Neuropeptide-E-1 Antagonizes the Action of Melanin-Concentrating Hormone on Stress-Induced Release of Adrenocorticotropin in the Rat

The physiological role of melanin-concentrating hormone (MCH) in mammals is still very elusive, but this peptide might participate in the central control of the hypothalamopituitary adrenal (HPA) axis during adaptation to stress. Cloning and sequencing of the rat MCH (rMCH) cDNA revealed the existence of additional peptides encoded into the MCH precursor. Among these peptides, neuropeptide (N) glutamic acid (E) isoleucine (I) arnide (NEI) is co-processed and secreted with MCH in rat hypothalamus. In the present work we examined: (1) The pattern of rMCH mRNA expression during the light and dark conditions in the rat hypothalamus and (2) The effect of intracerebroventricular (ICV) injections of rMCH and NEI in the control of basal or ether stress-modified release of corticotropin (ACTH), prolactin (PRL) and growth hormone (GH) secretion in vivo in light-on and light-off conditions. Our data indicate that rMCH mRNA levels do not change during the light-on period, but increase after the onset of darkness. Either alone or co-administered, rMCH and NEI do not modify basal secretion of GH and PRL at any time tested nor do they alter ether stress-induced changes in these two hormonal secretions. At the end of the light on period corresponding to the peak of the circadian rhythm in ACTH, administration of rMCH but not NEI leads to a decrease in ACTH levels while MCH is not effective during the light off period of the cycle (i.e. when basal ACTH levels are already low). Using a moderate ether induced stress, ACTH levels are only stimulated during the dark phase of the cycle. rMCH (63 or 210 nmoles) prevents the rise in ACTH release while NEI alone does not modify the stress response. Co-administration of both peptides before stress results in an abolition of the rMCH induced inhibition of ACTH plasma levels. Taken together, these data indicate that rMCH may act as a central corticotropin inhibitory factor involved in the circadian rhythmicity of plasma ACTH levels and that NEI antagonizes its action.     

Exposure to long summer days affects the human melatonin and cortisol rhythms

Exposure of 8 human subjects in summer to a natural 16 h bright light photoperiod phase advanced the morning salivary melatonin decline and cortisol rise and shortened the nocturnal melatonin signal by 2 h relative to the winter patterns of the same subjects followed under a combined artificial and natural light 16 h photoperiod. The data suggest that summer days experienced from sunrise till sunset and not winter days with a combined artificial and natural light long photoperiod evoke a true long day response of the human circadian system.     

The Photoperiodic Response in Syrian Hamster Depends upon a Melatonin-Driven Circadian Rhythm of Sensitivity to Melatonin

The pineal gland, via the daily pattern of melatonin (MEL) secretion, is directly involved in the conduction of photoperiodic information. The duration of MEL secretion is proportional to the duration of the dark period and, whatever the photoperiod is, MEL synthesis occurs 3 or 4 h after the dark onset in Syrian hamsters. In order to determine the relative importance of the duration or the coincidence hypothesis, a daily infusion protocol was used in sexually active pinealectomized hamsters. Long duration of MEL infusion (10 h) completely inhibit testes whereas short duration infusion (5 h) had no effect. When the animals were infused twice within 2 h 30 min separated by 3 h, they presented a complete gonadal atrophy, similar to the one observed with the 10 h infusion. Measurement of plasma MEL during the infusion and seperation periods revealed that MEL reached physiological nighttime values during the infusion period and fell to daytime values 1 h after the end of an infusion period. Thus, the results could not be due to a time additive action of the two MEL pulses. An intermediate response was observed when the 2 signals were applied across the light/dark transition. Gonadal regression did not occur when the 2 periods of infusion were separated by 5 h 30 min. The efficiency of this type of infusion was not dependent on the ambiant photoperiod since similar results were obtained in long and short photoperiods. The infusion was also as effective during the day as well as during the night. These results suggest that there is a rhythm of sensitivity to MEL, based on the coincidence hypotheses, that are important for transmission of photoperiodic information. This rhythm of sensitivity to MEL seems to be entrained by MEL itself, since the efficiency of the two pulses of MEL is not dependent of time of application and/or of photoperiod.     

