Bosynthesis of proteoglycans and hyaluronate in human fibroblast cultures established from normal and pathological corneas

Purification and characterization of chick corneal beta-D-glucuronyltransferase involved in chondroitin sulfate biosynthesis

Beta-D-Glucuronyltransferase, which transfers D-glucuronic acid (GlcA) from UDP-GlcA to N-acetyl-D-galactosamine (GalNAc) at the nonreducing end of chondro-pentasaccharide-PA (pyridylamino-), GalNAcbeta1-(4GlcAbeta1-3GalNAcbeta1)2-PA, was purified 339-fold with an 11.0% yield from 2-d-old chick corneas by chromatography on DEAE-Sepharose, WGA-agarose, heparin-Sepharose, and 1st and 2nd UDP-GlcA-agarose (in the presence of Gal) columns. The activity was detected by fluorescence of PA residues of the product. The purified enzyme has an optimum pH of 7.0 (Mes buffer), and much higher activity toward chondro-heptasaccharide-PA than toward the chondro-pentasaccharide-PA, but no activity toward p-nitrophenyl-beta-GalNAc. The enzyme activity was almost completely inhibited by GalNAc (20 mm). Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) of the purified enzyme fraction showed one band of 38 kDa with many other bands. The amino acid sequence was determined for the tryptic digests of the 38 kDa band protein. The sequences determined showed no homology to those of several beta-glucuronyltransferases reported previously. It seems that the enzyme is involved in the elongation of chondroitin sulfate chains in vivo.

Purification and characterization of N-acetylglucosaminyl sulfotransferase from chick corneas

N-Acetylglucosaminyl(GlcNAc) sulfotransferase, which transfers sulfate from adenosine 3'-phosphate 5'-phosphosulfate to GlcNAc at the nonreducing end of oligosaccharides, was purified 887-fold with a 8.4% yield from 2-day-old chick corneas by chromatography on CM-Sepharose, WGA-agarose, GlcNAc-agarose, and 3',5'-ADP-agarose columns. The purified enzyme has an optimum pH of 6.0 (Mes buffer) and specifically transfers a sulfate to GlcNAc at the nonreducing end but not to internal GlcNAc. The enzyme was stimulated by protamine and Mn(2+). SDS-polyacrylamide gel electrophoresis of the purified enzyme still showed two main bands (66 and 55 kDa) with some minor bands. It appears that this enzyme competes with beta-galactosyltransferase in binding to the nonreducing GlcNAc residue on KS synthesis; this suggests that the sulfation of the GlcNAc residue is coupled with the elongation of the KS chain.

Comparative analysis of proteoglycans synthesized by chick corneal stromal cells in cell culture and organ culture

Proteoglycans (PGs) synthesized by chick corneal stromal cells in cell culture and organ culture were metabolically radiolabelled with [35S]sulfate (for glycosaminoglycans) and [3H]leucine (for core proteins). Media, cell extracts and organ extracts from cultures were chromatographed on DEAE-Sephacel columns and separated into three fractions: the pass-through fraction (Fraction 1: nonsulfated PGs, hardly sulfated PGs, or glycoproteins with some oligosaccharides), the fraction eluted with a low salt concentration (Fraction 2: undersulfated PGs), and the fraction eluted with a high salt concentration (Fraction 3: highly sulfated PGs). The PG compositions of each fraction of cell culture and organ culture were then compared. While the proportions of highly sulfated KSPG in Fractions 3 of medium and cell extract of cell culture were both very low compared with those of medium and organ extract of organ culture, respectively, the proportions of highly sulfated CS/DS PG in Fractions 3 of those of cell culture were higher than those of organ culture. On the other hand, the proportions in the 35S activities of nonsulfated or undersulfated KSPG in Fractions 1 and 2 of medium and cell extract of cell culture were comparable to those of organ culture. Furthermore, the proportions of core proteins of undersulfated KSPG in Fractions 2 were higher in cell culture than in organ culture. These results show that, when the cells are cell-cultured, the degree of sulfation of KS chains decreases markedly, but the syntheses of the glycosaminoglycan backbone and core protein are maintained.

Glycosyltransferase and sulfotransferase activities in chick corneal stromal cells before and after in vitro culture

The expression type of proteoglycan in corneal stromal cells is known to change markedly when the cells are cultured in vitro. To determine which enzyme is primarily responsible for this change in chick corneal stromal cells, the activities of various glycosyltransferases and sulfotransferases were determined before and after in vitro culture of the cells. The activities of N-acetylglucosaminyltransferase, galactosyltransferase, and sulfotransferase, which are involved in keratan sulfate synthesis, were assayed using pyridylaminated N-acetyllactosamine-containing oligosaccharides as acceptor substrate; the activities of N-acetylgalactosaminyltransferase, glucuronyltransferase, and sulfotransferase, which are involved in chondroitin sulfate synthesis, were assayed using pyridylaminated chondrooligosaccharides as acceptor substrate. Of these enzymes, the sulfotransferase activity toward degalactosylated, pyridylaminated lacto-N-neotetraose and N-acetyllactosamine dimer (probably GlcNAc-6-sulfotransferase) decreased markedly after in vitro culture, whereas the galactosyltransferase activity increased. The chondroitin sulfate-sulfotransferase activities toward pyridylaminated chondrooligosaccharides hardly changed after in vitro culture. The marked decrease in the activity of the keratan sulfate-sulfotransferase corresponds to the marked decrease in keratan sulfate biosynthesis when the cells are cultured in vitro. These findings suggest that keratan sulfate-sulfotransferase (GlcNAc-6-sulfotransferase) is a key enzyme in keratan sulfate biosynthesis and that its decrease is primarily responsible for the marked decrease in keratan sulfate synthesis after in vitro culture.

Effect of human corneal fibroblasts on lymphocyte proliferation in vitro

Cocultivation of human corneal fibroblasts (CFB) with allogeneic peripheral blood mononuclear cells (PBL) induced a minimal lymphocyte proliferative response, even when the fibroblasts had been pre-treated with interferon-gamma (IFN-gamma) and were HLA-DR positive. IFN-gamma-treated CFB did not express detectable HLA-DQ alloantigens. Cocultivation of CFB with PBL in the presence of Concanavalin A (Con A) or Phytohemagglutinin inhibited mitogen-induced lymphocyte proliferation by 40-90% relative to PBL cultured without CFB. Induction of HLA-DR expression on the CFB did not alter their inhibitory properties. Addition of indomethacin to cultures reversed the effect of CFB on Con A responses. However, no difference between proliferative responses to HLA-DR positive or negative CFB was seen in the presence of indomethacin. This weak response to induced Class II alloantigens on CFB suggests that induction of Class II alloantigens on corneal stroma alone may be insufficient to sensitize a recipient for Class II alloantigen-driven corneal graft rejection.