Binding and Endocytosis of a Monoclonal Antibody to a High Molecular Weight Human Milk Fat Globule Membrane-associated Antigen by Cultured MCF-7 Breast Carcinoma Cells1

The aim of this study was to analyze whether a monoclonal antibody to human milk fat globule membrane-associated antigens, recognized specifically and homogeneously by human breast carcinoma cells but also by normal epithelial cells active in secretion, could be used to restrict the access of antitumoral drugs to cells exposing the epitope. The drugantibody conjugate to be used is constructed by means of a covalent peptidic linkage stable in extracellular medium but hydrolyzed by lysosomal enzymes after endocytosis of the drug-carrierconjugate. This monoclonal antibody specifically immunoprecipitates radioactive material from MCF-7 cells biosynthetically radiolabeled with galactose, glucosamine, palmitic acid, or acetic acid but not with mannose, leucine, or methionine. Upon polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate and dithiothreitol, the label migrates as two bands with apparent molecular weights of about 350,000 and 400,000. These bands disappear, or their molecular weight is affected, after treatment of the cells with cycloheximide or of cell lysates with trypsin, Primase, or neuraminidase but not treatment of the immunoprecipitate with endoglycosidase F. This suggests that these antigens are glycopro1eins with O-linked oligosaccharides containing sialic acid in the epitope. By analogy, they should be similar, if not identical, to those recognized by the monoclonal antibodies designated HMFG1 (H. Burchell, H. Durbin, and J. Taylor-Papadimitriou, J. Immunol., 131:508-513, 1983) and DF3 (H. Sekine, T. Ohno, and D. W. Kufe, J. Immunol., 735:36103615,1985). Binding at 4°Cof the 3H-labeled antibody by MCF-7 cells indicates the specific attachment of about 1.2 x 1(1'' IgG molecules per cells with a AHof about 14 nM. At 37°C, cells take up the '11-labelcd antibody in amounts much higher than the binding capacity. In addition to cellassociated material, labeled digestion products are released into the culture medium. Cell fractionation by differential centrifugation and isopycnic equilibration on sucrose gradient indicates that the bulk of cellassociated antibody is distributed like the marker enzyme of lysosomes. Although the total uptake of the antibody by the cells is unaffected by either 50 MMchloroquine or 3 Mg/ml cycloheximide, the release of digestion products is completely inhibited by chloroquine. Antigen-anti body dissociation is pH dependent, since, respectively, 50 and 84% of membrane-bound antibody are released during washing at pH 4.6 and 4.1. These results support the hypothesis that once bound to a plasma membrane epitope, the antibody is rapidly endocytosed and delivered to lysosomes for digestion. Antigen-antibody dissociation could occur as a result of a lower pH in endosomes and/or lysosomes. Since the uptake at 37°Cfar exceeds the 4°Cbinding capacity of the plasma membrane and is continuous over extended periods of time, there must be a mechanism allowing the continuous supply of antigen to the plasma membrane. The absence of an effect of chloroquine, a drug known to increase the endosomal and lysosomal pH, on the uptake of the antibody suggests that dissociation of antibody from epitope at acid pH is not required for continuous supply to the plasma membrane. Since cycloheximide, a protein synthesis inhibitor, does not affect the uptake of the antibody, Received 6/19/87; revised 12/15/87; accepted 2/22/88. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1This work was supported by the Belgian Fonds de la Recherche Scientifique Medicale (Grant 3.4548.84). 1Present address: The Weizmann Institute of Sciences, Department of Chem ical Immunology, P. O. B. 26 Rehovot, 76100 Israel. 3To whom requests for reprints should be addressed. antigen neosynthesis does not seem to be involved. Therefore, the antigen could derive from a large intracellularpool, which is supportedby binding of labeled antibody to cells preincubated with unlabeled antibody and then washed out in the presence of cycloheximide. All these results strongly suggest that this antibody is endocytosed, gains access to lysosomes, and could therefore be an appropriate carrier for drug targeting based on the lysosomotropic concept.