The efficiency of small enzyme-labeled tracers for the demonstration of intracellular antigen was investigated in tissues fixed with picric acid-formaldehyde. The influence of fixation on the immunological activity was tested in vitro by radial immunodiffusion. The experimental model consisted of newborn pig jejunum after absorption of ferritin from the intestinal lumen. Ferritin was located after 1 hr in vacuoles scattered in the cytoplasm of the absorptive cells and represented an easily recognizable intracellular antigen. After immunohistochemical treatments with antiferritin preparations, the distribution of labeling enzyme reaction product was examined by morphometry. The ratio of the labeled volume to the total volume of vacuoles containing ferritin indicated the degree of specific labeling of the antigen. In both direct and indirect methods, the degree of labeling was low when enzyme-labeled immunoglobulin G was the tracer. With antigen binding fragments (Fab), the labeling was significantly increased. In the indirect method, the degree of labeling was influenced by the first-step reagents. Onlywhen the serum titer was optimum was a high degree of labeling obtained. With antigen binding fragments or papain-digested serum the effect of the titer was negligible and maximum labeling was achieved. In both methods, with peroxidase as the labeling enzyme, a diffuse nonspecific deposition of reaction product was observed. This could be avoided by using cytochrome c instead.

This content is only available as a PDF.