The dual-oscillator system of Drosophila melanogaster under natural-like temperature cycles

Dual-oscillator systems that control morning and evening activities can be found in a wide range of animals. The two coupled oscillators track dawn and dusk and flexibly adapt their phase relationship to seasonal changes. This is also true for the fruit fly Drosophila melanogaster that serves as model organism to understand the molecular and anatomical bases of the dual-oscillator system. In the present study, the authors investigated which temperature parameters are crucial for timing morning and evening activity peaks by applying natural-like temperature cycles with different daylengths. The authors found that the morning peak synchronizes to the temperature increase in the morning and the evening peak to the temperature decrease in the afternoon. The two peaks did not occur at fixed absolute temperatures, but responded flexibly to daylength and overall temperature level. Especially, the phase of the evening peak clearly depended on the absolute temperature level: it was delayed at high temperatures, whereas the phase of the M peak was less influenced. This suggests that the two oscillators have different temperature sensitivities. The bimodal activity rhythm was absent in the circadian clock mutants Clk(Jrk) and cyc(01) and reduced in per(01) and tim(01) mutants. Whereas the activity of Clk(Jrk) mutants just followed the temperature cycles, that of per(01) and tim(01) mutants did not, suggesting that these mutants are not completely clockless. This study revealed new characteristics of the dual-oscillator system in Drosophila that were not detected under different photoperiods.     

The Nocturnal Activity of Fruit Flies Exposed to Artificial Moonlight Is Partly Caused by Direct Light Effects on the Activity Level That Bypass the Endogenous Clock

Artificial moonlight was recently shown to shift the endogenous clock of fruit flies and make them nocturnal. To test whether this nocturnal activity is partly due to masking effects of light, we exposed the clock‐mutants per⁰¹, tim⁰¹, per⁰¹;tim⁰¹, cyc⁰¹, and Clk^JRK to light/dark and light/dim‐light cycles and determined the activity level during the day and night. We found that under moonlit nights, all clock mutants shifted their activity significantly into the night, suggesting that this effect is independent of the clock. We also recorded the flies under continuous artificial moonlight and darkness to judge the effect of dim constant light on the activity level. All mutants, except Clk^JRK flies, were significantly more active under artificial moonlight conditions than under complete darkness. Unexpectedly, we found residual rhythmicity of per⁰¹ and especially tim⁰¹ mutants under these conditions, suggesting that TIM and especially PER retained some activity in the absence of its respective partner. Nevertheless, as even the double mutants and the cyc⁰¹ and Clk^JRK mutants shifted their activity into the night, we conclude that dim light stimulates the activity of fruit flies in a clock‐independent manner. Thus, nocturnal light has a twofold influence on flies: it shifts the circadian clock, and it increases nocturnal activity independently of the clock. The latter was also observed in some primates by others and might therefore be of a more general validity.     

Role for Circadian Clock Genes in Seasonal Timing: Testing the Bünning Hypothesis

A major question in chronobiology focuses around the “Bünning hypothesis” which implicates the circadian clock in photoperiodic (day-length) measurement and is supported in some systems (e.g. plants) but disputed in others. Here, we used the seasonally-regulated thermotolerance of Drosophila melanogaster to test the role of various clock genes in day-length measurement. In Drosophila, freezing temperatures induce reversible chill coma, a narcosis-like state. We have corroborated previous observations that wild-type flies developing under short photoperiods (winter-like) exhibit significantly shorter chill-coma recovery times (CCRt) than flies that were raised under long (summer-like) photoperiods. Here, we show that arrhythmic mutant strains, per⁰¹, tim⁰¹ and Clk^Jrk, as well as variants that speed up or slow down the circadian period, disrupt the photoperiodic component of CCRt. Our results support an underlying circadian function mediating seasonal daylength measurement and indicate that clock genes are tightly involved in photo- and thermo-periodic measurements.     

