Using the whole-cell variation of the patch-clamp technique to record from mammalian NG108-15 cells, we have studied the ligand-gated ion channel current activated by a high concentration (100 microM) of local pressure-applied 5-hydroxytryptamine (5-HT). The response was induced at intervals of at least 90-120 s, which allowed the receptor to fully recover between activations. 2. The rapid inward current induced by pressure-applied 5-HT was reproducibly inhibited by the superfusion of low concentrations of 5-HT which evoked little or no detectable inward current alone (0.01-0.3 microM). This inhibitory effect was most likely to be due to a direct action on the 5-HT3 receptor as it could be recorded using intracellular solutions with or without adenosine triphosphate (ATP) and guanosine triphosphate (GTP). 3. The maximum inhibitory effect of a given concentration of 5-HT was not dependent on its superfusion time but on the number of activations of the receptor by pressure-applied 5-HT. This activation dependence was clearly evident, since the first inward current in the presence of 0.1 microM 5-HT was often unaffected in amplitude. 4. The inhibitory effect of 5-HT was evident at holding potentials of +60 and -60 mV; with the calcium chelator BAPTA in the recording pipette and with the nominal removal of extracellular calcium and magnesium ions. 5. The inhibitory effect was concentration dependent, with 50% inhibition of the inward current amplitude occurring at approximately 50 nM 5-HT. The slope factor of the inhibition curve was 1.3. The effect was mimicked by two other 5-HT3 receptor agonists, 2-methyl-5-HT and m-chlorophenylbiguanide (mCPBG) which gave 50% inhibition at approximately 600 nM and approximately 20 nM, respectively. These values are similar to the affinity values for these ligands determined in radioligand binding assays. 6. The 5-HT3 receptor "antagonists' (+)-tubocurarine and quipazine (both at 3 nM) reduced the inward current amplitude by approximately 50%. The rate of onset of the inhibitory effect of bath-applied 5-HT was slowed in the presence of (+)-tubocurarine but not in the presence of quipazine. This difference might be explained by the agonist properties seen only with quipazine. 7. The inhibition of the 5-HT3 receptor mediated inward current by low concentrations of bath-applied 5-HT3 receptor agonists is compatible with the cyclic model of receptor activation and desensitization. We conclude that we have been studying the high-affinity binding of agonists to the desensitized form of the 5-HT3 receptor.
