1. The aim of this study was to examine the possible role of the release of guanosine 3':5'-cyclic monophosphate (cyclic GMP) into the extracellular space in the regulation of rat aortic cyclic GMP content. 2. Rat aortic segments incubated in physiological solution released cyclic GMP into the medium in a time-dependent manner. This release was greatly enhanced when intact instead of tissues without endothelium were used. After 120 min of observation, a maximal 33 fold difference in extracellular cyclic GMP content was detected. 3. Treatment of rat aortic preparations with either a Ca2+-free solution or methylene blue, both conditions known to inhibit endothelium-derived relaxing factor (EDRF)-mediated responses, markedly reduced the extracellular accumulation of cyclic GMP from tissues with but not without endothelium. 4. Endothelium-dependent vasodilators such as acetylcholine (10 microM) and carbachol (10 microM) greatly increased tissue cyclic GMP content, in a time-dependent manner in rat aortic preparations with endothelium, but only slightly in tissues without. Maximal increases in intact tissues were obtained after about 1 min of agonist contact and amounted to about 35 and 15 fold respectively, thereafter tissue cyclic GMP content rapidly declined. Histamine (10 microM) elicited only minor effects on tissue cyclic GMP content of both intact preparations and those without endothelium. 5. Acetylcholine (10 microM), carbachol (10 microM) and histamine (10 microM) stimulated a time-dependent release of the cyclic nucleotide into the incubation medium from tissues with endothelium. After 120 min of observation, extracellular accumulation of cyclic GMP from intact tissues was increased by about 2.6, 6.6 and 1.7 fold respectively. Carbachol and histamine induced only minor effects on release from tissues without endothelium. 6. Sodium nitroprusside (0.3 and 10 microM), a direct activator of soluble guanylate cyclase, induced a concentration-dependent accumulation of cyclic GMP in tissues with and without endothelium that was associated with a concentration-dependent accumulation of cyclic GMP in the extracellular space. Peak tissue cyclic GMP content reached similar levels in preparations with and without endothelium, while extracellular cyclic GMP levels were about two times greater when experiments were performed with intact compared to endothelium-denuded tissues. 7. Atriopeptin II, an activator of particulate guanylate cyclase, increased tissue cyclic GMP content by about 8 and 18 fold respectively in tissues with and without endothelium. As was the case with sodium nitroprusside, atriopeptin II-stimulated release was markedly enhanced from intact tissues compared with those without endothelium. After 120 min of observation, there was a 16 fold difference in the amount of extracellular cyclic GMP.