Circadian expression of the presynaptic active zone protein bruchpilot in the lamina of Drosophila melanogaster

In the fly's visual system, the morphology of cells and the number of synapses change during the day. In the present study we show that in the first optic neuropil (lamina) of Drosophila melanogaster, a presynaptic active zone protein Bruchpilot (BRP) exhibits a circadian rhythm in abundance. In day/night (or light/dark, LD) conditions the level of BRP increases two times, in the morning and in the evening. The same pattern of changes in the BRP level was detected in whole brain homogenates, thus indicating that the majority of synapses in the brain peaks twice during the day. However, these two peaks in BRP abundance, measured as the fluorescence intensity of immunolabeling, seem to be regulated differently. The peak in the morning is predominantly regulated by light and involves the transduction pathway in the retina photoreceptors. This peak is present neither in wild‐type Canton‐S flies in constant darkness (DD), nor in norpA⁷ phototransduction mutant in LD. However, it also depends on the clock gene per, because it is abolished in the per⁰ arrhythmic mutant. In turn, the peak of BRP in the evening is endogenously regulated by an input from the pacemaker located in the brain. This peak is present in Canton‐S flies in DD, as well as in the norpA⁷ mutant in LD, but is absent in per⁰¹, tim,⁰¹ and cry⁰¹ mutants in LD. In addition both peaks seem to depend on clock gene‐expressing photoreceptors and glial cells of the visual system. © 2012 Wiley Periodicals, Inc. Develop Neurobiol, 2013     

Insights into the Molecular Mechanisms of Temperature Compensation from the Drosophila Period and Timeless Mutants

The relative constancy of the circadian period over a wide range of temperatures is a general property of circadian rhythms. Insights into the molecular mechanisms of temperature compensation are emerging from genetic and molecular genetic studies of the period (per) and timeless (tim) genes in Drosophila. These genes encode proteins that are thought to be part of a negative feedback cycle, which results in circadian oscillations of both per and tim mRNA, as well as a complex of the two proteins. Complex formation is temporally regulated and apparently necessary for nuclear localization of both per and tim proteins. While insights into the roles of per and tim in temperature compensation have been intriguing, they have also been somewhat perplexing. For instance, the interaction of wild-type per peptides is relatively insensitive to temperature in the yeast two-hybrid assay or in assays employing in-vitro-translated peptides, while the interaction of per^L mutant peptides is reduced at a high temperature. Apparently, the per^L mutation increases an intramolecular interaction between different parts of the per peptide in these assays, and this interaction reduces the amount of per homodimer. On the other hand, the same assays show that the intermolecular interaction between the per and tim peptides is reduced at a high temperature by the per^L mutation; this reduction does not require the competing intramolecular interaction. Despite this difference, in all of the experiments employing these assays the per^L mutation has rendered per-per and per-tim peptide interactions sensitive to high temperature, so it is likely that one or both of these reduced interactions contribute to the longer circadian periods at high temperature in per^L mutant flies. However, the tim^SL and per^S mutations, as well as deletion of the Thr-Gly repeats from per, affect temperature compensation but have not been shown to affect these molecular interactions of per and tim. Finally, a recent report of oscillating per and tim proteins in the cytoplasm (rather than the nuclei) of silk moth neurons may suggest an alternative mechanism for per and tim function in these cells. (Chronobiology International 14(5), 455–468, 1997)     

Adapting Test Timing to the Sleep-Wake Schedule: Effects on Diurnal Neurobehavioral Performance Changes in Young Evening and Older Morning Chronotypes