Using Per gene expression to search for photoperiodic oscillators in the hamster suprachiasmatic nucleus

The circadian pacemaker in the suprachiasmatic nucleus (SCN) is also believed to underlie photoperiodic (seasonal) timekeeping in mammals. This clock has been modeled as a complex pacemaker composed of two coupled circadian oscillators; variability in their mutual phase relationship could account for the ability to measure daylength, with putative morning and evening oscillators synchronized to dawn and dusk, respectively. Recently, several genes have been identified that are believed to be part of the clock's core oscillatory mechanism. Here, we investigate how such molecular oscillations are altered as a function of photoperiod by analyzing Period (Per1, Per2, and Per3) gene expression at the mRNA level using SCN tissue sections and in situ hybridization. Golden hamsters were entrained to complete 24-h light-dark (LD) cycles with either a long (16 h) or a short (8 h) photophase, or they were entrained to the long complete photoperiod and then allowed to free-run in constant darkness. The results show large photoperiod-dependent changes in the duration of high daytime SCN Per1 and Per2 mRNA levels and small changes in the phase difference between their rhythms.     

Intracerebroventricular administration of melanin-concentrating hormone suppresses pulsatile luteinizing hormone release in the female rat

Melanin-concentrating hormone (MCH) has been reported to be involved in the regulation of feeding behaviour in rats and mice. Because many neuropeptides that influence ingestive behaviour also regulate reproductive function, the present study was designed to determine if central administration of MCH changes pulsatile secretion of luteinizing hormone (LH) in the rats. Wistar-Imamichi strain female rats were ovariectomized and implanted with oestradiol to produce a moderate inhibitory feedback effect on LH release. The effects of i. c.v. injections of MCH on LH release were examined in freely moving animals. Blood samples were collected every 6 min for 3 h through an indwelling cannula. After 1 h of sampling, MCH (0.1, 1 or 10 microg/animal) or vehicle (saline) was injected into the third cerebroventricle. Because MCH is also reported to affect the hypothalamo-pituitary-adrenal (HPA) axis, which in turn, can influence reproductive function, plasma corticosterone concentrations were determined in the same animals at 30-min intervals during the first and last hours and every 12 min during the second hour of the 3-h sampling period. When expressed as per cent changes, mean plasma LH concentrations after MCH administration were significantly lower in the animals injected with all doses of the peptide compared with vehicle-treated animals; LH pulse frequency was significantly lowered by 1 microg of MCH. Per cent changes in mean plasma corticosterone levels were not significantly affected by MCH administration. These results in oestradiol-treated ovariectomized rats indicate that central MCH is capable of inhibiting pulsatile LH secretion. We have previously shown that 48-h fasting suppresses pulsatile LH release in the presence of oestrogen. Take together, these results raise the possibility that MCH could play a role in mediating the suppression of LH secretion during periods of reduced nutrition.     

The circadian rhythm of Per1 gene product in the rat suprachiasmatic nucleus and its modulation by seasonal changes in daylength

The suprachiasmatic nucleus (SCN) of rats maintained under a 12-h light, 12-h dark cycle (LD12:12) as well as of those released into darkness exhibited the rhythm of a clock gene Per1 product, PER1 protein, with the maximum late in the subjective day and early night and minimum in the morning. The rhythm was phase delayed by 6-8 h compared with the reported rhythm of Per1 mRNA in the rat SCN [L. Yan et al. Neuroscience 94 (1999) 141]. Under a long, LD16:8, artificial photoperiod, the interval of elevated PER1-immunoreactivity was at least 4 h longer than that under a short, LD 8:16 photoperiod, due mainly to an earlier PER1 day-time rise under the long photoperiod. Under a natural photoperiod, profiles of the PER1 rhythm in summer and in winter resembled those under corresponding artificial photoperiods; therefore, twilight did not affect the rhythm in a substantial way. Under all photoperiods, when PER1 immunoreactivity was elevated, immunopositive cells were localized in the dorsomedial rather than in the ventrolateral part of the SCN. As the Per1 gene is a part of a molecular clockwork and as the rhythm of its product is modulated by the photoperiod, it appears that the whole molecular clockwork in the rat SCN is photoperiod-dependent and thus shaped by the season of the year.     

