Kanagawa hemolysin (KH), an exotoxin produced from Kanagawa phenomenon-positive Vibrio parahemolyticus, has been shown to possess various biological activities including hemolysis, enterotoxicity, cytotoxicity, and cardiotoxicity. The aim of this study was to investigate the effect of KH on the cardiovascular system and its mechanism, employing in vivo and in vitro experiments of the rat. Intracerebroventricular (icv) administration of 100 mHU KH produced a marked and continuous pressor effect (icv KH-pressor effect), and the icv pressor effect was not repeatable. However, intravenous (iv) injection of the same dose of KH induced a prominent depressor effect (iv KH-depressor effect). The icv KH-pressor effect was inhibited by acid-denaturation, while the iv KH-depressor effect was not. Simultaneous icv administration of the three agents (ouabain, diltiazem, or bumetanide: 10μg/kg each) significantly reduced the pressor effect. The icv KH-pressor effect was inhibited by treatment with iv phentolamine or chlorisondamine, but was not affected by iv candesartan. The iv KH-depressor effect was repeatable and was attenuated by treatment with iv NAME or methylene blue. In vitro experiments using isolated thoracic aorta, 10⁣6 M phenylephrine (PE) and 50 mM KCl produced a sustained contraction. In rings contracted with either agents, KH showed relaxant responses in a concentration- dependent fashion and the relaxation (KH-vasorelaxation) was not dependent on the existence of the endothelium. The KH-vasorelaxation in the endothelium-intact rings contracted by PE was abolished by methylene blue treatment. In summary, the present findings suggest that in the icv KH-pressor effect the cation leak-inducing action of KH is implicated, which leads to the increased central sympathetic tone, that the iv KH-depressor effect results from the vasorelaxation via NO-guanylate cyclase system, and that the KH-vasorelaxation is independent of the endothelium and the guanylate cyclase system is involved in it. In conclusion, the mechanism of KH producing the icv pressor effect may not be identical to that of KH producing the iv depressor effect.