaccharide structures containing both 6-sulfated galactosamine residues and 4-sulfated galactosamine residues. This results in the accumulation of shortened oligosaccharides, most of which have galactosamine-4-SO4 residues at their nonreducing terminals. In the presence of the lactone inhibitors, polymeric chondroitin SO4 is broken down by the enzyme preparation to oligosaccharides which are 10 to 15 monosaccharides long, indicating that degradation of chondroitin SO4 chains is initiated by an endoglycosidase which generates oligosaccharide substrates for the recycling exoglycosidase system.     

Sulfation of sialyl lactosamine oligosaccharides by chondroitin 6-sulfotransferase

We have previously shown that chondroitin 6-sulfotransferase(C6ST) catalyzes transfer of sulfate not only to position 6of GalNAc residue of chondroitin but also to position 6 of Galresidue of keratan sulfate. In this study, we examined the sulfationof sialyl lactosamine oligosaccharides by C6ST. C6ST catalyzedtransfer of sulfate to NeuAc     

Heat-shock response in cultured chick embryo chondrocytes. Osteonectin is a secreted heat-shock protein.

We investigated the induction of specific protein expression by heat shock in dedifferentiated and hypertrophic chick embryo chondrocytes in a culture system that allows 'in vitro' differentiation of cartilage cells [Castagnola, P., Moro, G., Descalzi-Cancedda, F. and Cancedda, R. (1986) J. Cell. Biol. 102, 2310-2317]. As control, we used cultures of embryonic fibroblasts from the whole body and from the skin. In the cell lysates of all cultures we identified four major heat-shock proteins (HSP), with a molecular size corresponding to HSP families previously described (HSP 90, HSP 70, HSP 47 and HSP 26). Some of these proteins were constantly induced when the temperature was raised, others were expressed in a more variable manner. Differences also existed in the relative amount of the HSP synthesized by the four cultures. When we specifically investigated HSP species released into the culture medium, we observed a 43-45 kDa protein constantly expressed and secreted in large amount by the cells. On the basis of its biochemical characteristic and its precipitation by specific antibodies, this protein has been identified as osteonectin (SPARC, BM-40).     

Depression by hyaluronic acid of glycosaminoglycan synthesis by cultured chick embryo chondrocytes

Proteins synthesized during the preimplantation period of mouse embryogenesis were labeled with radioactive tyrosine and lysine and fractionated by electrophoresis on polyacrylamide disc gels containing sodium dodecyl sulfate. For interstage comparisons and comparisons of the incorporation of different amino acids at the same developmental stages, the embryos were incubated with either ³H- or ¹⁴C-labeled amino acids. The embryos were then combined, and the proteins were isolated and electrophoresed simultaneously. The data were analyzed with a dual isotope computer program and expressed in the form of ${}^{14}\text{C}{}^3\text{H}$ ratios.Approximately 20–25 labeled protein components of apparent molecular weights between 25,000 and 115,000 can be defined, and 5 are most significant quantitatively. Of the latter, there are developmental increases in the rates of synthesis of 3 (with apparent molecular weights of 35,000 to 37,000, 37,000 to 41,000, and 66,000 to 70,000), a decrease in the rate of synthesis of another (53,000 to 57,000), and little change in the last (46,000 to 49,000). Developmental changes in the rates of synthesis of several other components are also demonstrated by the ${}^{14}\text{C}{}^3\text{H}$ incorporation ratios. The relative amounts of the different proteins synthesized by day 3 (early blastocyst) embryos over an 8-hr period remain constant, as does the relative labeling by lysine and tyrosine at each developmental stage examined. Similarly, there is no change in the pattern of the radioactive proteins when day 2 (8–16 cell) embryos are labeled for 2 hr and then incubated for an additional 24 hr. The greatest change in the overall pattern of protein synthesis occurs quite early, between day 1 (2 cell) and day 2, and lesser changes occur at later stages. These findings are in contrast to the major change in the rate of protein synthesis which occurs after day 2.     

Purification and characterization of a 3'-phosphoadenylylsulfate:chondroitin 6-sulfotransferase from arterial tissue

