- Email: [email protected]
Exposure to long summer days affects the human melatonin and cortisol rhythms
Brain Research 759 Ž1997. 166–170
Short communication
Exposure to long summer days affects the human melatonin and cortisol rhythms Dana Vondrasova, Helena Illnerova´ ˇ ´ Ivan Hajek, ´
)
Institute of Physiology, Academy of Sciences of the Czech Republic, Vıdenska ´ ˇ ´ 1083, 14220 Prague 4, Czech Republic Accepted 11 March 1997
Abstract 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. Keywords: Circadian rhythm; Photoperiod; Melatonin; Cortisol; Human
In all mammals studied so far, the waveform of the circadian rhythm in melatonin production and secretion depends on day length, i.e., on photoperiod w11,14x. On long summer days, light intruding into the late evening hour phase delays the evening melatonin production onset, light intruding into the early morning hour phase advances the morning production offset; consequently, the melatonin signal is shortened. On short winter days, the signal is extended w11x. Hence the melatonin signal duration may provide mammals with information about the season of the year. In humans, the melatonin signal duration appears to depend just slightly on the ambient photoperiod, and only at higher latitudes, e.g. at 608N in Sweden w1x and 688S in Antarctica w18x. However, in temperate zones, e.g. at a latitude 358S w13x, 398N w26x and 508N w12x, respectively, no difference between the summer and winter melatonin signal duration has been found, though in summer the natural photoperiod may last even 16 h and in winter it may be only 8 h. A hypothesis has been put forward that even during the winter time the actual photoperiod may be as long as 16 h, as humans usually spend 8 hrday in darkness during their nighttime sleep and otherwise are exposed to artificial and ambient natural light for 16 h w27x. Therefore, no difference in the melatonin signal duration between the summer and winter time can be expected.
)
Corresponding author. Fax: q42 Ž2. 471-9517.
Indeed, when human subjects experience a lighting regime with just 10 h of light and 14 h of darkness ŽLD 10:14 h. with forced inactivity during the dark period, the melatonin signal duration is extended by about 2 h as compared with the pattern under the artificial and natural light–dark regime ŽLD 16:8. w25,27x. All human subjects studied so far were students and employees who were not exposed to the whole long natural photoperiod during the summer time. Intensity of light under the natural summer photoperiod may be by more than 2 orders higher than intensity under an artificial lighting regime. As the extent of the photic entrainment of the human pacemaker controlling circadian rhythms depends on light intensity w3x, it may be expected that the actual ambient summer photoperiod might shape the melatonin signal more markedly than the combined artificial and natural 16 h photoperiod, the so-called ‘‘short night’’ regime w27x. The present study was undertaken to find out whether exposure to the whole natural summer photoperiod would shorten the melatonin signal and reset the melatonin rhythm relative to the winter ‘‘LD 16:8’’ pattern. Another circadian rhythm controlled by the same pacemaker as the rhythmic melatonin production, namely the rhythm in cortisol secretion w14x, was followed as well, in order to compare phase shifts of both the circadian rhythms. To be able to study the rhythms under field conditions in the country, a non-invasive sampling method was used, i.e., melatonin and cortisol were assayed in the
0006-8993r97r$17.00 Copyright q 1997 Elsevier Science B.V. All rights reserved. PII S 0 0 0 6 - 8 9 9 3 Ž 9 7 . 0 0 3 5 8 - 2
D. VondrasoÕa ˇ ´ et al.r Brain Research 759 (1997) 166–170
