1. The radioligand binding characteristics of the 3H-derivative of the novel 5-HT3 receptor antagonist BRL46470 were investigated and directly compared to the well characterized 5-HT3 receptor radioligand [3H]-granisetron, in tissue homogenates prepared from rat cerebral cortex/hippocampus, rat ileum, NG108-15 cells, HEK-5-HT3As cells and human putamen. 2. In rat cerebral cortex/hippocampus, rat ileum, NG108-15 cell and HEK-5-HT3As cell homogenates, [3H]-BRL46470 bound with high affinity (Kd (nM): 1.57 +/- 0.18, 2.49 +/- 0.30, 1.84 +/- 0.27, 3.46 +/- 0.36, respectively; mean +/- s.e. mean, n = 3-4) to an apparently homogeneous saturable population of sites (Bmax (fmol mg-1 protein): 102 +/- 16, 44 +/- 4, 968 +/- 32 and 2055 +/- 105, respectively; mean +/- s.e. mean, n = 3-4) but failed to display specific binding in human putamen homogenates. 3. In the same homogenates of rat cerebral cortex/hippocampus, rat ileum, NG108-15 cells, HEK-5-HT3As cells and human putamen as used for the [3H]-BRL46470 studies, [3H]-granisetron also bound with high affinity (Kd (nM): 1.55 +/- 0.61, 2.31 +/- 0.44, 1.89 +/- 0.36, 2.03 +/- 0.42 and 6.46 +/- 2.58 respectively; mean +/- s.e. mean, n = 3-4) to an apparently homogeneous saturable population of sites (Bmax (fmol mg-1 protein): 39 +/- 4, 20 +/- 2, 521 +/- 47, 870 +/- 69 and 18 +/- 2, respectively; mean +/- s.e. mean, n = 3-4). 4. Competition studies with a range of structurally different 5-HT3 receptor ligands indicated that in both rat cerebral cortex/hippocampus and rat ileum homogenates, [3H]-BRL46470 binding exhibited a pharmacological profile consistent with the labelling the 5-HT3 receptor with compounds competing with Hill coefficients close to unity.5 In HEK-5-HT3As cell homogenates, [3H]-BRL46470 and [3H]-granisetron associated rapidly((3.84+/-0.4)106 M-1S-1 and (5.85+/-0.2)106 M-1S-1, respectively, mean+/-s.e.mean, n=3-4) in an apparently monophasic manner. Following the establishment of equilibrium, both [3H]-BRL46470 and [3H]-granisetron at a saturating concentration ([3H]-BRL46470 approximately 16 nM; [3H]-granisetron approximately 18 nM) and at a sub-Kd concentration (approximately 1 nm for both radioligands)dissociated biphasically in HEK-5-HT3As cell homogenates (saturating concentration; [3H]-BRL464704.05 x 10-3+/-2.53 x I0-3 s-1 and 5.83 x 10-5+0.91 x I0-5 s-1; [3H]-granisetron 3.20 x 10-3+ 1.70 x IO-3 s-1 and18.58 x 10-5 +/- 4.19 x I0-5 s-1: sub-Kd concentration; [3H]-BRL46470 2.47 x 10-3+/- 1.18 x 10-3 s-1 and 9.30x 10-5+/-2.59x 10-5 S-1; [3H]-granisetron 65.91 x 10-3+/-22.14x I0-3 s-1 and 49.96x 10-5+/-12.26x 10-5s- 1 mean+/- s.e.mean, n = 4-8) when induced by a 300 fold dilution in ice-cold Tris/Krebs.6 In conclusion, the present study provides evidence that [3H]-BRL46470 specifically labels the 5-HT3receptor in rat cerebral cortex/hippocampus, rat ileum, NG108-15 cell and HEK-5-HT3As cell homogenates, but fails to label the 5-HT3 receptor expressed in human putamen. Whilst the pharmacological profile of the site labelled by [3H]-BRL46470 is directly comparable to that labelled by [3H]-granisetron, [3H]-BRL46470 consistently labelled approximately twice the density of sites compared to [3H]-granisetron in the same tissue homogenates prepared from rat cortex/hippocampus, ratileum, NG108-15 cells and HEK-5-HT3As cells.