The aim of the present study was to examine the effects of sarpogrelate, a 5-HT2 antagonist, on 5-HT-induced endothelium-dependent relaxation in isolated porcine coronary artery preincubated with ketanserin (3 x 10(-6) M) and precontracted by U 46619 (5 x 10(-9) M) and compare its effects with other 5-HT2 antagonists such as ritanserin and cyproheptadine. The investigation showed that sarpogrelate (10(-7) - 10(-5) M) had a weak antagonistic effect on 5-HT-induced relaxation and its effect was weaker than that of ritanserin (10(-9) -10(-7) M) and cyproheptadine (10(-8) -10(-6) M). The rank order of the antagonistic effects was: ritanserin > cyproheptadine > sarpogrelate. The study also showed that both sarpogrelate and ritanserin had no inhibitory effect on bradykinin-induced relaxation. In our previous study, we investigated the binding affinity of sarpogrelate, ritanserin and cyproheptadine to the 5-HT2A-receptor in rabbit cerebral cortex membranes and the pK(i) values found were 7.22, 8.98 and 7.54, respectively (M. Rashid et al., Jpn J Pharmacol 87, 189 - 194, 2001). Rank order of the calculated ratio of concentration of pA(2) or pD'(2) vs K-i was: sarpogrelate > ritanserin > cyproheptadine. Thus, these findings suggest that sarpogrelate has the lowest antagonistic effect on 5-HT-induced endothelium-dependent relaxation and the highest selectivity towards 5-HT2A receptor and might also be the safest drug with respect to its clinical implications in comparison with ritanserin and cyproheptadine.