saliva. Salivary melatonin and cortisol levels have been reported to be highly correlated with serum levels w15,19,20,24x. Eight high-school students, 4 boys and 4 girls aged 15 years, participated in the study. All the volunteers were high melatonin secretors, with peak values higher than 25 pg of melatoninrml of saliva. The subjects gave informed consent after the nature of the study was fully explained. Both the winter and the summer parts of the study were performed during the school year. The students were synchronized to a similar schedule: they slept daily from 22.00 to 23.00 h until 07.00 h of the local time and were thus exposed to artificial and natural light for 15–16 hrday. During winter, the local time was the same as the standard time, whereas during summer the daylight saving time was introduced and hence the local time was 1 h ahead the standard time. All subsequent data are given in the standard time. Accordingly, before the start of the summer experiment the students slept daily from 21.00 to 22.00 until 06.00 of the standard time. The study was performed near Prague, at 508N latitude. The winter part of the study was performed from January 14 to January 15 ŽFig. 1.. Sunrise occurred at 07.54 h, sunset at 16.24 h; consequently, the ambient natural photoperiod lasted just 8.5 h. The students gathered in the afternoon of January 14 in the house of D.V. Since 16.00 h, they were sampled in 1–2 h intervals until 10.00 h of the next day when they left. They slept in darkness from 22.00 h until 07.00 h; otherwise, they experienced light of an intensity lower than 100 lux. The summer part of the study was performed from June 24 until June 27. At that time, the sunrise occurred at 03.52 h and sunset at 20.13 h; consequently, the natural photoperiod lasted 16.3 h. The students gathered in the afternoon of June 24 in a cabin in the country. From June 24 until June 27, they slept from 21.00 h until 04.00 in darkness and were outdoors until
Fig. 1. Experimental paradigm. Black bars represent sleep in darkness, cross-hatched bars intervals when individual subjects started to sleep, hatched bars light periods of intensity lower than 100 lux and open bars bright light periods. Open triangles indicate times of sunset and sunrise, respectively, arrows the sampling times and double lines times when the outdoor light intensity fell bellow 2000 lux in the evening, and rose above 2000 lux in the morning, respectively.
167
20.00 in the evening and again from 04.15 h in the morning; the intensity of the indoor light between 20.00 h and 21.00 h and between 04.00 h and 04.15 h, respectively, was lower than 100 lux. In the morning, the intensity of the outdoor light exceeded 2000 lux at 04.18 h, 04.15 h and 04.40 h on June 25, 26 and 27, respectively. In the evening, the intensity fell below 2000 lux at 19.45 h, 19.45 h and 19.54 h on June 24, 25 and 26, respectively. The sampling started at 17.00 h on June 26 and continued until 09.00 h on June 27; the subjects had just 2 full days to adjust to the long natural photoperiod. The highest ambient light intensity during the study was 104 000 lux. Saliva samples Ž3–5 ml. were collected without stimulation. The subjects were asked not to eat and drink during the preceding 30 min, to rince their mouth with water immediately before spitting directly into a plastic tube and to avoid coffee and usage of tooth paste throughout the whole sampling period. The samples were stored at y208C until required for assay, when they were defrosted, kept at room temperature for 30 min, vortexed and centrifuged at 3000 = g for 10 min at 48C; the supernatant was retained. Melatonin was measured by a direct radioimmunoassay w6x, validated for use in saliva w19,20x. w 3 HxMelatonin, specific activity 3.15 Tbqrmmol, was purchased from the Radiochemical Centre ŽAmersham, UK.. The antiserum, batch GrS 704-8483, was kindly provided by Dr. J. Arendt via Stockgrand Ltd., Department of Biochemistry, University of Surrey. Samples of 25 and 100 pg melatoninrml had interassay coefficients of variation of 13 and 9%, respectively. The limit of assay detection was 6 pgrml for 500 m l saliva assayed. The value of 0 pgrml was arbitrarily assigned to all baseline daytimes valued which were below the level of detection. Samples were assayed in duplicate. For construction of the calibration curve, an afternoon saliva of the corresponding subject sampled between 14.00 h and 17.00 h was used; no melatonin was ever found in any afternoon saliva sample assayed separately against the charcoal-stripped saliva standard curve. Cortisol was measured by a direct radioimmunoassay, validated for