Intracellular electrophysiological study of suprachiasmatic nucleus neurons in rodents: inhibitory synaptic mechanisms.

1. The mechanisms responsible for evoked and spontaneous fast inhibitory postsynaptic potentials (IPSPs) in the hypothalamic suprachiasmatic nucleus (SCN) were studied with intracellular recording. SCN neurons, primarily ones identified as receiving excitatory optic nerve input, were recorded in rat and guinea‐pig brain slice preparations maintained in vitro. 2. In normal medium, electrical stimulation of a site dorsocaudal to the SCN evoked IPSPs from thirty‐three of thirty‐six neurons. The evoked IPSPs rose to the peak quickly (8.7 +/‐ 0.9 ms, mean +/‐ S.E.M.; n = 15 neurons) and decayed gradually with a time constant of 25 +/‐ 3 ms. Spontaneous IPSPs were present in each of the thirty‐six neurons. These IPSPs had a rise‐to‐peak time of 7.2 +/‐ 1.0 ms (n = 6 neurons) and a decay time constant of 14 +/‐ 5 ms. 3. When recorded with potassium acetate‐filled electrodes, the evoked and spontaneous IPSPs were hyperpolarizing at resting membrane potential (less negative than ‐70 mV) and had a reversal potential of around ‐75 mV. On the other hand, when recorded with potassium chloride‐filled electrodes, the IPSPs were depolarizing at membrane potentials more negative than ‐50 mV and had an estimated reversal potential less negative than spike threshold. 4. Bath application of bicuculline (10‐50 microM), a gamma‐aminobutyric acidA (GABAA) receptor antagonist, resulted in a complete blockade of both the evoked (n = 16) and spontaneous (n = 13) IPSPs. The bicuculline effects were reversible, and were not associated with any significant and consistent change in baseline membrane potential or input resistance. The neurons impaled in bicuculline‐containing medium (n = 11) exhibited neither spontaneous IPSPs nor evoked IPSPs. 5. In some neurons bicuculline‐resistant hyperpolarizing potentials, which were similar to the fast IPSPs in time course, occurred spontaneously or were evoked by electrical stimulation of the optic nerve or the dorsocaudal site. A fast prepotential always preceded the hyperpolarizing potential, and hyperpolarizing currents blocked these events, indicating that they were not synaptic in origin. No slow IPSPs were detected. 6. The results suggest that fast IPSPs from non‐retinal afferents exist in virtually all SCN neurons receiving excitatory retinal input, and that GABAA receptors associated with Cl‐ channels mediate the fast IPSPs.