graphic

Synchronization (top) of mPer1 transcription in the SCN is lost when neuronal activity is blocked (bottom).

Okamura/AAAS

The brain's clock is actually an assembly of individually cycling cells, according to results from Shun Yamaguchi, Hitoshi Okamura (Kobe University, Kobe, Japan), and colleagues. But the discrete clocks can work as one because they are synchronized by electrical impulses.

Our internal clock, which controls circadian behaviors and physiology, is a circuitry of many thousands of neurons in the brain called the suprachiasmatic nucleus (SCN). The Kobe group looked at communication within this circuitry in cultured brain slices of transgenic mice using a fluorescent reporter of transcription of a central clock gene, Period (mPer1). They found that mPer1 transcription cycled in nearly every neuron in the SCN. Transcription occurs independently in each cell, and yet in all cells the transcriptional peaks were synchronized: mPer1 levels were high in the day and low at night. The only variation was a dorsal-to-ventral wave, with dorsal regions peaking a few hours earlier. The function of the wave is unclear but may relate to the fact that portions of the SCN talk to different areas of brain.

The synchronized gene oscillation is controlled by neuronal activity. Blocking Na+-dependent action potentials desynchronized the SCN, thus producing some cells with day and some with night mPer1 peaks. But synchronization reasserted itself automatically in the neurons—resumption of action potentials returned the organized oscillations to the SCN.

The results mean that the SCN must be reading out as a sum of its parts to “announce circadian time to the rest of the organism,” says Okamura. In the future, he would like to define the circuits that mediate both coordination within the SCN and communication with peripheral clocks. ▪

Reference:

Yamaguchi, S., et al.
2003
.
Science.
302
:
1408
–1412.