use in saliva w24x. Iodinated cortisol Žw 125 Ixiodohistaminyl-carboxymethyloxiiminocortisol. and rabbit antiserum against cortisol-3-CMObovine serum albumin were prepared by the Institute of Endocrinology and Biochemica ŽPrague., respectively. Diluted saliva samples, equivalent to 25 m l of saliva were incubated with 100 m l of antiserum Ždiluted 1:4000– 1:16 000. and labeled cortisol Žapproximately 15 000 c.p.m.rtube.. The measuring range of the method was 7–500 pg cortisol per tube. The intra-assay variability was 7–12% for the given range. Samples were assayed in duplicate and read against the calibration curve based on cortisol ŽSigma. w2x. Two phase markers of the melatonin rhythm were followed, namely the time of the evening melatonin rise and the time of morning melatonin decline. The rise was defined as the phase of melatonin rhythm at which the
168
D. VondrasoÕa ˇ ´ et al.r Brain Research 759 (1997) 166–170
salivary melatonin concentration reached a value of 12 pgrml during its evening increase. Similarly, the decline was defined as the phase of the melatonin rhythm at which the concentration reached the 12 pg level during its morning decrease. Alternatively, the times of the rise and decline were read at the level of 40% of the maximum nocturnal melatonin concentration for each subject. The time interval between the evening rise and the morning decline represents the duration of the melatonin signal. For the cortisol rhythm, just one phase marker was used, namely the time when cortisol during its morning rise reached the concentration of 3 ngrml of saliva. Alternatively, the time of the rise was read at the level of 50% of the maximum morning concentration. The paired t-test was employed for analysis of the difference between the winter and summer melatonin and cortisol rhythm characteristics, with a s 0.05 required for significance. Data for the winter and summer study are summarized in Table 1 and Fig. 2. The evening melatonin rise occurred at about the same time in summer as in winter. The morning melatonin decline in summer was, however, phase advanced by about 2 h as compared with the winter pattern. Consequently, the summer duration of the nocturnal melatonin signal of about 6 h was by 2 h shorter than the winter duration of about 8 h. The summer morning melatonin decline from the maximum level at 01.00 h occurred already at 03.00 h Ž P - 0.001., almost 1 h before sunrise, and was thus not triggered by the morning light. The morning cortisol rise in summer, similarly as the melatonin decline, was also phase advanced by about 2 h as compared with the winter pattern. The pacemaker controlling the morning melatonin decline and the cortisol rise was thus phase advanced by 2 h in summer as compared with the winter phase. A 1 h phase advance might be due to the effect of daylight saving time w26x. A further phase advance was rather due to the early morning bright light exposure than to earlier awakenings during the summer study, as the photoperiod appears to be a much more stronger entraining agent than the sleep-wake
Fig. 2. Salivary melatonin ŽA. and cortisol ŽB. rhythms in winter Žclosed circles. and summer Žopen circles.. Each point represents the mean" S.E.M. from 8 Žmelatonin. or 7 Žcortisol. subjects. Full Žwinter. and dashed Žsummer. arrows indicate times of sunset and sunrise, respectively.
cycle alone w7x. As the morning melatonin decline occurred spontaneously before exposure to bright ambient light, it was actually due to the phase advancing effect and not to the suppressive effect of the morning bright light w17x. In the evening, the situation was less clear. The melatonin rise occurred at about the same time in summer as in winter, around 21.30 h or little later. In summer, bright light in the evening until 20.00 h might have a phase
Table 1 Melatonin rise, decline and signal duration and cortisol rise in winter and summer
Time of winter melatonin rise Ž8. Time of summer melatonin rise Ž8. Time of winter melatonin decline Ž8. Time of summer melatonin decline Ž8. Winter melatonin signal duration Žh. Ž8. Summer melatonin signal duration Žh. Ž8. Time of winter cortisol rise Ž7. Time of summer cortisol rise Ž7.