The synchrony effect refers to the beneficial impact of temporal matching between the timing of cognitive task administration and preferred time-of-day for diurnal activity. Aging is often associated with an advance in sleep-wake timing and concomitant optimal performance levels in the morning. In contrast, young adults often perform better in the evening hours. So far, the synchrony effect has been tested at fixed clock times, neglecting the individual's sleep-wake schedule and thus introducing confounds, such as differences in accumulated sleep pressure or circadian phase, which may exacerbate synchrony effects. To probe this hypothesis, the authors tested older morning and young evening chronotypes with a psychomotor vigilance and a Stroop paradigm once at fixed morning and evening hours and once adapting testing time to their preferred sleep-wake schedule in a within-subject design. The authors observe a persistence of synchrony effects for overall median reaction times during a psychomotor vigilance task, even when testing time is adapted to the specific individual's sleep-wake schedule. However, data analysis also indicates that time-of-day modulations are weakened under those conditions for incongruent trials on Stroop performance and the slowest reaction times on the psychomotor vigilance task. The latter result suggests that the classically observed synchrony effect may be partially mediated by a series of parameters, such as differences in socio-professional timing constraints, the amount of accumulated sleep need, or circadian phase, all leading to differential arousal levels at testing. (Author correspondence: f.collette@ulg.ac.be)     

Short-day and long-day expression patterns of genes involved in the flesh fly clock mechanism: period,timeless, cycle and cryptochrome

Though our knowledge of the molecular details of the circadian clock has advanced rapidly, the functional elements of the photoperiodic clock remain largely unknown. As a first step to approach this issue, we report here the sequences and expression patterns of period (per), timeless (tim), cycle (cyc) and cryptochrome (cry) mRNAs in the flesh fly Sarcophaga crassipalpis. Nucleotide and deduced amino acid sequences of the genes in S. crassipalpis show high similarity to homologous genes in other insects that have been investigated. S. crassipalpis TIM has a unique C-terminus that contains a poly Q region. A diel rhythmicity of per and tim mRNA abundance was detected in the adult heads (peak during scotophase), while cry and cyc mRNA abundance remained fairly constant throughout. The abundance of cyc mRNA was quite low when compared to per, tim and cry mRNA. Rearing temperature affected the amount of per and tim mRNAs: abundance of per mRNA increased at 20 °C when compared to 25 °C, but that of tim mRNA decreased. Photoperiod influenced the expression patterns of per and tim mRNA: the peak of per mRNA expression shifted in concert with onset of the scotophase, while a shift in tim mRNA expression was less pronounced. The amplitude of tim mRNA was severely dampened under long daylength, but that of per mRNA was not affected. These distinct patterns of expression suggest that this information could be used to determine photoperiodic responses such as diapause.     

Orcadian Rhythm of Serotonin Binding in Rat Brain—I. Effect of the Light-Dark Cycle

Moving rapidly from a supine to a standing posture is a common daily activity, yet a significant physiological challenge. Syncope can result from the development of initial orthostatic hypotension (IOH) involving a transient fall in systolic/diastolic blood pressure (BP) of >40/20?mm Hg within the first 15 s, and/or a delayed orthostatic hypotension (DOH) involving a fall in systolic/diastolic BP of >20/10?mm Hg within 15?min of posture change. Although epidemiological data indicate a heightened syncope risk in the morning, little is known about the diurnal variation in the IOH and DOH mechanisms associated with postural change. The authors hypothesized that the onset of IOH and DOH occurs sooner, and the associated cardiorespiratory and cerebrovascular changes are more pronounced, in the early morning. At 06:00 and 16:00?h, 17 normotensive volunteers, aged 26?±?1 yrs (mean?±?SE), completed a protocol involving supine rest, an upright stand, and a 60° head-up tilt (HUT) during which continuous beat-to-beat measurements of middle cerebral artery velocity (MCAv), mean arterial BP (MAP), heart rate, and end-tidal Pco₂ (P_ETco₂) were obtained. Mean MCAv was ～12% lower at baseline in the morning (p?≤?.01) and during the HUT (p?<?.01), despite a morning elevation in P_ETco₂ by ～2.2?mm Hg (p?=?.01). The decline in MAP during initial standing (morning vs. afternoon: 50%?±?4% vs. 49%?±?3%) and HUT (39%?±?3% vs. 38%?±?3%) did not vary with time-of-day (p?>?.30). In conclusion, although there is a marked reduction in MCAv in the morning, there is an absence of diurnal variation in the onset of and associated physiological responses associated with IOH and DOH. These responses, at least in this population, are unlikely contributors to the diurnal variation in orthostatic tolerance. (Author correspondence: N.C.Lewis@2004.ljmu.ac.uk)     

