There is a good evidence for an input to the hypothalamic suprachiasmatic nucleus (SCN) from the retrochiasmatic area including the arcuate nucleus (ARCN), which may carry non-photic neuroendocrine cues to the SCN. The present study was undertaken to determine if melatonin influenced the spontaneous activity of cells in the SCN or their responses to stimulation of the ARCN. The experiments were carried out in accordance with the Animal (Scientific Procedures) Act, 1986. Animals were killed humanely. Conventional extracellular recordings were made from single units in the SCN in 500 µm hypothalamic slices from male Wistar rats. Responses of the SCN neurones to the ARCN stimulation were assessed by peristimulus time histograms. Twenty-seven of 47 SCN cells tested responded to single pulse stimulation of the ARCN. Both excitatory (n = 10) and inhibitory (n = 11) responses were observed. Additionally in some cases (n = 6) complex responses (excitation followed by inhibition or vice versa) were seen. These may reflect both excitatory and inhibitory projections. Typically the responses had a short latency (< 20 ms) after the stimulus pulse. Excitatory responses were usually short (▓le│ 25 ms) and inhibitory responses were longer (▓ge│ 40 ms).

Administration of melatonin (1 nM) to the bath solution changed neither the mean spike frequency of SCN cells nor mean log interspike interval (ISI). However, more sensitive indices of neuronal coding, e.g. the entropy of the ISIs (P < 0.05, paired t test) and the mutual information between adjacent ISIs (P < 0.02, Wilcoxon signed rank test; Bhumbra & Dyball, 2002) were significantly increased. The input to the SCN from the ARCN was significantly altered by melatonin in 19 of 47 experiments. Both excitatory (n = 5) and inhibitory (n = 6) responses were significantly reduced by addition of melatonin. In six cases excitatory (n = 3) and inhibitory (n = 3) responses became evident only in the presence of melatonin; in one case melatonin enhanced an inhibitory response and, in one additional case, an initial excitatory response became inhibitory. The effects of melatonin developed relatively slowly (> 15 min) and once established were not usually reversible.

The present results showed that melatonin at a very low concentration can modulate the input to the SCN from the ARCN. Melatonin also had significant effects on the spontaneous firing of SCN neurones that involved changes in ISI coding without significant changes in firing rate. Differences in responsiveness of individual SCN cells to the ARCN stimulation may reflect the heterogeneity of the neurones of the ARC but it is clear that melatonin can affect the behaviour of cells in the SCN and probably also daily endocrine rhythms.

This work was supported by a grant from the EPSRC.