A
B
21 h 48 min " 30 min 21 h 39 min " 12 min 6 h 18 min " 18 min 3 h 54 min " 12 min 8.1 " 0.6 6.2 " 0.4 a 5 h 0 min " 12 min 2 h 30 min " 12 min
21 h 54 min " 24 min 21 h 36 min " 12 min 6 h 00 min " 12 min 3 h 42 min " 6 min b 8.1 " 0.5 6.1 " 0.2 a 4 h 42 min " 12 min 2 h 42 min " 18 min b
b
b
Standard times of the evening melatonin rise, of the morning melatonin decline and cortisol rise and duration of the nocturnal melatonin signal were determined from individual winter and summer salivary melatonin and cortisol rhythm profiles Ždata not shown.. The values for melatonin were read either at the level of 12 pgrml saliva ŽA. or at the level of 40% of the maximum nocturnal concentration ŽB.. The values for cortisol were read either at the level of 3 ngrml saliva ŽA. or at the 50% of the maximum morning concentration ŽB.. Data are expressed as means" S.E.M.; figures in parentheses indicate the number of subjects. a P - 0.01, b P - 0.001, respectively, paired t-test, comparing values for winter with those for summer.
D. VondrasoÕa ˇ ´ et al.r Brain Research 759 (1997) 166–170
delaying effect on the melatonin rise and hence on an underlying evening component of a complex pacemaker controlling the rise w11x. The delaying effect could have been counterbalanced by a phase advancing effect of the early morning bright light on the morning pacemaker component controlling the melatonin decline via interaction of both components. Alternatively, the evening bright light might suppress the beginning of the nocturnal melatonin production w17x. Whatever is the case, the summer melatonin signal was markedly reduced as compared with the winter duration, due to the advanced melatonin decline and no change in the phase of the rise. Though in summer a phase advance of the morning melatonin decline relative to the winter pattern was reported in previous studies, no difference was found in the melatonin signal duration, due to a similar phase advance of the evening melatonin rise; in these studies, however, subjects were not exposed to the actual natural summer photoperiod from sunrise until sunset w4,10,12,13x. Also, our study did not confirm a summer–winter difference in the melatonin rhythm amplitude found in human pineal glands obtained at autopsy w8x. The winter melatonin pattern might not, however, reflect the actual short winter days. The photoperiod combined of artificial and natural light lasted 15–16 h. When such a photoperiod is shortened to 10 h and subjects are inactive during the remaining 14 h in darkness, the melatonin signal is extended by about 2 h as compared with the 16 h photoperiod signal w25,27x. It is therefore plausible to expect that difference between the summer and winter melatonin signal duration in humans exposed to actual summer and winter days at 508N latitude might be as long as 4 h. Such an exposure might have been normal, e.g. in the Middle Ages for our ancestors working in the field. In summer, they had to work from sunrise until sunset due to their unefficient and primitive tools whereas in winter they could sleep for a longer time w22x. These medieval people might be more affected by seasonal cycles and consequently be more photoperiodic than the present civilized society w21x. However, even nowadays humans can experience the true long summer days, e.g. during camping. The summer duration of the melatonin signal of about 6 h corresponds well with the reported duration in subjects exposed during the winter time to a 16 h bright light skeleton photoperiod, with 3 h bright light pulse in the evening and again in the morning w5x. Also, the 2 h summer phase advance of the morning melatonin decline and cortisol rise is in agreement with a similar phase advance found in a Finnish study at 608N Žlatitude; as only urban residents participated in the study, difference between the summer and winter melatonin signal duration was not significant w16x. The 2 day exposure to the natural summer photoperiod appears to be sufficient for adjustment to summer days, as the morning melatonin decline occurred spontaneously before sunrise. In animals, adjustment to a transition from a short to a long photoperiod occurs almost immediately, whereas adjustment to a transi-
169
tion from a long to a short photoperiod is gradual only and may take more days or even weeks w11x. Animal studies also indicate that not just the melatonin rhythm, but the controlling circadian pacemaking system itself located in the suprachiasmatic nucleus ŽSCN. of the hypothalamus is photoperiod dependent w23x. Even human studies suggest that the summer and winter SCNs may differ w9x. In conclusion, the present data show that exposure to an actual natural summer photoperiod from sunrise until sunset phase advances the circadian oscillator controlling the morning melatonin decline and cortisol rise and shortens the melatonin signal duration as compared with the winter pattern. Thus true summer days evoke a long photoperiod response even in humans.
