Synaptic input from the retina to the suprachiasmatic nucleus changes with the light-dark cycle in the Syrian hamster.

1. Single cell extracellular recordings were made from the suprachiasmatic nucleus (SCN) in urethane-anaesthetized Syrian hamsters at different times of the light-dark cycle. Peristimulus time histograms (PSTHs) were created following stimulation of the optic nerve. 2. Both short-latency (< 50 ms) and long-latency (> 50 ms) excitatory responses were seen. Almost all inhibitory responses had a short latency. 3. A total of 288 SCN neurones were recorded. Taking all types of response together, 55 (36.9%) of the 149 neurones tested in the dark period responded to optic nerve stimulation while only 23 (16.6%) of the 139 neurones tested in the light period responded. The difference between the proportion of all responsive and non-responsive neurones in the dark and light periods was highly significant (P < 0.01, Fisher's exact probability test). The difference in the proportion of excitatory responses was also significant (P < 0.01). 4. During the dark period, the mean spontaneous firing rate (5.00 +/- 0.88 spikes s-1; mean +/- S.E.M., n = 55) of the responsive cells was significantly higher than that of the non-responsive cells (2.65 +/- 0.33 spikes s-1; mean +/- S.E.M., n = 74; P < 0.01; Student's unpaired t test). 5. Injection of APV (20 mM, 2 microliters, I.C.V.; n = 6), an antagonist for the NMDA receptor, or CNQX (10 mM, 2 microliters, I.C.V.; n = 5), an antagonist of the non-NMDA receptor, significantly reduced the responses of all the neurones tested. 6. We conclude that there is daily variation in the firing of SCN neurones in vivo and the variation is restricted to those cells receiving optic nerve inputs. The change in the responsiveness of the SCN to optic nerve stimulation at different times of day suggests that there is a rapidly changing cycle of synaptic function in the SCN. The action of the antagonists suggests that the excitatory retinal projections to the SCN which show this variation are mediated by glutamate and that both NMDA and non-NMDA receptors are involved.