A 3'-phosphoadenylylsulfate:chondroitin sulfotransferase (EC 2.8.2.5) was purified to homogeneity (about 760-fold) from the cytosolic fraction of calf arterial tissue by Con A-Sepharose, ion exchange and affinity chromatography. The enzyme has a molecular mass of 38000 Da, optimal activity at pH 6.0 (100%) and 7.25 (75%), requires divalent cations for maximal activity (Mn2+ greater than Mg2+, Ca2+) and exhibits specificity towards desulfated chondroitin sulfate and oligosaccharides derived therefrom. The enzyme transfers sulfate groups from [35S]phosphoadenylylsulfate exclusively to C-6 OH groups of N-acetylgalactosamine units of the acceptor substrates. Maximal sulfate transfer occurs at 2mM chondroitin disaccharide units (100%), the transfer rates decreasing with decreasing chain length in the order deca (55%), octa (17%) and hexasaccharides (4%). Lineweaver-Burk plots revealed equal maximal velocities for chondroitin, deca-, octa- and hexasaccharide, but decreasing Km values. Chondroitin 4-sulfate has 21% of the acceptor potency exhibited by chondroitin, whereas dermatan sulfate, heparan sulfate and hyaluronate and the chondroitin tetrasaccharide showed no acceptor properties. Analysis of the reaction products formed by prolonged enzymatic sulfation of a reduced chondroitin hexasaccharide [GlcA-GalNAc]2-GlcA-GalNAc-ol revealed that the preterminal N-acetylgalactosamine from the non-reducing end and the internal N-acetylgalactosamine but not the N-acetylgalactosaminitol were sulfated and that no hexasaccharide disulfate was formed by the action of chondroitin 6-sulfotransferase. Chondroitin 6-sulfotransferase is considered to possess a binding region capable of accommodating a nonsulfated oligosaccharide sequence of at least six sugars and is believed to act in the course of chondroitin sulfate synthesis in cooperation with, but shortly after, the enzymes involved in the chain elongation reaction.     

The effect of embryo extract on the types of collagen synthesized by cultured chick chondrocytes.

Growth of embryonic chick chondrocytes in dialyzed embryo extract results in both a change in morphology of the cells toward that of a fibroblast and a change in the type of collagen synthesized from the cartilage-specific Type II collagen (chain composition [α1(II)]₃) to a mixture of Type I collagen (chain composition [α1(I)]₂α2) and the Type I trimer (chain composition [α1(I)]₃). Analyses after 6 days of growth in embryo extract show that the synthesis of only Type I collagen and the Type I trimer can be detected. However, on subculturing the cells to a low density and allowing a period of growth without embryo extract, colonies of chondrocytes reappear and the synthesis of Type II collagen apparently resumes. It is suggested that the observed changes represent a “modulation” in cell behavior, this being expressed not only by the morphological changes but also by changes in cell-specific protein synthesis as demonstrated by the changes in the type of collagen synthesized.     

Enzymatic sulfation of galactose residue of keratan sulfate by chondroitin 6-sulfotransferase

We have previously found that the purified chondroitin 6-sulfotransferase(C6ST), which transfers sulfate from 3'-phosphoadenosine 5'-phosphosulfate(PAPS) to position 6 of N-acetylgalactosamine in chondroitin,catalyzed the sulfation of keratan sulfate, and that both theC6ST activity and the keratan sulfate sulfotransferase (KSST)activity were expressed in COS-7 cells when C6ST cDNA was transfected.In this report we describe some properties of the KSST activitycontained in the purified C6ST, and characterize the sulfatedproducts formed from keratan sulfate and partially desulfatedkeratan sulfate. Optimal pH, requirement for cationic activators,and K_m value for PAPS of the KSST activity were very similarto those of the C6ST activity. ³⁵S-Labeled glycosaminoglycansformed from keratan sulfate and partially desulfated keratansulfate were N-deacetylated by treatment with hydrazine/hydrazinesulfate and then cleaved with HNO₂ at pH 4, and the resultingproducts were reduced with NaB³H₄. Analysis of the degradationproducts with paper chromatography and high performance liquidchromatography provided evidence that C6ST transferred sulfateto position 6 of galactose residue which was glycosidicallylinked to N-acetylglucosamine 6-sulfate residue or to N-acetylglucosamineresidue. Northern blot analysis using poly (A)⁺ RNA from 12-d-oldchick embryos indicated that the message of C6ST was expressednot only in the cartilage but also in the cornea in which keratansulfate is actively synthesized. chondroitin sulfate keratan sulfate glycosaminoglycan sulfotransferase hydrazinolysis deaminative cleavage     

Sulfation of N-acetylglucosamine by chondroitin 6-sulfotransferase 2 (GST-5)

Based on sequence homology with a previously cloned human GlcNAc 6-O-sulfotransferase, we have identified an open reading frame (ORF) encoding a novel member of the Gal/GalNAc/GlcNAc 6-O-sulfotransferase (GST) family termed GST-5 on the human X chromosome (band Xp11). GST-5 has recently been characterized as a novel GalNAc 6-O-sulfotransferase termed chondroitin 6-sulfotransferase-2 (Kitagawa, H., Fujita, M., Itio, N., and Sugahara K. (2000) J. Biol. Chem. 275, 21075-21080). We have coexpressed a human GST-5 cDNA with a GlyCAM-1/IgG fusion protein in COS-7 cells and observed four-fold enhanced [(35)S]sulfate incorporation into this mucin acceptor. All mucin-associated [(35)S]sulfate was incorporated as GlcNAc-6-sulfate or Galbeta1-->4GlcNAc-6-sulfate. GST-5 was also expressed in soluble epitope-tagged form and found to catalyze 6-O-sulfation of GlcNAc residues in synthetic acceptor structures. In particular, GST-5 was found to catalyze 6-O-sulfation of beta-benzyl GlcNAc but not alpha- or beta-benzyl GalNAc. In the mouse genome we have found a homologous ORF that predicts a novel murine GlcNAc 6-O-sulfotransferase with 88% identity to the human enzyme. This gene was mapped to mouse chromosome X at band XA3.1-3.2. GST-5 is the newest member of an emerging family of carbohydrate 6-O-sulfotransferases that includes chondroitin 6-sulfotransferase (GST-0), keratan-sulfate galactose 6-O-sulfotransferase (GST-1), the ubiquitously expressed GlcNAc 6-O-sulfotransferase (GST-2), high endothelial cell GlcNAc 6-O-sulfotransferase (GST-3), and intestinal GlcNAc 6-O-sulfotransferase (GST-4).     