Pituitary Hormone Gene Expression in Male Golden Hamsters: Interactions Between Photoperiod and Testosterone

Daylength regulates neuroendocrine function in male golden hamsters. Exposure to short days triggers gonadal regression and decreases serum luteinizing hormone (LH), prolactin and testosterone concentrations. Inhibitory photoperiods also amplify the negative feedback actions of androgens upon gonadotropin secretion. To examine whether these changes arise from altered adenohypophyseal gene expression, we measured the abundance of the messenger ribonucleic acids (mRNAs) encoding β-LH, prolactin and proopiomelanocortin in anterior pituitaries of male golden hamsters which were either left intact, castrated, castrated and implanted with testosterone, or pinealectomized and maintained in either long (14 h light/10 h dark) or short (5 h light/19 h dark) days. Short days caused testicular atrophy in intact male hamsters and reduced serum LH in intact and castrated, testosterone-replaced hamsters. The relative abundance of β-LH mRNA was suppressed by exposure to short days only in castrated hamsters. Serum prolactin was decreased by short days regardless of circulating testosterone concentrations. Prolactin mRNA abundance was decreased by short days in all pineal-intact groups. Castration reduced proopiomelanocortin mRNA abundance in long days and testosterone replacement reversed this effect. In the presence of testosterone, photoperiod influenced serum LH concentrations without altering hypophyseal abundance of β-LH mRNA. In contrast, photoperiodic influences on prolactin secretion were correlated with alterations in steady-state mRNA abundance.     

Dynamics of noradrenergic circadian input to the chicken pineal gland

To analyze the dynamics of sympathetic input to the chicken pineal the concentrations of catecholamines, indoleamines and some of their metabolites were determined by high performance liquid chromatography with electrochemical detection (HPLC-EC) in the pineal glands of young chickens killed at different times of day. Rhythmic variations over 24 h were observed in tissue levels of dopamine (DA), 5-hydroxytryptamine (5-HT), N-acetylserotonin (NAS) and 5-hydroxyindoleacetic acid (5-HIAA), while norepinephrine (NE) concentrations exhibited no significant change. DA content peaked 2 h after onset of darkness and NAS was detectable only during the night. A bimodal pattern of 5-HT and 5-HIAA levels was observed with peak tissue levels occurring at dawn and dusk. To determine the possible differential effects of light on these biogenic amines, birds were sacrificed at midday, midnight and at midnight following a 1 h exposure to light, and their pineals processed for HPLC-EC. NE, DA and 5-HT levels were similar at midday and midnight, while 5-HIAA and NAS were elevated during the night. Midnight illumination decreased NE and NAS levels, increased 5-HT and 5-HIAA levels and had no effect on DA levels. Temporal variations in NE turnover were determined by pretreating young chickens with alpha-methyl-p-tyrosine, a tyrosine hydroxylase inhibitor, and measuring the rates of decline in NE content over 2 h at midday and midnight in birds held on light cycles and at mid-subjective day in birds held in constant darkness (DD).(ABSTRACT TRUNCATED AT 250 WORDS)     

Somatostatin release as measured by in vivo microdialysis: circadian variation and effect of prolonged food deprivation

In vivo microdialysis was used to determine SRIF release from the hypothalamus in unanesthetized male rats over a period of 24 h and in rats deprived of food for 72 h, in relation to changes in plasma GH levels. Before the experiment, a microdialysis probe was inserted into the anterior pituitary gland of the rats with an indwelling right atrial cannula. Dialysates and blood samples were collected serially, after normal feeding or 72-h deprivation of food. Normal rats implanted with the microdialysis probe showed an episodical pattern of GH secretion at intervals of 3 h. SRIF was secreted in a pulsatile fashion in the dark period in a similar manner to the light period. Mean SRIF pulse amplitude and mean SRIF level were significantly increased in the dark period. There was no significant correlation between the SRIF and GH pulses in the light period. SRIF levels in dialysates obtained from fed rats and food-deprived rats showed a pulsatile pattern. Food deprivation resulted in significant increases in mean SRIF level and mean SRIF pulse amplitude. These results suggest that the existence of circadian rhythm in SRIF release and the increase in SRIF release play an important role in suppressing GH secretion during prolonged food deprivation.     

Adjustment of the human melatonin and cortisol rhythms to shortening of the natural summer photoperiod

Fifteen human subjects were exposed to natural outdoor summer light from 0415 h until 2000 h for 4 days and then from 0800 h until 1600 h for another 4 days. Following shortening of the natural summer photoperiod, times of the morning salivary melatonin decline and cortisol rise did not change whereas the time of the evening melatonin rise phase advanced by about 1.5 h within 1 day and further did not change significantly. Consequently, the melatonin signal duration extended markedly within 1 day. The data show that the compressed melatonin rhythm waveform in humans experiencing a long natural summer photoperiod from sunrise until sunset may change rapidly following a shortening of the photoperiod.     