Atypical expression of circadian clock genes in denervated mouse skeletal muscle

Muscle force production and power output in active males, regardless of the site of measurement (hand, leg, or back), are higher in the evening than in the morning. This diurnal variation is attributed to motivational, peripheral and central factors, and higher core and, possibly, muscle temperatures in the evening. This study investigated whether increasing morning rectal temperatures to evening resting values, by active or passive warm-ups, leads to muscle force production and power output becoming equal to evening values in motivated subjects. Ten healthy active males (mean ± SD: age, 21.2 ± 1.9 yrs; body mass, 75.4 ± 8 kg; height, 1.76 ± .06 m) completed the study, which was approved by the University Ethics Committee. The subjects were familiarized with the techniques and protocol and then completed four sessions (separated by at least 48 h): control morning (07:30 h) and evening (17:30 h) sessions (with an active 5-min warm-up) and then two further sessions at 07:30 h but proceeded by an extended active or passive warm-up to raise rectal temperature to evening values. These last two sessions were counterbalanced in order of administration. During each trial, three measures of handgrip strength, isokinetic leg strength measurements (of knee flexion and extension at 1.05 and 4.19 rad.s^?1 through a 90° range of motion), and four measures of maximal voluntary contraction (MVC) on an isometric ergometer (utilizing the twitch-interpolation technique) were performed. Rectal and intra-aural temperatures, ratings of perceived exertion (RPE) and thermal comfort (TC) were measured. Measurements were made after the subjects had reclined for 30 min and after the warm-ups and prior to the measurement of handgrip and isokinetic and isometric ergometry. Muscle temperature was taken after the warm-up and immediately before the isokinetic and MVC measurements. Warm-ups were either active (cycle ergometer at 150 W) or passive (resting in a room at 35°C, relative humidity 45%). Data were analyzed using analysis of variance models with repeated measures. Rectal and intra-aural temperatures were higher at rest in the evening (.56°C and .74°C; p < .05) than in the morning, but there were no differences after the active or passive warm-ups, the subjects' ratings of thermal comfort reflecting this. Muscle temperatures also displayed significant diurnal variation, with higher values in the evening (～.31°C; p < .05). Grip strength, isokinetic knee flexion for peak torque and peak power at 1.05 rad.s^?1, and knee extension for peak torque at 4.19 rad.s^?1 all showed higher values in the evening. All other measures of strength or power showed a trend to be higher in the evening ( .10 > p > .05). There was no significant effect of active or passive warm-ups on any strength or power variable, and subjects reported maximal values for effort for each strength measure. In summary, effects of time of day were seen in some measures of muscle performance but, in this population of motivated subjects, there was no evidence that increasing morning rectal temperature to evening values by active or passive warm-up increased muscle strength to evening values. (Author correspondence: B.J.Edwards@ljmu.ac.uk)     

Scheduled Feeding Alters the Timing of the Suprachiasmatic Nucleus Circadian Clock in Dexras1-Deficient Mice

