Beta-1,3-glucuronyltransferase 3 (glucuronosyltransferase I) (B3GAT3)

Abstract

β-1,3-glucuronyltransferase 3 (glucuronosyltransferase I: GlcAT-I) transfers GlcA from UDP-GlcA to the trisaccharide-serine, Galβ1-3Galβ1-4Xylβ1-O-Ser, to form the glycosaminoglycan (GAG)-protein linkage region, GlcAβ1-3Galβ1-3Galβ1-4Xylβ1-O-Ser, common to various proteoglycans. GlcAT-I comprises 335 amino acids and one N-glycan; it has a type II transmembrane orientation characteristic of many other glycosyltransferases (Kitagawa et al. 1998; Wei et al. 1999). The molecular mass of the GlcAT-I peptide backbone is 37 kDa, but the mass of the mature protein increases with glycosylation to around 47 kDa. Another characteristic feature in the amino acid sequence of GlcAT-I is a proline-rich domain (from Pro-30 to Pro-75) next to the transmembrane region; this juxtaposition occurs in several other glycosyltransferases including glucuronyltransferase-P (GlcAT-P) and glucuronyltransferase-D (GlcAT-D), which synthesize the precursor structure GlcAβ1-3Galβ1-4GlcNAc-R for the HNK-1 carbohydrate epitope GlcA(3-O-sulfate)β1-3Galβ1-4GlcNAc-R (Seiki et al. 1999; Shimoda et al. 1999; Terayama et al. 1997). Database searches indicate that the amino acid sequence of GlcAT-I is 43-46 % identical to GlcAT-P and GlcAT-D, respectively. The highest sequence identity is found in the COOH-terminal catalytic domain of each GlcAT. The catalytic domain of GlcAT-I is positioned immediately C-terminal to the proline-rich domain and stretches for 252 amino acids from Pro-68 to Glu-319; this domain has about 60 % identity with the catalytic domains from other GlcAT (Kitagawa et al. 1998; Wei et al. 1999), and the domain contains the four previously identified, highly conserved motifs (I-IV) found in putative GlcAT from different animal species (Shimoda et al. 1999; Terayama et al. 1997).