Redefining the limits of day length responsiveness in a seasonal mammal.

At temperate latitudes, increases in day length in the spring promote the summer phenotype. In mammals, this long-day response is mediated by decreasing nightly duration of melatonin secretion by the pineal gland. This affects adenylate cyclase signal transduction and clock gene expression in melatonin-responsive cells in the pars tuberalis of the pituitary, which control seasonal prolactin secretion. To define the photoperiodic limits of the mammalian long day response, we transferred short day (8 h light per 24 h) acclimated Soay sheep to various longer photoperiods, simulating those occurring from spring to summer in their northerly habitat (57 degrees N). Locomotor activity and plasma melatonin rhythms remained synchronized to the light-dark cycle in all photoperiods. Surprisingly, transfer to 16-h light/day had a greater effect on prolactin secretion and oestrus activity than shorter (12 h) or longer (20 and 22 h) photoperiods. The 16-h photoperiod also had the largest effect on expression of circadian (per1) and neuroendocrine output (betaTSH) genes in the pars tuberalis and on kisspeptin gene expression in the arcuate nucleus of the hypothalamus, which modulates reproductive activity. This critical photoperiodic window of responsiveness to long days in mammals is predicted by a model wherein adenylate cyclase sensitization and clock gene phasing effects of melatonin combine to control neuroendocrine output. This adaptive mechanism may be related to the latitude of origin and the timing of the seasonal transitions.

Gonadotrophin-releasing hormone drives melatonin receptor down-regulation in the developing pituitary gland.

Melatonin is produced nocturnally by the pineal gland and is a neurochemical representation of time. It regulates neuroendocrine target tissues through G-protein-coupled receptors, of which MT(1) is the predominant subtype. These receptors are transiently expressed in several fetal and neonatal tissues, suggesting distinct roles for melatonin in development and that specific developmental cues define time windows for melatonin sensitivity. We have investigated MT(1) gene expression in the rat pituitary gland. MT(1) mRNA is confined to the pars tuberalis region of the adult pituitary, but in neonates extends into the ventral pars distalis and colocalizes with luteinizing hormone beta-subunit (LH beta) expression. This accounts for the well documented transient sensitivity of rat gonadotrophs to melatonin in the neonatal period. Analysis of an upstream fragment of the rat MT(1) gene revealed multiple putative response elements for the transcription factor pituitary homeobox-1 (Pitx-1), which is expressed in the anterior pituitary from Rathke's pouch formation. A Pitx-1 expression vector potently stimulated expression of both MT(1)-luciferase and LH beta-luciferase reporter constructs in COS-7 cells. Interestingly, transcription factors that synergize with Pitx-1 to trans-activate gonadotroph-associated genes did not potentiate Pitx-1-induced MT(1)-luciferase activity. Moreover, the transcription factor, early growth response factor-1, which is induced by gonadotrophin-releasing hormone (GnRH) and trans-activates LH beta expression, attenuated Pitx-1-induced MT(1)-luciferase activity. Finally, pituitary MT(1) gene expression was 4-fold higher in hypogonadal (hpg) mice, which do not synthesize GnRH, than in their wild-type littermates. These data suggest that establishment of a mature hypothalamic GnRH input drives the postnatal decline in pituitary MT(1) gene expression.