Restricted feeding (RF) schedules are potent zeitgebers capable of entraining metabolic and hormonal rhythms in peripheral oscillators in anticipation of food. Behaviorally, this manifests in the form of food anticipatory activity (FAA) in the hours preceding food availability. Circadian rhythms of FAA are thought to be controlled by a food-entrainable oscillator (FEO) outside of the suprachiasmatic nucleus (SCN), the central circadian pacemaker in mammals. Although evidence suggests that the FEO and the SCN are capable of interacting functionally under RF conditions, the genetic basis of these interactions remains to be defined. In this study, using dexras1-deficient (dexras1^?/?) mice, the authors examined whether Dexras1, a modulator of multiple inputs to the SCN, plays a role in regulating the effects of RF on activity rhythms and gene expression in the SCN. Daytime RF under 12L:12D or constant darkness (DD) resulted in potentiated (but less stable) FAA expression in dexras1^?/? mice compared with wild-type (WT) controls. Under these conditions, the magnitude and phase of the SCN-driven activity component were greatly perturbed in the mutants. Restoration to ad libitum (AL) feeding revealed a stable phase displacement of the SCN-driven activity component of dexras1^?/? mice by ～2?h in advance of the expected time. RF in the late night/early morning induced a long-lasting increase in the period of the SCN-driven activity component in the mutants but not the WT. At the molecular level, daytime RF advanced the rhythm of PER1, PER2, and pERK expression in the mutant SCN without having any effect in the WT. Collectively, these results indicate that the absence of Dexras1 sensitizes the SCN to perturbations resulting from restricted feeding. (Author correspondence: haiying.cheng@utoronto.ca)     

Pigment‐dispersing factor affects circadian molecular oscillations in the cricket Gryllus bimaculatus

Pigment‐dispersing factor (PDF) is an important neurotransmitter in insect circadian systems. In the cricket Gryllus bimaculatus, it affects nocturnal activity, the free‐running period and photic entrainment. In this study, to investigate whether these effects of PDF occur through a circadian molecular machinery, we measured mRNA levels of clock genes period (per) and timeless (tim) in crickets with pdf expression knocked‐down by pdf RNAi. The pdf RNAi decreased per and tim mRNA levels during the night to reduce the amplitude of their oscillation. The phase of the rhythm advanced by about 4 h in terms of trough and/or peak phases. On the other hand, pdf mRNA levels were little affected by per and tim RNAi treatment. These results suggest that PDF affects the circadian rhythm at least in part through the circadian molecular oscillation while the circadian clock has little effect on the pdf expression.     

Role of Patient Chronotype on Circadian Pattern of Myocardial Infarction: A Pilot Study

Population-based studies indicate the risk of acute myocardial infarction (AMI) is greatest in the morning, during the initial hours of diurnal activity. The aim of this pilot study was to determine whether chronotype, i.e., morningness and eveningness, impacts AMI onset time. The sample comprised 63 morning- and 40 evening-type patients who were classified by the Horne-Östberg Morningness-Eveningness Questionnaire (MEQ) in the hospital after experiencing the AMI. The average wake-up and bed times of morning types were ～2?h earlier than evening types. Although the lag in time between waking up from nighttime sleep and AMI onset during the day did not differ between the two chronotypes, the actual clock-hour time of the peak in the 24-h AMI pattern did. The peak in AMI of morning types occurred between 06:01 and 12:00?h and that of the evening types between 12:01 and 18:00?h. Although the results of this small sample pilot study suggest one's chronotype influences the clock time of AMI onset, larger scale studies, which also include assessment of 24-h patterning of events in neither types, must be conducted before concluding the potential influence of chronotype on the timing of AMI onset. (Author correspondence: dryavuzselvi@yahoo.com).     

Modeling Temperature Compensation in Chemical and Biological Oscillators

All physicochemical and biological oscillators maintain a balance between destabilizing reactions (as, for example, intrinsic autocatalytic or amplifying reactions) and stabilizing processes. These two groups of processes tend to influence the period in opposite directions and may lead to temperature compensation whenever their overall influence balances. This principle of “antagonistic balance” has been tested for several chemical and biological oscillators. The Goodwin negative feedback oscillator appears of particular interest for modeling the circadian clocks in Neurospora and Drosophila and their temperature compensation. Remarkably, the Goodwin oscillator not only gives qualitative, correct phase response curves for temperature steps and temperature pulses, but also simulates the temperature behavior of Neurospora frq and Drosophila per mutants almost quantitatively. The Goodwin oscillator predicts that circadian periods are strongly dependent on the turnover of the clock mRNA or clock protein. A more rapid turnover of clock mRNA or clock protein results, in short, a slower turnover in longer period lengths. (Chronobiology International, 14(5), 499–510, 1997)     

Seasonal Variation of Temporal Niche in Wild Owl Monkeys (Aotus azarai azarai) of the Argentinean Chaco: A Matter of Masking?

