Thirteen sera from children with ulcerative colitis were examined for antibodies reacting with constituents of human colonic tissue by means of immunofluorescent methods. 3 out of 10 sera reacted positively when tested by the direct staining method while 6 out of 13 reacted positively when tested by the indirect method with conjugates of rabbit anti-human gamma globulin. The specificity of the reactions could be confirmed by inhibition tests. 16 sera from healthy children and adults yielded completely negative results. The staining capacity of various sera was correlated to their hemagglutinating titer when they were tested with sheep erythrocytes, coated with phenol-water extract of human colon. Absorption experiments indicated that the stainable antigen was also present in the extracts used for the hemagglutination experiments.

In unfixed tissue sections, fluorescent antibodies were adsorbed onto the epithelial cells of the mucosa. Adsorption on epithelial basement membranes could not be demonstrated. Fluorescent H agglutinins, isolated from eel serum, were adsorbed onto the same mucosal structures of human colon (blood group O) as the antibodies in the sera of patients with ulcerative colitis. However, any immunological relationship between H substance and the colonic antigen of ulcerative colitis could be ruled out by cross-inhibition and hemagglutination inhibition experiments.

Fluorescent serum from patients with rheumatoid arthritis also stained sections of human colon but the localization of the stainable antigens was different from that visualized with the ulcerative colitis sera. Inhibition experiments indicated that the rheumatoid arthritis serum contained antibodies staining colon antigens different from those reacting with antibodies in the ulcerative colitis sera. Sera from patients with systemic lupus erythematosus or with the nephrotic syndrome, which all hemagglutinated erythrocytes coated with colon extract, did not stain the sections of the colon tissue either because of suboptimal antibody concentration or because of a difference in type or localization of the antigen.