LONG-TERM POTENTIATION OF NICOTINIC TRANSMISSION BY A HETEROSYNAPTIC MECHANISM IN THE STELLATE GANGLION OF THE CAT

In the anesthetized, scopolamine-treated cat, the compound action potential (CAP) evoked by a single supramaximal shock to the third thoracic white ramus (T3WR) was recorded in the inferior cardiac nerve (ICN). The CAP was depressed in a dose-dependent manner by the intravenous administration of the nicotinic antagonist hexamethonium (C6). During steady intravenous infusion of C6, which reduced the amplitude of the CAP by 80-90%, a short train of stimuli (few seconds, 10-40 Hz) to the sympathetic trunk just below T4WR potentiated the CAP for periods of tens of minutes to 1-2 h (heterosynaptic long-term potentiation, LTP). An LTP of similar time course was obtained when both train and single shock were applied to T3WR (homosynaptic LTP). Magnitude and duration of the heterosynaptic LTP were dependent on number, frequency, and intensity of the stimuli. No LTP was produced by a train to the ICN. Heterosynaptic LTP was also observed in the absence of C6. Because of the limited subliminal fringe of the test input under this condition, the LTP was of small magnitude. Heterosynaptic LTP also of the heart rate (HR) response to a test stimulus was observed after a conditioning train. The conditioning train produced a displacement to the right of the dose-response curve for C6. The intravenous dose of C6 required for 50% attenuation of the test CAP increased from 0.84 ± 0.15 (SE) mg/kg pretrain to 2.56 ± 0.46 mg/kg posttrain (n = 5, P < 0.01). The magnitude of the HR increase evoked by a bolus of acetylcholine (ACh) injected into the arterial supply of the stellate ganglion was enhanced by a train to the sympathetic trunk for a time comparable to the duration of LTP of the CAP. The heterosynaptic or homosynaptic LTP was not appreciably modified by the intravenous administration of the β-adrenergic antagonist propranolol, the α-adrenergic antagonist phentolamine, or the dopamine antagonist haloperidol. The conditioning input produced no potentiation, or a potentiation of small amplitude lasting only 1-2 min, after prolonged stimulation (e.g., 40 Hz for 2-3 h). At this time nicotinic transmission was unchanged. Only the axons that underwent prolonged stimulation lost the ability to produce LTP. For example, by the use of the heterosynaptic paradigm, after prolonged stimulation of the sympathetic trunk below T4WR, the LTP produced by the trunk (but not that produced by T3WR) was lost, whereas the converse occurred after prolonged stimulation of T3WR. After the LTP was lost, there was no recovery over a period of 4-8 h. These results show that a set of converging preganglionic axons produce LTP of the ganglion cell firing evoked, through a nicotinic mechanism, by another set of preganglionic axons. The data suggest that a postsynaptic mechanism is involved in the expression of this LTP. The pharmacology of LTP suggests that the inducer is neither ACh nor a catecholamine. The finding of loss of LTP after prolonged stimulation suggests that the inducer has limited capacity to maintain its stores, and this, in turn, suggests the possibility of a peptide.