The Influence of Photoperiod on the Hypothalamic Content of Beta-Endorphin and the Luteinizing Hormone Responses to Naloxone and to Steroid Withdrawal in the Male Syrian Hamster

The aim of this study was to investigate the influence of inhibitory photoperiods upon opioidergic function, as determined by changes in the hypothalamic content of β-endorphin and the luteinizing hormone response to opioidergic receptor blockade, in the male Syrian hamster over the course of gonadal involution and spontaneous gonadal recrudescence. Animals exposed to an 8 h light: 16 h dark cycle (8L: 16D) for 14 weeks underwent gonadal regression. Regression was also observed in animals held for 7 weeks on one of a range of short daylengths of between 11.5 h and 13.5 h, the degree of atrophy being greatest in those animals on the shortest daylength. The tissue concentration of β-endorphin within the mediobasal hypothalamus was significantly higher in animals exposed to 8L: 16D for 14 weeks than in gonadally active controls held on long days (16L: 8D). Exposure to photoperiods of less than 13.5 h for 7 weeks also caused a significant increase in the β-endorphin content of the mediobasal hypothalamus and there was a positive correlation between the concentration of β-endorphin, the degree of gonadal atrophy and the shortness of the photoperiod. Endorphin levels within the preoptic area were not affected by photoperiodic treatments. Exposure of intact animals to 8L: 16D for 12 weeks caused gonadal atrophy and an associated loss of the luteinizing hormone responses to both naloxone and castration. Castrated animals receiving testosterone replacement (cast + T) also exhibited photoinhibition, in the form of reduced serum levels of luteinizing hormone, and this was similarly accompanied by a loss of sensitivity to naloxone and to withdrawal of steroid. Prolonged exposure to 8L:16D led to spontaneous reactivation of the gonadotrophic axis as a consequence of the development of scotorefractoriness. In both gondally intact animals and in cast + T groups, this was associated with a restoration, in parallel, of the luteinizing hormone responses to naloxone and to castration/ steroid withdrawal. The time-course of the restoration of the response to steroid withdrawal in castrates was not significantly different to that observed in intact animals. The luteinizing hormone response to naloxone took significantly longer to redevelop in cast + T groups than it did in gonadally intact animals. The data demonstrate that central opioid systems are sensitive to photoperiod and are consistent with the hypothesis that opioids are involved in the neuroendocrine regulation of reproductive responses to daylength.     

Postnatal light alters hypothalamic‐pituitary‐adrenal axis function and induces a depressive‐like phenotype in adult mice

The postnatal light environment that a mouse experiences during the critical first three postnatal weeks has long‐term effects on both its circadian rhythm output and clock gene expression. Furthermore, data from our lab suggest that postnatal light may also impact the hypothalamic‐pituitary‐adrenal (HPA) axis, which is a key regulator of stress. To test the effect of postnatal light exposure on adult stress responses and circadian rhythmicity, we raised mice under either 24‐h light – dark cycles (LD), constant light (LL) or constant dark (DD) during the first three postnatal weeks. After weaning we then exposed all animals to LD cycles (basal conditions), followed by LL (stressed conditions) environments. We examined brain neuropeptide and glucocorticoid receptor (GR) expression, plasma corticosterone concentration rhythm and body temperature rhythm, together with depression‐ and anxiety‐related behaviour. Results showed that LL‐ and DD‐raised mice exhibited decreased GR expression in the hippocampus, increased plasma corticosterone concentration at the onset of the dark phase and a depressive phenotype when exposed to LD cycles later in life. Furthermore, LL‐raised mice showed increased corticotrophin‐releasing hormone mRNA expression in the paraventricular nucleus of the hypothalamus. When exposed to LL as adults, LL‐raised mice showed a significant circadian rhythm of plasma corticosterone concentration, together with a shorter period and stronger circadian rhythm of body temperature compared to DD‐raised mice. Taken together, these data suggest that altered postnatal light environments have long‐term effects on the HPA axis and the circadian system, which can lead to altered stress responses and a depressive phenotype in adulthood.     

