Recombinant nicotinic receptors, expressed in Xenopus oocytes, do not resemble native rat sympathetic ganglion receptors in single-channel behaviour.

1. In order to establish the subunit composition of neuronal nicotinic receptors in rat superior cervical ganglia (SCG), their single-channel properties were compared with those of recombinant receptors expressed in Xenopus oocytes, using outside-out excised patch recording. 2. The mean main conductance of SCG channels from adult and 1-day-old rats was 34.8 and 36.6 pS, respectively. Less frequent openings to lower conductances occurred both as isolated bursts and as events connected to the main level by direct transitions. There was considerable interpatch variability in the values of the lower conductances. 3. Nicotinic receptors from oocytes expressing alpha3beta4 and alpha4beta4 subunits had chord conductances lower than that of SCG neurones (22 pS for alpha3beta4 and 29 pS for alpha4beta4). 4. Prolonged recording from both native and recombinant channels was precluded by 'run-down', i.e. channel activity could be elicited for only a few minutes after excision. Nevertheless, SCG channel openings were clearly seen to occur as short bursts (slowest component, 38 ms), whereas recombinant channels opened in very prolonged bursts of activity, the major component being the slowest (480 ms). 5. Addition of the alpha5 subunit to the alpha3beta4 pair produced channels with a higher conductance than those observed after injection of the pair alone (24.9 vs. 22 pS), suggesting incorporation of alpha5 into the channel. Addition of the beta2 subunit did not change alpha3beta4 single-channel properties. In one out of fourteen alpha3alpha5beta4 patches, both ganglion-like, high conductance, short burst openings and recombinant-type, low conductance, slow burst openings were observed. 6. Channels produced by expression in Xenopus oocytes of neuronal nicotinic subunits present in rat SCG as a rule differ from native ganglion receptors in single-channel conductance and gross kinetics. While it is possible that an essential nicotinic subunit remains to be cloned, it is perhaps more likely that oocytes either cannot assemble neuronal nicotinic subunits efficiently into channels with the correct composition and stoichiometry, or that they produce post-translational channel modifications which differ from those of mammalian neurones.