Studies on the carcinoma-associated mucin type molecules MUC1 and CD43

The first two papers in this dissertation describe a novel monoclonal antibody (designated as 9H8) reacting with a carcinoma-associated epitope. Initial characterization of this antigen revealed that this is a high molecular mass protein appearing as a smear in western blot analysis. Preliminary epitope analysis showed that the epitope for this antibody is of carbohydrate nature by its sensitivity to periodate oxidation and by the possibility to block the antibody binding site by lectins (PNA). In the second paper a more detailed characterization of this epitope and its co-expression with other known carcinoma-associated oligosaccharide epitopes was undertaken. This epitope was shown to be closely related to the T-antigen (Thomsen-Friedenreich antigen) by its sensitivity to O-glycanase treatment, specifically cleaving the T-antigen. By using antibodies specific for the MUC1 apoprotein it was shown that the MUC1 protein is one carrier of the 9H8-epitope in carcinoma patients. By combining antibodies specific to 9H8, Si-Lex and Si-Lea epitopes, it was shown that the epitope for 9H8 was rarely co-expressed with Si-Lex and Si-Lea on the same molecule, though all were expressed on MUC1 mucins. This result shows that 9H8 epitope and Si-Lex/Si-Lea epitopes can be considered as mutually exclusive. This also indicates that sera of cancer patients can contain different populations of MUC1 mucin distinguishable by different sets of oligosaccharides. This knowledge should be taken into account when designing immunoassays exploiting MUC1-reactive antibodies.The third paper describes the effect of the MUC1 mucin on NK cell cytotoxicity when secreted or expressed on the cell membrane of target cells. The target cells showed increased resistance to lysis by NK cells when transfected with MUC1 cDNA compared to nontransfected cells. The same effect was observed when soluble MUC1 mucin was added to the incubation medium. The main conclusion of this study is that the MUC1 mucin when expressed by tumour cells may protect them from attack by NK cells.The last two papers are dealing with another mucin-type molecule, CD43. This protein has hitherto been considered as an exclusive leukocyte marker. By using CD43-specific monoclonal antibodies and digoxigenin-labelled RNA probes we were able to show that CD43 was expressed (both on protein and mRNA levels) in all studied colorectal adenomas (21/21) and in about 50% of carcinomas (18/34). Direct evidence showing a causal role of CD43 in colon tumourigenesis is lacking, but its involvement in leukocyte activation and impairment of apoptotic response suggests a role for CD43 in tumour development. In the last paper, we describe new monoclonal antibodies raised against the cytoplasmic tail of CD43 protein to facilitate detection of CD43 in a glycosylation independent manner in paraffin-embedded human nonhematopoietic tumours. By using these antibodies we observed that, in colon tumours, CD43 was predominantly located intracellularly and not concentrated to the plasma membrane as observed in leukocytes