Stimuli which entrain the circadian clock of the neonatal Syrian hamster in vivo regulate the phosphorylation of the transcription factor CREB in the suprachiasmatic nucleus in vitro.

Photic resetting of the adult mammalian circadian clock in vivo is associated with phosphorylation of the Ser133 residue of the calcium/cyclic AMP response-element binding-protein (CREB) in the retinorecipient region of the suprachiasmatic nucleus (SCN). Western blotting and immunocytochemistry were used to investigate whether agonists known to reset the clock of neonatal hamsters in vivo are also able to influence the phosphorylation of CREB in the suprachiasmatic hypothalamus in vitro. Antisera raised against synthetic CREB peptide sequences were used to differentiate between total CREB and the Ser133 phosphorylated form of CREB (pCREB). Western blot analysis of proteins isolated from suprachiasmatic tissue of 1-day-old Syrian hamsters revealed bands at approximately 45 kDa corresponding to total CREB and pCREB. Treatment of the tissue with a mixture of glutamatergic agonists [N-methyl-D-aspartate (NMDA), amino-methyl proprionic acid (AMPA) and kainate, all at 1 microM], or native glutamate (1 microM) had no effect on the total CREB signal, but increased the pCREB signal, indicative of agonist-stimulated phosphorylation of CREB on Ser133. A similar effect was seen following treatment of the suprachiasmatic blocks with either dopamine (1 microM) or forskolin (1 microM). Simultaneous treatment with melatonin (1 microM) significantly attenuated stimulation by forskolin. The effect of the agonists on nuclear pCREB-immunoreactivity (-ir) was investigated in primary cultures which contained a mixture of cell types characteristic of the suprachiasmatic nuclei in vivo. Basal expression of nuclear total CREB-ir was high, whereas expression of pCREB-ir was low. Treatment with glutamate (1 microM) or dopamine (1 microM) had no effect on total CREB-ir, but increased pCREB-ir in approximately 50 and 30% of cells, respectively, whereas forskolin (1 microM) increased pCREB-ir in almost all cells (> 90%). The effects of all three agonists were rapid (< 15 min), and dose and time dependent. Melatonin reversed the effects of forskolin in mixed cultures, but not in pure astrocyte cultures. Dual-immunocytochemistry (ICC) revealed that glutamate (1 microM) increased nuclear pCREB-ir in cells immunoreactive for microtubule-associated protein II (MAP II-ir), but not other cells, indicating an effect predominantly on neurons. This occurred equally in gamma-amino butyric acid (GABA)-ir and non-GABA-ir neurons. Dopamine (1 microM) was more selective, increasing pCREB-ir only in GABA-ir neurons, whereas forskolin increased pCREB-ir in all cells. The specific stimulation of pCREB-ir in GABA-ir neurons by dopamine was reversed by melatonin, but melatonin had no effect on the increase in pCREB-ir induced in GABA-ir neurons by glutamate. These results demonstrate that agonists known to entrain the circadian clock in vivo modulate phosphorylation of CREB in GABA-ir neurons derived from the neonatal suprachiasmatic nuclei.