It was once thought possible that the brain clock, located in the suprachiasmatic nucleus (SCN) of the hypothalamus, could be understood as a homogeneous population of cells that produced a synchronous daily oscillatory signal. Instead, it is now clear that SCN subregions exhibit orderly phase dispersal. The mechanisms enforcing regional phase differences, however, are not well understood. Hong et al. (in press) propose that calcium contributes to synchronization through two mechanisms acting over different time scales and distances. Using all possible oscillating cell pairs as data points, the plot of temporal phase difference against pair separation distance suggests the coexistence of two modes of signaling: progressively propagating waves of a diffusing signal in adjacent cells, and phase-synchronizing neural networks acting at long range. In the first, a sharp wedge-shaped boundary in the region of small pair separation distances was inferred to represent a calcium wave sweeping through the SCN. The slope of this boundary represents the travel velocity of the wave, which, by itself, was calculated to be too slow to pass through the SCN in 24 h. A second mode of signaling was indicated by the finding that some cell pairs showed large spatial separations but nevertheless had small phase differences. For these cell pairs, the Fluorescence Resonance Energy Transfer signal was sufficiently bright to illuminate cell processes, revealing that anatomically joined cells oscillated in phase.
How does the fast, long-distance mechanism work? Ca2+(and/or other diffusing ions or molecules) could flow from one cell to another through gap junctions (Long et al., 2005), in addition to modulating the rate of neurotransmitter release. The slow Ca2+wave presumably indicates time-dependent release from Ca2+stores, with the usual long list of Ca2+-dependent metabotropic pathways, including gene activation, coming into play. The data of Long et al. (2005) are consistent with substantial evidence highlighting the importance of calcium and cAMP production acting through cAMP-dependent transcription factors upon which clock gene expression and SCN synchronization depend (O’Neill & Reddy, 2012). In the shell region of the SCN, there is an orderly daily sequence of high-amplitude oscillations, which begins in the dorsomedial region and encompasses serial activation of specific SCN subregions, followed by a silent interval (Yamaguchi et al., 2003; Foley et al., 2011). RGS16, a modulator of G protein signaling, which inactivates a negative regulator of cAMP production, is first expressed in the dorsomedial region (Doi et al., 2011). This mechanism appears to function synergistically with the activation of adenylate cyclase evoked by vasoactive intestinal polypeptide released from SCN core neurons (An et al., 2011). The fast signal depends, at least partially, on spiking activity, as tetrodoxin (TTX) dampens the oscillations (Welsh et al., 2010). A surprising datum of Hong et al. (in press) is that TTX induced oscillations in some cells that were formerly silent. This suggests that some spike-dependent inhibitory influence is removed by TTX, a problem for further investigation. What is also not yet clear is the mechanistic basis for the slow spread of signal from one region to the next. Hong et al. (in press) implicate a calcium wave and future work will show whether calcium is a primary player, or perhaps it shares the spotlight with other ions or small molecules mediating slow signal spread.