1 A series of isoquinolines have been identified as 5-HT3 receptor antagonists. One of these, RS 25259-197 [(3aS)-2-[(S)-1-azabicyclo[2.2.2]oct-3-yl]-2,3,3a,4,5,6-hexahydro-1-oxo-1H-benzo[de]isoquinoline-hydro- chloride], has two chiral centres. The remaining three enantiomers are denoted as RS 25259-198 (R,R), RS 25233-197 (S,R) and RS 25233-198 (R,S). 2 At 5-HT3 receptors mediating contraction of guinea-pig isolated ileum, RS 25259-197 antagonized contractile responses to 5-HT in an unsurmountable fashion and the apparent affinity (pK(B)), estimated at 10 nM, was 8.8 +/- 0.2. In this tissue, the - log K-B values for the other three enantiomers were 6.7 +/- 0.3 (R,R), 6.7 +/- 0.1 (S,R) and 7.4 +/- 0.1 (R,S), respectively. The apparent affinities of RS 25259-197 and RS 25259-198, RS 25233-197 and RS 25233-198 at 5-HT3 receptors in membranes from NG-108-15 cells were evaluated by a [H-3]-quipazine binding assay. The - log K-i values were 10.5 +/- 0.2, 8.4 +/- 0.1, 8.6 +/- 0.1 and 9.5 +/- 0.1, respectively, with Hill coefficients not significantly different from unity. Thus, at these 5-HT, receptors, the rank order of apparent affinities was (S,S) > (R,S) > (S,R) = (R,R). 3 RS 25259-197 displaced the binding of the selective 5-HT3 receptor ligand, [H-3]-RS 42358-197, in membranes from NG-108-15 cells, rat cerebral cortex, rabbit ileal myenteric plexus and guinea-pig ileal myenteric plexus, with affinity (pK(i)) values of 10.1 +/- 0.1, 10.2 +/- 0.1, 10.1 +/- 0.1 and 8.3 +/- 0.2, respectively. In contrast, it exhibited low affinity (pK(i) < 6.0) at 28 other receptors in binding assays, including adrenoceptors (alpha(1A), alpha(1B),alpha(2A), alpha(2B), beta(1), beta(2)), muscarinic (M(1)-M(4)),dopamine (D-1, D-2),opioid and other 5-HT (5-HT1A, 5-HT1D, 5-HT2C, 5-HT4) receptors. 4 RS 25259-197 was tritium labelled (specific activity: 70 Ci mmol(-1)) and evaluated in pharmacological studies. Saturation studies with [H-3]-RS25259-197 in membranes from NG-108-15 and cloned homomeric alpha subunits of the 5-HT3 receptor from N1E-115 cells expressed in human kidney 293E1 cells, revealed an equilibrium dissociation constant (K-d) of 0.05 +/- 0.02 and 0.07 +/- 0.01 nM, and B-max of 610 +/- 60 and 1068 +/- 88 fmol mg(-1), respectively. Competition studies in NG-108-15 cells indicated a pharmacological specificity entirely consistent with labelling a 5-HT3 receptor, i.e. RS 25259-197 > granisetron > (S)-zacopride > tropisetron > (R)-zacopride > ondansetron > MDL 72222. 5 In contrast to the majority of radioligands available to label 5-HT3 receptors, [H-3]-RS 25259-197 labelled a high affinity site in hippocampus from human post-mortem tissue with an equilibrium dissociation constant (K-d) of 0.15 +/- 0.07 nM and density (B-max) of 6.8 +/- 2.4 fmol mg(-1) protein. Competition studies in this tissue indicated a pharmacological specificity consistent with labelling of a 5-HT3 receptor. 6 Quantitative autoradiographic studies in rat brain indicated a differential distribution of 5-HT3 receptor sites by [H-3]-RS 25259-197. High densities of sites were seen in nuclear tractus solitaris and area postrema, a medium density in spinal trigeminal tract, ventral dentate gyrus and basal medial amygdala, and a low density of sites in hippocampal CA1, parietal cortex, medium raphe and cerebellum. 7 In conclusion, the functional, binding and distribution studies undertaken with the radiolabelled and non-radiolabelled RS 25259-197 (S,S enantiomer) established the profile of a highly potent and selective 5-HT3 receptor antagonist.