HISTOCHEMICAL DEMONSTRATION OF INSULIN IN THE ISLETS OF LANGERHANS1

Abstract

The present studies demonstrate by in vitro experiments that insulin, precipitated from solution by Romeis'' fluid, remains insoluble during the histological procedures necessary for the preparation of tissue sections and during the histochemical steps necessary for the staining of sulfhydryl and disulfide groups. When the results of the in vitro experiments were applied to sections of the pancreas of 6 spp. of normal animals, the beta cells of the islets of Langerhans were strongly positive for sulfhydryl and disulfide groups whereas, the alpha cells were not. Intensity of histochemical staining of the islet beta cells of the normal animals corresponded to the insulin content as determined by bioassay by other investigators. If the pancreas was fixed in media which did not precipitate insulin irreversibly, under the conditions used, both alpha and beta cells stained weakly. The specificity of the histochemical method for sulfhydrvls and disulfide groups utilizing DDD as the reagent was proven for precipitated insulin in vitro as well as on sections of pancreas. A less specific method for sulfhydryls and disulfides utilizing tetrazolium salts, for which the specificity was not proven, gave similar staining results on sections of rat pancreas. Experiments designed to alter the insulin content of the pancreas were conducted so that the alterations were revealed by changes in the amount of reactive material with the histochemical method. These experiments on rabbits included the administration of alloxan, glucose, insulin, or diethyldithiocarbamate, starvation and partial pancreatectomy. Experiments on mice included both obese, hyperglycemic and Swiss mice, both controls and treated with diethyldithiocarbamate. In both normal and experimental animals when the histochemical staining of the sulfhydryl and disulfide groups was compared with that of aldehyde-fuchsin, results were similar in that the same cells stained. In experimental animals, results of both stains were directly proportional. The two methods differed in some other respects, the most notable of which was that aldehyde-fuchsin did not stain insulin in vitro. On the basis of in vitro and histochemical studies of the pancreas of normal and experimentally treated animals and comparison of these results with those of others who stained beta cells selectively or who bioassayed the pancreas for insulin content, it is concluded that the disulfide groups of insulin in islet beta cells have been demonstrated.