Photoperiodic Control of the Concentration of Luteinizing Hormone, Prolactin and Testosterone in the Male Emu (Dromaius novaehollandiae), a Bird that Breeds on Short Days

The objective of this study was to establish, for a short-day breeding bird, the male emu, whether the breeding season is principally controlled by changes in photoperiod, and to investigate the endocrine mechanisms involved. Two groups of adult males were subjected to three alternating periods of 150-185 days of 14 h light/day (LD) and 10 h light/day (SD) terminating in a 360-day period of LD or SD. Transfer from LD to SD led to increases in plasma concentrations of luteinizing hormone (LH) and testosterone, after 82 +/- 8 and 73 +/- 3 (SEM) days, and an increase in prolactin concentrations after 115 +/- 12 days. Concentrations of LH and testosterone began to decrease before transfer back to LD, at a time when prolactin concentrations were approaching peak values. Transfer from LD to 360 days of SD resulted in increases in LH and testosterone concentrations, and these terminated after an increase in prolactin concentrations. After transfer from SD to 360 days of LD, plasma concentrations of LH and testosterone began to increase, after delays of 222 +/- 24 and 225 +/- 13 days, and were high at the end of the study, while prolactin values remained depressed throughout. These observations clearly show that seasonal breeding in the emu is directly controlled by changes in photoperiod. The dynamics of the hormonal responses to change of photoperiod suggest that, despite being short-day breeders, the photoregulation of breeding in emus involves mechanisms that are currently accepted for birds, rather than mechanisms that have been proposed for short-day breeding mammals. The initiation of breeding in emus is due to dissipation of photorefractoriness by short days which leads to an increase in the secretion of gonadotrophins to levels that are sufficient to support full reproductive condition. The termination of breeding, while days are still short, is due to the antigonadotrophic action of prolactin which, unusually for birds, increases while the days are still short. In conclusion, breeding activity in male emus is strongly controlled by photoperiod. Emus are short-day breeders, but the central mechanisms that regulate the secretion of reproductive hormones seem to be similar to those previously proposed for long-day breeding birds. The pattern of prolactin secretion in emus suggests an important role for this hormone in the termination of the breeding cycle.     

Circadian variation of brain histamine in goldfish

Teleosts may make an excellent model to study brain histamine function. Fishes are phylogenetically closer to the basic vertebrate blueprint than higher vertebrates. They appear to have a simpler histaminergic system in terms of central nervous system distribution and, contrary to higher vertebrates, brain histamine appears to be strictly neuronal. In this preliminary study, we examined circadian variation of brain histamine in goldfish, Carassius auratus, as this neurotransmitter correlates with circadian behavior of some mammals. Two groups of juvenile goldfish were held in 24 60L aquaria, six fish per aquarium, on reversed photoperiods; L:D 12:12 with light onset either at 0700 or 1900h. Fish were sampled every 4h. At a sampling time, all the fish in a tank were taken; each sampling, for both groups, was done in replicate. Brain histamine was determined by immunoassay. There was a significant circadian variation in histamine on both photoperiod regimes with the highest levels during the photophase. These results support the hypothesis of an early phylogenic role for histamine in vertebrate circadian physiology.     

Differential Melanin-Concentrating Hormone Gene Expression in Two Hypothalamic Nuclei of the Teleost Tilapia in Response to Environmental Changes

For some teleosts, a role has been established for melanin-concentrating hormone (MCH) background adaptation and stress response. In teleost fishes, prepro-MCH (ppMCH) rnRNA is expressed in the hypothalamus, predominantly in neurons of the nucleus lateralis tuberis (NLT) and in scattered cells of the nucleus recessus lateralis (NRL). The response of mature tilapia to different environmental challenges was studied by assessing ppMCH rnRNA levels in these two hypothalamic nuclei by quantitative dot blot analysis. Changes in background colour induced pronounced differences in ppMCH mRNA expression in the NLT, but not in the NRL. The NLT of tilapia adapted to a white background contained 2.5 to 3 times more ppMCH mRNA than the NLT of black-adapted fish. The NLT of fish kept on neutral background contained intermediate levels of ppMCH mRNA, which were significantly lower than the levels in white-adapted fish. Oral administration of dexamethasone lowered plasma cortisol concentrations, but had no effect on ppMCH mRNA levels in white- and black-adapted fish. In tilapia exposed to strongly acidified water (pH 3.5), plasma cortisol and ACTH concentrations were highly elevated, and plasma chloride concentrations considerably lower than in controls. These fish responded with a 70% rise in ppMCH mRNA levels in the NLT, which is most probably associated with a stress response evoked by inadequate osmoregulation. After exposure to a milder acidification (pH 4.0) or to seawater no significant changes in ppMCH mRNA levels occurred in either the NLT or the NRL, nor in plasma chloride, cortisol and ACTH levels. A specific increase of ppMCH mRNA levels in the NRL was observed in repeatedly disturbed tilapia. We conclude that MCH neurons in the NLT and NRL of this teleost differentially respond to background colour, acidification and disturbance stress, and that this response is not strictly associated with changes in plasma ions and activity of the pituitary-interrenal axis.     

