Detection of tumor antigens in mouse and man by immunofluorescence

Antibodies to tumor antigens have been detected by several means including complement fixation, cytotoxic test, skin tests, precipitin tests, and immunofluorescence (IF) tests. This work employed the IF technique to detect antibodies in mice against Sarcoma 37 (S-37), Ehrlich's (EAT), and Krebs 11 (K-11) ascites tumors. Autochthonous and allologous antibodies to malignant melanoma antigens in man were also studied by the IF procedure. The fluorescent antibody FA method employed test sera which was allowed to react with target cells. The target cells were prepared from washed ascites tumor cells spread on coverslips and fixed in ether-alcohol or used as live cells. Explanted tumor cells (or other cells) from humans were grown on coverslips in tissue culture medium and used as target cells either fixed or alive as needed. The sera to be tested were reacted with the target cells and the antigen-antibody reactions were detected by the indirect IF technique. Photographs were used to document results. It was found that tumor-bearing mice and circulating antibodies in their serum to the ascites tumors as early as 9 to 10 days post challenge with a tumorogenic dose of approximately 10/6 cells ip. Subcutaneous tumors also induced antibody production to the tumor cells n the mouse. No antibodies to the ascites tumor cells were detected in the ascites fluid of mice bearing the tumor. Furthermore, the anti-tumor antibodies were directed against intracytoplasmic tumor antigens and no tumor surface or cell membrane antigens were detected by the IF procedure. Sera of rabbits immunized against the mouse ascites tumor cells exhibited the presence of cytotoxic antibodies; this indicated histoincompatibility differences between rabbit and mouse tumor cells, a condition not detected in sera of tumor-bearing mice. Tumor explants were grown in tissue culture media from two patients with malignant melanoma metastasis. These two patients had circulating antibodies to their own and the corresponding tumor cells of the other patient. Thirteen other patients with the same disease also had serum antibodies that reacted with the explanted malignant melanoma tumor cells. The tumor antigen(s) appeared as granular material in the cytoplasm of the explanted tumor cells. These antigens were not found to be present in human benign nevus cells, HeLa, human kidney, or guinea pig kidney cells. Antibodies against the intracellular tumor antigens of malignant melanoma could not cross the cell membrane of a live tumor cell; the antigens inside the tumor cells were detected only when fixed cells were employed. The significance of surface antigens versus intracellular antigens was considered. It appears that the tumor antigens of the mouse cancers studies as well as the human malignant melanoma tumor antigens are intracellular. Since gamma globulins cannot pass the cell membrane of living cells, the possibility is lessened that the antibody detected in malignant melanoma is related to resistance to this disease.