The suprachiasmatic nuclei (SCN) of the hypothalamus are widely regarded as the part of the brain that generates circadian rhythms in mammals. Secretion of the pineal hormone melatonin is one of a number of neuroendocrine events that show a daily rhythm (Perreau-Lenz et al. 2003), and melatonin in turn affects the circadian clock by mechanisms that are unclear (Pevet et al. 2002). It has recently been demonstrated using single-unit recordings from the SCN in intact animals that SCN cells show a daily rhythm of mean spike frequency (Saeb-Parsy & Dyball, 2003). The present study was undertaken to extend the earlier study and to explore the role played by melatonin in the generation of rhythms in the SCN by making recordings from pinealectomised rats.

Using the ventral surgical approach, single-cell recordings were made in anaesthetised (urethane, 1.2 g/kg I.P.) rats in vivo between 3 and 10 weeks after pinealectomy under isofluorane. The rats were killed humanely at the end of the experiments.

In pinealectomised rats, recordings were made from 192 SCN cells at different times of the day night cycle and the activity of each of them, recorded for a minimum of 10 min, was recorded. Activity was expressed both in terms of mean spike frequency and in terms of the entropy of the log interspike interval distribution. The entropy of the log interspike interval distribution quantifies the variability in the interval distribution and thus gives a measure of the coding capacity of each spike. Entropy was plotted against the respective Zeitgeber time. In intact animals such a plot showed an oscillation that could be described using a second order Fourier expansion with a peak in the mid-dark period. Monte-Carlo significance testing showed a significant rhythm in intact animals (Bhumbra et al. 2003). In pinealectomised animals, however, the maximum amplitude (0.4092 bits) of the curve describing the SCN cell activity was below the amplitude found after fitting the corresponding curves to randomised data (maximum amplitude after 10,000 randomisations, 0.8139 bits), and thus the oscillation seen in real data was not statistically different from a flat line (P < 0.001, Monte-Carlo significance testing). Therefore, after pinealectomy no rhythm in the activity of the SCN cells with the dark/light period was detected. Mean spike frequency showed a similar lack of change in pinealectomised animals. It thus appears that at least some aspects of the rhythmic activity of the SCN do not persist after pinealectomy.

This work was supported by St John’s College, Cambridge.