Among the more than 40 genera of anthropoid primates (monkeys, apes, and humans), only the South American owl monkeys, genus Aotus, are nocturnal. However, the southernmostly distributed species, Aotus azarai azarai, of the Gran Chaco may show considerable amounts of its 24-h activity during bright daylight. Due to seasonal changes in the duration of photophase and climatic parameters in their subtropical habitat, the timing and pattern of their daily activity are expected to show significant seasonal variation. By quantitative long-term activity recordings with Actiwatch AW4 accelerometer data logger devices of 10 wild owl monkeys inhabiting a gallery forest in Formosa, Argentina, the authors analyzed the seasonal variation in the temporal niche and activity pattern resulting from entrainment and masking of the circadian activity rhythm by seasonally and diurnally varying environmental factors. The owl monkeys always displayed a distinct bimodal activity pattern, with prominent activity bouts and peaks during dusk and dawn. Their activity rhythm showed distinct lunar and seasonal variations in the timing and daily pattern. During the summer, the monkeys showed predominantly crepuscular/nocturnal behavior, and a crepuscular/cathemeral activity pattern with similar diurnal and nocturnal activity levels during the cold winter months. The peak times of the evening and morning activity bouts were more closely related to the times of sunset and sunrise, respectively, than activity-onset and -offset. Obviously, they were better circadian markers for the phase position of the entrained activity rhythm than activity-onset and -offset, which were subject to more masking effects of environmental and/or internal factors. Total daily activity was lowest during the two coldest lunar months, and almost twice as high during the warmest months. Nighttime (21:00–06:00?h) and daytime (09:00–18:00?h) activity varied significantly across the year, but in an opposite manner. Highest nighttime activity occurred in summer and maximal daytime activity during the cold winter months. Dusk and dawn activity, which together accounted for 43% of the total daily activity, barely changed. The monkeys tended to terminate their nightly activity period earlier on warm and rainy days, whereas the daily amount of activity showed no significant correlation either with temperature or precipitation. These data are consistent with the dual-oscillator hypothesis of circadian regulation. They suggest the seasonal variations of the timing and pattern of daily activity in wild owl monkeys of the Argentinean Chaco result from a specific interplay of light entrainment of circadian rhythmicity and strong masking effects of various endogenous and environmental factors. Since the phase position of the monkeys' evening and morning activity peaks did not vary considerably over the year, the seasonal change from a crepuscular/nocturnal activity pattern in summer to a more crepuscular/cathemeral one in winter does not depend on a corresponding phase shift of the entrained circadian rhythm, but mainly on masking effects. Thermoregulatory and energetic demands and constraints seem to play a crucial role. (Author correspondence: hans.erkert@t-online.de)     

Photoreception in decapitated larvae of silkworm Bombyx mori

To investigate the photoreception that controls daily oscillations at the periphery in insects, we decapitated larvae of the silkworm Bombyx mori (Lepidoptera: Bombycidae) by ligature, and observed rhythms in their peripheral tissues under several light conditions. We measured the mRNA expression of period (per) and timeless (tim), which are homologues of Drosophila clock genes that function in the core oscillator of the circadian clock system. The expression of both per and tim significantly changed in the midgut, Malpighian tubules and silk glands of decapitated larvae exposed to photophase and scotophase that were reversed from the original daily light–dark cycle under which the larvae were housed. Under constant darkness, the daily expression of tim mRNA persisted for at least one cycle in the midgut and silk gland. In addition, an appropriate light stimulus under constant darkness induced a significant phase shift in the endogenous timing system (probably a circadian clock) that determined peak levels of tim mRNA expression in the midgut and silk glands of decapitated larvae. Since light regulated the gene expression rhythm in peripheral tissues of decapitated silkworm larvae, neither the brain nor eyes were essential for photoreception to control daily oscillations in these tissues. Thus, peripheral tissues in insects might directly use light even at the larval stage.     

