To determine the distribution of gap junction protein connexin 43 (Cx43) and the expression of Cx43 mRNA in different regions of canine small intestine and colon, modified Western blot and Northern blot techniques were used with a site-directed antibody raised against a synthetic peptide of Cx43 Anti-Cx43 (252-271) and a probe of 1.3-kb Cx43 cDNA. Equal amounts of plasma membrane enriched fraction (10 mug protein) from small intestinal and colonic circular muscle, longitudinal muscle, and colonic submucous plexus border of circular muscle (interstitial cells of Cajal rich layer, ICC) of the dog were resolved by 12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis, transferred onto nitrocellulose, and blotted by chemiluminescent assay. Different yet characteristic ratios of the immunoreactive gap junction proteins located at 43 and 40 kDa were found in membranes from small intestinal and colonic circular smooth muscle, and ICC. These results suggest that gap junction Cx43 in dog heart, dog small intestinal and colonic circular muscle, and ICC is homologous to rat heart. Furthermore, with the use of Northern blot hybridization with a 1.3-kb Cx43 cDNA, a 3.0-kb message was observed in small intestinal and colonic circular muscles, longitudinal muscles, and ICC. However, the mRNA signal of small intestinal circular muscle was the strongest and that of longitudinal muscle was weakest, especially from colon longitudinal muscle. This study provided an independent and molecular basis for our previous observation that with the use of electron microscopy visible gap junctions were found in large numbers only in canine intestinal circular muscle and colonic ICC but were rare in colonic circular muscle and were nearly absent from longitudinal muscle. Colon longitudinal muscle has so far not been demonstrated by any technique to have gap junctions; this raises the possibility that it may be poorly coupled because it lacks gap junctions. This study also indicated that Cx43 mRNA was observed in intestinal longitudinal muscle, which has few or no visible gap junctions. These data suggest that small gap junctions invisible by electron microscopy may be present.