Abstract
Ever since the locus of the brain clock in the suprachiasmatic nucleus (SCN) was first described, methods available have both enabled and encumbered our understanding of its nature at the level of the cell, the tissue, and the animal. A combination of in vitro and in vivo approaches has shown that the SCN is a complex heterogeneous neuronal network. The nucleus is composed of cells that are retinorecipient and reset by photic input; those that are reset by nonphotic inputs; slave oscillators that are rhythmic only in the presence of the retinohypothalamic tract; endogenously rhythmic cells, with diverse period, phase, and amplitude responses; and cells that do not oscillate, at least on some measures. Network aspects of SCN organization are currently being revealed, but mapping these properties onto cellular characteristics of electrical responses and patterns of gene expression are in early stages. While previous mathematical models focused on properties of uniform coupled oscillators, newer models of the SCN as a brain clock now incorporate oscillator and gated, nonoscillator elements.