Labelling of colloidal gold with IgE. A quantitative study using monoclonal IgE anti-beta-lactoglobulin and evaluation of the biological activity of the gold complex with RBL-1 cells.

Immunocytochemical markers prepared by labelling colloidal gold with antibodies are gaining wide acceptance both in transmission and scanning electron microscopy. However, detailed information on the process and extent of adsorption of IgG and IgE in particular are still lacking. The adsorption isotherm of mouse monoclonal 125I-IgE antibovine milk beta-lactoglobulin was studied quantitatively with colloidal gold buffered at pH 6.1-8.8 (28 nm in particle diameter). At low coverage of the particles (less that or equal to 5 molecules per particle), the isotherm was independent of pH. In the presence of a large excess of IgE, the highest coverage was obtained at pH 6.1 near the pI of IgE (5.2-5.8). The binding constants were higher at low coverage (side-on adsorption) than at high coverage where desorption was observed. IgE-Au markers were unreactive towards the immobilized antigen and did not bind to receptors for IgE of rat basophilic leukemia cells (RBL-1). The reactivity of immobilized anti-IgE antibodies with IgE-Au markers increased as a function of particle coverage. Mapping of RBL-1 cell membrane IgE receptors was achieved by incubating successively IgE-sensitized RBL-1 cells with anti-IgE antibodies and a protein A-gold marker at 4 degrees C. Surface clusters developed when the cells were incubated at 37 degrees C.

Effect of trimming inhibitors on the secretion and biological activity of a murine IgE monoclonal antibody.

Since secretion of IgE antibodies is known to be blocked by tunicamycin, the first aim of the present study was to determine at which step of glycosylation or processing secretion was restored. For this purpose, murine hybridoma cells secreting an anti beta-lactoglobulin IgE were incubated either in the presence of inhibitors of glucosidase I (castanospermine or N-methyl-1-deoxynojirimycin), or of an inhibitor of Golgi mannosidase II (swainsonine). Terminal galactoses predominate on the native IgE N-linked carbohydrate chains. The action of the trimming inhibitors, which results in changes in these terminal galactose residues, was monitored through detecting binding modifications to Concanavalin A and to the lectin of Ricinus communis. The antibody activity was also evaluated by a radioimmunoassay. It was shown that neither secretion nor anti beta-lactoglobulin activity of the IgE antibody are modified in the presence of any of the trimming inhibitors, whereas secretion is blocked in the presence of tunicamycin. Other biological activities of this IgE were investigated: no difference was observed in the binding of the carbohydrate-modified IgE molecules to normal mouse mast cells, nor to RBL-1 cells, as demonstrated by passive cutaneous anaphylaxis and in vitro binding tests respectively. However, traces of unglycosylated epsilon chain (mol. wt 61,000) found in tunicamycin treated cell supernatant did not bind to RBL-1 cell Fc epsilon receptors. These findings globally suggest that secretion occurs only if the tetradecasaccharide precursor of N-linked carbohydrate chains is transferred from its lipid-carrier to the polypeptide. Further, the presence of such non-processed oligosaccharides (Glc3Man9GlcNAc2) on IgE, does not seem to modify any of the biological activities of this molecule.

A mouse monoclonal IgE antibody anti bovine milk beta-lactoglobulin allows studies of allergy in the gastrointestinal tract.

A monoclonal IgE antibody directed against bovine milk beta-lactoglobulin (beta-LG) was produced by fusion of NSI myeloma cells with spleen cells of Balb/c mice immunized with alum-precipitated beta-LG. This antibody was found by radioimmunoassay to react with both native and aggregated beta-LG, but a positive passive cutaneous anaphylaxis reaction (PCA) was evident only with aggregated beta-LG. 1 ng of this purified antibody was capable of eliciting a PCA. The chemical and physical properties were characterized by amino acid and carbohydrate analysis and by ultracentrifugation. The epsilon-chain had an apparent molecular weight of 86,000 +/- 2,000 by polyacrylamide gel electrophoresis. An immediate hypersensitivity reaction within the gut was elicited by intravenous (i.v.) administration of the hybridoma ascitic fluid followed by feeding with aggregated beta-LG. Accumulation of liquid within the small intestine and diarrhoea were evident 30-90 min later. Intravenous injection of carbon particles revealed an increased permeability of the venulae from the submucosa and serosa. Histological examination showed oedema within the villae, without modification of the epithelium.

Experimental myasthenia in Balb/c mice immunized with rat acetylcholine receptor from rat denervated muscle.

A new model of an autoimmune disease of the neuromuscular junction was obtained by injection of acetylcholine receptor purified from rat denervated muscles into Balb/c mice. Anti-rat, then anti-mouse acetylcholine receptor antibodies, appear in mouse serum during the immunization procedure. Electrophysiological investigations performed on immunized mice reveal a neuromuscular block similar to that found in myasthenia gravis. Not a single mouse with objective signs of muscular weakness was lacking anti-mouse acetylcholine receptor antibodies but no correlation was found between their level and the severity of the disease.