Acknowledgements The authors are grateful to Mr. M. Klicpera for his excellent technical assistence. The work was supported by Grant Agency of the Academy of Sciences of the Czech Republic Grant A 7011604r1996 through contribution of PRO.MED.CS a.s.
References w1x J. Beck-Friis, D. Van Rosen, B.F. Kjellman, J.G. Ljunggren, L. Wetterberg, Melatonin in relation to body measures, sex, age, season and the use of drugs in patients with major affective disorders and healthy subjects, Psychoneuroendocrinology 10 Ž1985. 173–183. w2x M. Bickova, Comparison of isotope ˇ ´ R. Hampl, Z. Putz, L. Starka, ´ and non-isotope variants of direct cortisol immunoassay, in: S. Goro ŽEd.., Advances in Steroid Analysis, Academiai Kiado, ´ ´ Budapest, 1988, pp. 101–106. w3x D.B. Boivin, J.F. Duffy, R.E. Kronauer, C.A. Czeisler, Dose–response relationships for resetting of human circadian clock by light, Nature 379 Ž1996. 540–542. w4x J.J. Broadway, J. Arendt, S. Folkard, Bright light phase shifts the human melatonin rhythm during the Antarctic winter, Neurosci. Lett. 29 Ž1987. 185–189. w5x M. Buresova, ˇ ´ M. Dvorakova, ˇ´ ´ P. Zvolsky, ´ H. Illnerova, ´ Human circadian rhythm in serum melatonin in short winter days and in simulated artificial long days, Neurosci. Lett. 136 Ž1992. 173–176. w6x S. Frazer, P. Cowen, M. Franklin, Direct radioimmunoassay for melatonin in plasma, Clin. Chem. 29 Ž1983. 396–401. w7x T.M. Hoban, A.J. Lewy, R.L. Sack, C.M. Singer, The effects of shifting sleep two hours within a fixed photoperiod, J. Neural Transm. 85 Ž1991. 61–68. w8x M.A. Hofman, D.J. Skene, D.F. Swaab, Effect of photoperiod on the melatonin and 5-methoxytryptophol rhythms in the human pineal gland, Brain Res. 671 Ž1995. 254–260. w9x M.A. Hofman, D. Swaab, Seasonal changes in the suprachiasmatic nucleus of man, Neurosci. Lett. 139 Ž1992. 257–260. w10x K.S. Honma, S. Honma, M. Koshaka, N. Fukuda, Seasonal variation in the human circadian rhythm: dissociation between sleep and temperature rhythm, Am. J. Physiol. 262 Ž1992. R885–R891. w11x H. Illnerova, ´ The suprachiasmatic nucleus and the rhytmic pineal melatonin production, in: D.C. Klein, R.J. Moore, S.M. Reppert ŽEds.., Suprachiasmatic Nucleus: the Mind’s Clock, Oxford University Press, New York, 1991, pp. 197–216.
170
D. VondrasoÕa ˇ ´ et al.r Brain Research 759 (1997) 166–170
w12x H. Illnerova, The circadian rhythm in ´ P. Zvolsky, ´ J. Vanecek, ˇˇ plasma melatonin concentration of the urbanized man, the effect of summer and winter time, Brain Res. 328 Ž1985. 186–189. w13x D.J. Kennaway, P. Royles, Circadian rhythms of 6-sulphatoxymelatonin, cortisol and electrolyte excretion at the summer and winter solstices in normal men and women, Acta Endocrinol. ŽCopenh.. 113 Ž1986. 450–456. w14x D.C. Klein, R.J. Moore, S.M. Reppert ŽEds.., Suprachiasmatic Nucleus: the Mind’s Clock, Oxford University Press, New York, 1991. w15x M.-L. Laakso, T. Porkka-Heiskanen, A. Alila, D. Stenberg, D. Johansson, Correlation between salivary and serum melatonin: dependence on serum melatonin levels, J. Pineal Res. 9 Ž1990. 39–50. w16x M.-L. Laakso, T. Porkka-Heiskanen, A. Alila, D. Stenberg, D. Johansson, Twenty-four-hour rhythms in relation to the natural photoperiod: a field study in humans, J. Biol. Rhythms 9 Ž1994. 283–293. w17x A.J. Lewy, T.A. Wehr, F.K. Goodwin, D.A. Newsome, S.P. Markey, Light suppresses melatonin production in humans, Science 210 Ž1980. 1267–1269. w18x I. Makkison, J. Arendt, Melatonin secretion in humans on two different arctic bases Ž688 and 758S., J. Interdisc. Cycle Res. 22 Ž1991. 149–150. w19x A. Miles, D. Philbrick, S.F. Tidmarsh, D.M. Shaw, Direct radioimmunoassay for melatonin in saliva, Clin. Chem. 31 Ž1985. 1412– 1413.
