Immunohistochemical localization of a gap junction protein (connexin43) in the muscularis externa of murine, canine, and human intestine

Electron-microscopic studies have revealed a heterogeneous distribution of gap junctions in the muscularis externa of mammalian intestines. This heterogeneity is observed at four different levels: among species; between small and large intestines; between longitudinal and circular muscle layers; and between subdivisions of the circular muscle layer. We correlated results obtained with two immunomethods, using an antibody to the known gap-junctional protein (connexin43) with ultrastructural findings, and further evaluated the respective sensitivity of these two approaches. For comparative reasons we also included the vascular smooth muscle of coronary arteries into our study. Two versions of the immunotechnique (peroxidase-antiperoxidase and fluorescence methods) were applied to frozen sections of murine, canine, and human small and large intestines, as well as to pig coronary artery. In the small intestine of all three species a very strong reactivity marked the outer main division of the circular muscle layer, while the longitudinal muscle layer as well as the inner thin division of the circular muscle layer were negative. In murine and human colon both muscle layers were negative, while in canine colon the border layer between the circular muscle and the submucosa reacted strongly, and scattered activity was found in the portion of the circular muscle layer (one tenth of its thickness) closest to the submucosa. The remainder of the circular muscle layer and the entire longitudinal muscle layer were negative in the canine colon. In the coronary artery we could not confirm the positive, specific labeling reported by other investigators (l.c.). In conclusion, we found close correlations at all four above-mentioned levels in the distribution of gap junctions in the gut musculature, as determined by binding of anticonnexin43 in comparison to conventional ultrastructural studies. Since no significant immunostaining was found in (i) the outer border of the circular muscle layer of the canine colon and (ii) the border layer between the submucosa and the circular muscle layer of human colon, where rare gap junctions have been identified at the ultrastructural level, we conclude that the electron-microscopic analysis is the more sensitive of the two methods.