The vasoactive properties of the neurokinins (substance P (SP), neurokinin A (NKA), neurokinin B (NKB)) and some selective analogues were assessed in the arterial and venous mesenteric beds of the rat. Although both sides of the mesenteric vasculature displayed endothelium-dependent relaxation in response to acetylcholine (ACh) or bradykinin (BK) (1 and 10 nmol), SP and the selective NK-1 analogue, [Sar9,Met(O2)11]SP were inactive. Of the three selective neurokinin agonists used, [Sar9,Met(O2)11]SP (NK-1), [beta-Ala8]NKA-(4-10) (NK-2) and [MePhe7]NKB (NK-3), only the latter induced a dose-dependent pressor effect in the venous mesenteric vasculature. Injections of SP and the selective NK-1 and NK-2 analogues at high doses (10 nmol), did not change the perfusion pressure in the mesenteric bed even when the mesenteric vasculature was treated with methylene blue (50-mu-AM) to inhibit the effects of endothelium-derived relaxing factor (EDRF) or with N(G)-nitro-L-arginine (L-NNA) (20-mu-M) to inhibit the formation of EDRF or with 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate] (CHAPS 20 mM, 30 s) to remove the endothelial layer. In contrast, the vasoconstrictor effects of noradrenaline (NA), angiotensin II (AT(II)), NKB and [MePhe7]NKB on the venous side of the circulation were enhanced following treatment with L-NNA, methylene blue or CHAPS. The present results suggest that neurokinins act on the rat mesenteric bed by increasing the perfusion pressure of the venous vasculature via activation of NK-3 receptors. Neurokinins are inactive on the arterial mesenteric vasculature. The endothelial layer of the venous vasculature may interfere by acting as a physical and physiological barrier (basal release of EDRF) to the pressor response induced by NA, AT(II) NKB and [MePhe7]NKB.