The actions of agonists at α
2
-adrenoceptors were investigated on single cells of the submucous plexus of the guinea pig small intestine. Intracellular recordings were made from neurons
in vitro
, and noradrenaline and other agonists were applied by adding them to the superfusion solution. The actions of noradrenaline released from terminals of sympathetic nerves was also studied by stimulating the nerves and recording the inhibitory postsynaptic current; this current can be mimicked by brief applications of noradrenaline from a pipette tip positioned within 50 μm of the neuron. The α
2
-adrenoceptor-bound noradrenaline with an apparent dissociation constant of 15 μM, determined by the method of partial irreversible receptor inactivation: clonidine and 5-bromo-6-(2-imidazolin-2-ylamino)-quinoxaline (UK 14304) had dissociation constants of 36 nM and 2.5 μM respectively. Noradrenaline and UK 14304 caused maximal hyperpolarizations, or outward currents; clonidine was a full agonist in only 4 of 35 cells, a partial agonist in 25 cells, and without effect in 4 cells. Clonidine acted as a competitive antagonist of noradrenaline in those cells in which it lacked agonist action; its dissociation equilibrium constant determined by Schild analysis was about 20 nM. The potassium conductance increased by the α
2
-adrenoceptor agonists, whether they were applied exogenously or released by stimulation of presynaptic nerves, showed marked inward rectification. The neurons showed inward rectification also in the absence of agonist; both types of rectification were eliminated by rubidium (2 mM), barium (3–30 μM) and caesium (2 mM). When the recording electrodes contained the non-hydrolysable derivative of guanosine 5′-triphosphate (GTP), guanosine 5′-
O
-(3-thiotriphosphate, GTP-γ-S), the effects of applied α
2
-adrenoceptor agonists did not reverse when they were washed from the tissue, implying that GTP hydrolysis is necessary for the termination of agonist action. Pretreatment with pertussis toxin abolished the inhibitory synaptic potential (IPSP) and agonist-induced hyperpolarizations. Phorbol 12, 13-dibutyrate, forskolin, cholera toxin and sodium fluoride did not affect the responses to α
2
-adrenoceptor agonists. The synaptic hyperpolarization resulting from sympathetic nerve stimulation, or the hyperpolarization evoked by a brief (3–5 ms) application of noradrenaline, began after a latency of about 30 and 60 ms respectively. The decline of the synaptic current was exponential with time constant about 300 ms: when a high concentration of the antagonist idazoxan was applied suddenly (by applying pressure to a pipette tip positioned near the neuron), a steady-state hyperpolarization evoked by superfusion with noradrenaline was terminated with a similar time-course. This result suggests the decline of the synaptic response may be determined by the dissociation rate of noradrenaline; it is also possible that an intermediate biochemical process may underlie the decay of the synaptic potential. Many of the features of the response to noradrenaline are noted to be the same as for inhibitory synaptic potentials caused by acetylcholine acting on the cardiac type of M
2
muscarinic receptor. It is proposed that synaptically released noradrenaline binds to the α
2
receptor and brings about neuronal inhibition by activating a GTP binding protein within the membrane, which in turn leads to an increased opening of inwardly rectifying potassium channels.
The structural basis for agonist and partial agonist action on a β1-adrenergic receptor