Cytoskeleton and adhesion patterns of cultured chick embryo chondrocytes during cell spreading and Rous sarcoma virus transformation

The cytoskeleton and the adhesion complex of chick embryo chondrocytes maintained in vitro have been studied by fluorescence and interference reflection microscopy during the process of cell spreading. The pattern of actin-containing microfilaments and the distribution of vinculin speckles on adhesion plaques have been found to change as a function of the culture time. Newly plated chondrocytes adhere to the substratum mostly around a peripheral ring-like region and show a complex tridimensional array of microfilaments. When chondrocytes flatten, they develop stress fibres and show a diffuse system of vinculin-containing adhesion plaques scattered over the entire ventral side of the cells. Upon infection with Rous sarcoma virus (RSV) chondrocytes display one or more actin-containing ruffles located on the dorsal side similar to the 'actin flowers' earlier described in other cell types. These structures have been found to accumulate vinculin too. In chondrocytes infected with two td-ts mutants of RSV, 'actin flowers' have been found to persist at the restrictive temperature. At this temperature, however, in the majority of cells, stress fibres and adhesion plaques reappear.     

Functional analysis of the chondroitin 6-sulfotransferase gene in relation to lymphocyte subpopulations, brain development, and oversulfated chondroitin sulfates.

Chondroitin 6-sulfotransferase (C6ST) catalyzes the transfer of sulfate to position 6 of the N-acetylgalactosamine residue of chondroitin. To obtain direct evidence regarding the function of C6ST and its product, chondroitin 6-sulfate, in vivo, we isolated the mouse C6ST gene (C6st) and generated mice deficient in this gene (C6st(-/-)) by embryonic stem cell technology. C6st(-/-) mice were born at approximately the expected frequency and were viable through adulthood. In the spleen of C6st(-/-) mice, the level of chondroitin 6-sulfate became almost undetectable. Analyses of these knockout mice provided insights into the biosynthesis of oversulfated chondroitin sulfates in mice; chondroitin sulfate D in the brain of null mice and the cartilage and telencephalon of null embryos disappeared, whereas the chondroitin sulfate E level in the spleen and brain of the null mice was unchanged. Despite the disappearance of chondroitin sulfate D structure, brain development was normal in the C6st(-/-) mice. Further analysis revealed that the number of CD62L(+)CD44(low) T lymphocytes corresponding to naive T lymphocytes in the spleen of 5-6-week-old C6st(-/-) mice was significantly decreased, whereas those in other secondary lymphoid organs were unchanged. This finding suggested that chondroitin 6-sulfate plays a role in the maintenance of naive T lymphocytes in the spleen of young mice.     

Properties of chick embryo chondrocytes grown in serum-free medium

Chick embryo tibial chondrocyte growth and activities were compared in serum-free and serum-supplemented media. A basal salts medium containing equal volumes of Ham's F-12 and Dulbecco's modified Eagle's medium was supplemented with 10% fetal calf serum or with a mixture of bovine insulin, transferrin, fibroblast growth factor, dexamethasone, a prostaglandin E1 supplement, and a liposome supplement. Chondrocytes grew at identical rates in both media. Insulin, liposomes, and fibroblast growth factor were required for optimum growth in the serum-free medium, but removal of transferrin, dexamethasone, or prostaglandin E1 had little effect on the growth rate. In the serum-supplemented medium, the chondrocytes synthesized Type II collagen, Mr = 59,000 collagen, and both the large, cartilage-specific and the small ubiquitous proteochondroitin SO4 species typically produced by cultured chondrocytes. In the serum-free medium there was a shift toward synthesis of Type I collagen and a loss of the capacity to synthesize Mr = 59,000 collagen and the cartilage-specific proteochondroitin SO4. The loss of capacity for cartilage-specific proteochondroitin SO4 synthesis began immediately after replacement of the serum with the mixture of defined growth factors and the rate of loss was retarded but not reversed when serum was added back in place of the growth factors. When the serum and the mixture of growth factors were added together to the basal medium at the time of cell plating, the chondrocytes grew rapidly and retained their normal phenotype observed in serum-supplemented cultures. Thus, the serum appears to contain factors which are required for retention of the chondrocyte phenotype in culture over and above those factors necessary for cell growth.