24 Hour changes in catecholamine content of rat thyroid and submaxillary glands

Summary The occurrence of diurnal rhythmicity in tissue norepinephrine (NE), epinephrine (E) and dopamine (DA) levels of rat thyroid and submaxillary glands was assessed in animals killed at eight time intervals during a 24 hour period. In both tissues significant peak values of NE content were found at the third hour of dark onset (at 23:00 h). Additionally a minimum of NE content of thyroid gland was apparent in rats killed at the fifth hour of photophase (at 11:00 h). As in the case of NE, peak values of E content were found during the scotophase in both tissues examined. In contrast, DA levels exhibited no diurnal rhythmicity in the rat thyroid gland, and an early rise at the beginning of the photophase in the submaxillary glands. These results support an activation of the sympathetic nervous system innervating the thyroid and submaxillary glands at the beginning of the scotophase. Tissue DA levels do not resemble the activity of peripheral sympathetic nerves in the tissues examined.This study was supported by grants PID 3-078500 and PID 3-0667500 from the Consejo Nacional de Investigaciones Clientificas y Técnicas, Argentina.     

The Biosynthesis of Melanin-Concentrating Hormone in a Fish

This work investigated the biosynthesis of a neurohypophysial hormone, melanin-concentrating hormone (MCH), in the trout. Sephadex G-75 chromatography showed the presence of several large MCH-immunoreactive molecules in hypothalamic and pituitary gland extracts, with different retention times on high-performance liquid chromatography from the mature MCH^1–17. About 10% of the total MCH-immunoreactivity in the hypothalamus was attributable to large molecular weight forms but these contributed less than 1% to the immunoreactivity in the neurointermediate lobe. Both [³⁵ S]methionine and [³ H]leucine were injected into the hypothalamus near the MCH perikarya (nucleus lateralis tuberis region) of anaesthetized fish, after which the fish were killed at intervals of up to 8 h post-injection and the basal hypothalami, pituitary pars distales and neurointermediate lobes were extracted in acid. MCH-related immunoprecipitates from these extracts were fractionated by sodium dodecyl sulphate polyacrylamide gel electrophoresis or by Sephadex G-50 chromatography. The results show the incorporation of radiolabel into 15.3 K and 11.3 K precursors within 0.75 h, and their conversion, via several smaller intermediates, to a molecule resembling MCH^1–17. The results are discussed in relation to the known cDNA sequence of salmon MCH. Labelled MCH first appeared in the neurointermediate lobe 4 h after injection, giving an estimated transit rate of 0.4 mm/h.     

Glucocorticoid-mediated nycthemeral and photoperiodic regulation of tph2 expression

In the Syrian hamster dorsal and median raphé nuclei, the tryptophan hydroxylase 2 gene (tph2), which codes the rate-limiting enzyme of serotonin synthesis, displays daily variations in its expression in animals entrained to a long but not to a short photoperiod. The present study aimed to assess the role of glucocorticoids in the nycthemeral and photoperiodic regulation of daily tph2 expression. In hamsters held in long photoperiod from birth, after adrenalectomy and glucocorticoid implants the suppression of glucocorticoid rhythms induced an abolition of the daily variations in tph2-mRNA concentrations, a decrease in the amplitude of body temperature rhythms and an increase in testosterone levels. All these effects were reversed after experimental restoration of a clear daily rhythm in the plasma glucocorticoid concentrations. We conclude that the photoperiod-dependent rhythm of glucocorticoids is the main regulator of tph2 daily expression.