Immunohistochemical evidence for cell surface and Golgi localization of galactosyltransferase in human stomach, jejunum, liver and pancreas.

Abstract

Immunohistochemical localization of galactosyltransferase (UDP-galactose: 2-acetamido-2-deoxy beta-D-glucopyranose beta (1-4) transferase) in human tissue specimens of gastric and jejunal mucosa, exocrine pancreas, and liver was carried out at the light microscopic level using affinity purified rabbit anti-human milk galactosyltransferase antibodies. Intracellular localization of galactosyltransferase in epithelial cells appeared as a triangular compact structure close to the apical pole of the nucleus. In hepatocytes, the enzyme was found in discrete spots in the cytoplasm between the nuclei and the bile canaliculi. In addition to the intracellular, juxtanuclear location an intense reaction at the luminal part of the cell surface was found in the lining epithelium of the stomach, in enterocytes of the jejunal villus tips, and in ductular cells of the pancreas. Enterocytes located in the middle portion along the cryptvillus gradient exhibited cytoplasmic staining adjacent to the brush borders. Basolateral membranes appeared negative. Little or no enzyme could be demonstrated in cells belonging to the connective tissue. These results show that secretory cells contain a Golgi apparatus which can be visualized at the light microscopic level by virtue of its content in galactosyltransferase. Presence of galactosyltransferase antigen on the surface of certain cells supports the assumption that ectoglycosyltransferases do exist, at sites, however, apparently not involved in cell contact and adhesion.

Immunocytochemical demonstration of ecto-galactosyltransferase in absorptive intestinal cells.

“…Different antibody preparations, i.e., affinity purified according to conventional methods (3) as well as the one recovered from blotted enzyme, gave the same staining pattern and a comparable intensity of labeling. Thus, the immunostaining described in earlier work (17) corresponds to the findings presented here and represents bona fide galactosyltransferase antigenic sites. The immunocytochemical technique employed permits a lateral resolution that is at the best the size of the antibody molecule.…”

“…The preferential external localization of galactosyltransferase was also evident along the basolateral plasma membrane (Figs. 9-12), although the labeling here was less intense and had escaped detection at the light microscopic level (17). The density of gold particle labeling along the apical plasma membrane including the glycocalyx exceeded the labeling degree observed over positive trans-Golgi cisternae.…”

“…However, galactosyltransferase does not seem to be absent from all cell surfaces. Recent light microscopical immunohistochemical studies on intestinal cells (17) and the fallopian tube epithelium (7) strongly suggest the presence of galactosyltransferase on the apical membrane. Due to the limited resolution of this approach it remains unclear whether cell surface-related staining truly reflected ecto-galactosyltransferase associated with the outer side of the plasma mem-brane, or alternatively, enzyme trapped into the mucus or associated with intracellular structures.…”

Bovine galactosyltransferase: identification of a clone by direct immunological screening of a cDNA expression library.

“…Galactosyltransferase has also been localized to the plasma membrane of a diverse variety of cells and tissues by immunohistochemical (5)(6)(7)(8)(9) and biochemical procedures [comprehensively reviewed in Pierce et al (10) and Shur (11)]. This cell surface distribution has led to the postulate that, in addition to its biosynthetic role, galactosyltransferase has a functional role in intercellular recognition and/or adhesion (12).…”

A protein-specific monoclonal antibody to rat liver beta 1-->4 galactosyltransferase and its application to immunohistochemistry.

“…Immunohistochemical localization of galactosyltransferase has already been studied with affinity-purified polyclonal antibodies to this soluble enzyme from human milk (Roth et al, 1985;%asin et al, 1985;Davis et al, 1984;Pestalozzi et al, 1982;Roth and Berger, 1982;Berger et al, 1981). However, since the human milk enzyme has various abundant kinds of sugar chains, including blood group-related structures (Amano et al, 1991;Childs et al, 1986), the polyclonal antibodies may contain antibodies directed to such nonspecific carbohydrate chains.…”