Circadian oscillations outside the optic lobe in the cricket Gryllus bimaculatus

Although circadian rhythms are found in many peripheral tissues in insects, the control mechanism is still to be elucidated. To investigate the central and peripheral relationships in the circadian organization, circadian rhythms outside the optic lobes were examined in the cricket Gryllus bimaculatus by measuring mRNA levels of period (per) and timeless (tim) genes in the brain, terminal abdominal ganglion (TAG), anterior stomach, mid-gut, testis, and Malpighian tubules. Except for Malpighian tubules and testis, the tissues showed a daily rhythmic expression in either both per and tim or tim alone in LD. Under constant darkness, however, the tested tissues exhibited rhythmic expression of per and tim mRNAs, suggesting that they include a circadian oscillator. The amplitude and the levels of the mRNA rhythms varied among those rhythmic tissues. Removal of the optic lobe, the central clock tissue, differentially affected the rhythms: the anterior stomach lost the rhythm of both per and tim; in the mid-gut and TAG, tim expression became arrhythmic but per maintained rhythmic expression; a persistent rhythm with a shifted phase was observed for both per and tim mRNA rhythms in the brain. These data suggest that rhythms outside the optic lobe receive control from the optic lobe to different degrees, and that the oscillatory mechanism may be different from that of Drosophila.     

Melatonin Treatment Effects on Adolescent Students' Sleep Timing and Sleepiness in a Placebo-Controlled Crossover Study

During the last few decades, the incidence of sleep-onset insomnia, due to delay of circadian phase, has increased substantially among adolescents all over the world. We wanted to investigate whether a small dose of melatonin given daily, administered in the afternoon, could advance the sleep timing in teenagers. Twenty-one students, aged 14–19 yrs, with sleep-onset difficulties during school weeks were recruited. The study was a randomized, double blind, placebo (PL)-controlled crossover trial, lasting 5 wks. During the first 6 d in wks 2 and 4, the students received either PL or melatonin (1 mg) capsules between 16:30 and 18:00 h. During the first 6 d of wk 5, all students received melatonin. Wks 1 and 3 were capsule-free. In the last evening of each week and the following morning, the students produced saliva samples at home for later melatonin analysis. The samples were produced the same time each week, as late as possible in the evening and as early as possible in the morning. Both the student and one parent received automatic mobile text messages 15 min before saliva sampling times and capsule intake at agreed times. Diaries with registration of presumed sleep, subjective sleepiness during the day (Karolinska Sleepiness Scale, KSS) and times for capsule intake and saliva samplings were completed each day. Primary analysis over 5 wks gave significant results for melatonin, sleep and KSS. Post hoc analysis showed that reported sleep-onset times were advanced after melatonin school weeks compared with PL school weeks (p < .005) and that sleep length was longer (p < .05). After the last melatonin school week, the students fell asleep 68 min earlier and slept 62 min longer each night compared with the baseline week. Morning melatonin values in saliva diminished compared with PL (p < .001) and evening values increased (p < .001), indicating a possible sleep phase advance. Compared with PL school weeks, the students reported less wake up (p < .05), less school daytime sleepiness (p < .05) and increased evening sleepiness (p < .005) during melatonin weeks. We conclude that a small dose of melatonin given daily, administered in the afternoon, could advance the sleep timing and make the students more alert during school days even if they continued their often irregular sleep habits during weekends. (Author correspondence: arne.lowden@stress.su.se)