Molecular cloning and expression of endo-beta-N-acetylglucosaminidase D, which acts on the core structure of complex type asparagine-linked oligosaccharides.

Endo-beta-N-acetylglucosaminidase D (Endo D) produced by Streptococcus pneumoniae cleaves the di-N-acetylchitobiose structure in asparagine-linked oligosaccharides. The enzyme generally acts on complex type oligosaccharides after removal of external sugars by neuraminidase, beta-galactosidase, and beta-N-acetylglucosaminidase. We cloned the gene encoding the enzyme and expressed it as a periplasm enzyme in Escherichia coli. The first 37 amino acids in the predicted sequence are removed in the mature enzyme, yielding a protein with a molecular mass of 178 kDa. The substrate specificity of the recombinant enzyme is indistinguishable from the enzyme produced by S. pneumoniae. Endo-beta-N-acetylglucosaminidase A (Endo A) from Arthrobacter protophormiae, the molecular mass of which is 72 kDa, had 32% sequence identity to Endo D, starting from the N-terminal sides of both enzymes, although Endo A hydrolyzes high-mannose-type oligosaccharides and does not hydrolyze complex type ones. Endo D is not related to endo-beta-N-acetylglucosaminidases H, F(1), F(2), or F(3), which share common structural motifs. Therefore, there are two distinct groups of endo-beta-N-acetylglucosaminidases acting on asparagine-linked oligosaccharides. The C-terminal region of Endo D shows homology to beta-galactosidase and beta-N-acetylglucosaminidase from S. pneumoniae and has an LPXTG motif typical of surface-associated proteins of Gram-positive bacteria. It is possible that Endo D is located on the surface of the bacterium and, together with other glycosidases, is involved in virulence.

Genomic organization and promoter activity of embigin, a member of the immunoglobulin superfamily.

Embigin is a transmembrane glycoprotein belonging to the immunoglobulin superfamily, which is preferentially expressed in early stages of mouse embryogenesis and enhances integrin-mediated cell-substratum adhesion. The mouse embigin gene, which we cloned, spanned more than 50kb, in which nine exons were present. All exons contained protein-coding sequences. Each of the two immunoglobulin domains was encoded by two exons, and the C-proximal half of the second immunoglobulin domain and the transmembrane domain were in the same exon. These features are shared by the basigin gene; together with protein sequence homology, our results defined a family in the immunoglobulin superfamily, to which embigin and basigin both belong. The major transcriptional initiation site of embigin gene was 103 bases upstream from the translation initiation site, as determined by 5' rapid amplification of cDNA ends. A 3kb DNA fragment upstream from the transcriptional initiation site contained three Sp1 binding sites and had a promoter sequence capable of expressing the downstream gene not only in F9 embryonal carcinoma cells which express the gene, but also in L and G401 cells which do not, indicating the presence of a regulatory region outside the 3kb DNA region. Deletion analysis of the 3.5kb DNA fragment revealed that the region between -125 to +1, containing a single Sp1 binding site, is essential for transcription of the embigin gene.

Identification of the Ca(2+)-binding domains in reticulocalbin, an endoplasmic reticulum resident Ca(2+)-binding protein with multiple EF-hand motifs.

Reticulocalbin (RCN) is a member of the EF-hand Ca(2+)-binding protein family and is a luminal protein of the endoplasmic reticulum (ER) with a molecular weight of 44,000 [Ozawa, M. and Muramatsu, T. (1993) J. Biol. Chem. 268, 699-705]. Although RCN has six repeats of a domain containing an EF-hand motif, the varying degrees of divergence of the amino acid sequences of these domains from the EF-hand consensus sequences suggested that some domains might have lost their Ca(2+)-binding capability and adopted new functions. To identify the domains involved in Ca(2+)-binding, discrete domains of RCN were expressed in Escherichia coli, using the glutathione S-transferase fusion protein system. 45Ca2+ blot analysis of the resultant fusion proteins revealed that the first, fourth, fifth, and sixth domains bind Ca2+, however, the second and third ones do not. The fusion proteins containing all six domains, and the first and second domains, respectively, showed Ca(2+)-dependent increases in their electrophoretic mobilities, suggesting that Ca2+ induces a conformational change in reticulocalbin.