1 A combined study of receptor binding in central neuronal cell membranes and functional responses in isolated segments of guinea-pig small intestine allowed characterization of the interaction of four antidepressant drugs with central and peripheral 5-HT3 and 5-HT4 receptors. 3 In whole ileal segments, concentration-response curves to 5-HT were biphasic, with the high-and low-potency phases involving 5-HT4 and 5-HT3 receptors, respectively. Curves to 2-methyl-5-hydroxytryptamine (2-methyl-5-HT: a 5-HT3 receptor agonist) and 5-methoxytryptamine (5-MeOT: a 5-HT4 receptor agonist) were monophasic. All antidepressants were used at concentrations lacking anticholinoceptor properties, as demonstrated in both electrically stimulated longitudinal musclemyenteric plexus preparations (LMMPs) and in unstimulated LMMPs following addition of acetylcholine (100 nM). 4 Fluoxetine (0.1-1 JM) and litoxetine (0.3-3JM) antagonized both the high-and low-potency phases of the 5-HT curve. Schild analysis for the low-potency phase yielded pA2 estimates of 6.6 ± 0.3 (Schild slope of 1.1) and of 6.6 ± 0.1 (Schild slope of 1.1), respectively. At higher concentrations (3 JM), fluoxetine markedly inhibited the 5-HT response maximum. Clomipramine (10-300 nM) inhibited, by a mechanism independent of concentration, both phases of the 5-HT curve with a reduction of the maximum response. Paroxetine (1 JM) was ineffective on the high-potency phase, but caused a rightward shift of the low-potency phase (pKB: 6.1 ± 0.01).5 Responses to 2-methyl-5-HT were inhibited by 1 JAM fluoxetine (pKB: 5.4 ± 0.02). Like clomipramine (30 and 100 nM), litoxetine (1 and 3 JAM) produced rightward displacements of 2-methyl-5-HT-induced contractions, which were virtually independent of antidepressant concentration (pKB values: 6.0 ± 0.02 and 5.5 + 0.01, respectively). At higher concentrations, fluoxetine (3 JM) and clomipramine (300 nM) markedly reduced the 2-methyl-5-HT response maximum. Paroxetine (1 JAM) was ineffective.6 Responses to 5-MeOT were shifted to the right by fluoxetine (0.1-1 JAM) and litoxetine (1 and 3fJM) in a concentration-dependent manner. At higher concentrations, fluoxetine (3 JlM) markedly reduced the 5-MeOT response maximum, an effect also observed with 100 and 300 nM clomipramine. Paroxetine (1 JM) was ineffective.7 In unstimulated LMMPs, the excitatory effects evoked by 5-HT, 2-methyl-5-HT and 5-MeOT and the antagonism produced by 300 nM clomipramine were comparable to those obtained in whole ileal segments. This suggests that 5-HT contained in the mucosa of whole preparations does not interfere with agonist-induced contractile responses and with the inhibitory effect of antidepressant drugs.