w20x R. Nowak, I.C. McMillen, J. Redman, R.V. Short, The correlation between serum and salivary melatonin concentrations and urinary 6-hydroxymelatonin sulphate excretion rates: two non-invasive techniques for monitoring human circadian rhythmicity, Clin. Endocrinol. ŽCopenh.. 27 Ž1987. 445–452. w21x T. Roenneberg, J. Aschoff, Annual rhythm of human reproduction II. Environmental correlations, J. Biol. Rhythms 5 Ž1990. 217–239. w22x Z. Smetanka, Z., Legenda o Ostojovi, Mlada´ Fronta, Prague, 1992, ´ 290 pp. w23x A. Sumova, ´ Y. Travnıckova, ´ ´ˇ ´ R. Peters, W.J. Schwartz, H. Illnerova, ´ The rat suprachiasmatic nucleus is a clock for all seasons, Proc. Natl. Acad. Sci. USA 92 Ž1995. 7754–7758. w24x T. Umeda, R. Hiramatsu, T. Iwaoka, T. Shimada, F. Miura, T. Sato, Use of saliva for monitoring unbound free cortisol levels in serum, Clin. Chim. Acta 110 Ž1981. 245–253. w25x T.A. Wehr, The durations of human melatonin secretion and sleep respond to changes in daylength Žphotoperiod., J. Clin. Endocrinol. Metab. 73 Ž1991. 1276–1280. w26x T.A. Wehr, H.A. Giesen, D.E. Moul, E.H. Turner, P.J. Schwartz, Suppression of men’s responses to seasonal changes in day length by modern artificial lighting, Am. J. Physiol. 269 Ž1995. R173–R178. w27x T.A. Wehr, D.E. Moul, G. Barbato, H.A. Giesen, J.A. Seidel, C. Barker, C. Bender, Conservation of photoperiod-response mechanisms in humans, Am. J. Physiol. 265 Ž1993. R846–R857.
Short communication
Exposure to long summer days affects the human melatonin and cortisol rhythms Dana Vondrasova, Helena Illnerova´ ˇ ´ Ivan Hajek, ´
)
Institute of Physiology, Academy of Sciences of the Czech Republic, Vıdenska ´ ˇ ´ 1083, 14220 Prague 4, Czech Republic Accepted 11 March 1997
Abstract 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. Keywords: Circadian rhythm; Photoperiod; Melatonin; Cortisol; Human
In all mammals studied so far, the waveform of the circadian rhythm in melatonin production and secretion depends on day length, i.e., on photoperiod w11,14x. On long summer days, light intruding into the late evening hour phase delays the evening melatonin production onset, light intruding into the early morning hour phase advances the morning production offset; consequently, the melatonin signal is shortened. On short winter days, the signal is extended w11x. Hence the melatonin signal duration may provide mammals with information about the season of the year. In humans, the melatonin signal duration appears to depend just slightly on the ambient photoperiod, and only at higher latitudes, e.g. at 608N in Sweden w1x and 688S in Antarctica w18x. However, in temperate zones, e.g. at a latitude 358S w13x, 398N w26x and 508N w12x, respectively, no difference between the summer and winter melatonin signal duration has been found, though in summer the natural photoperiod may last even 16 h and in winter it may be only 8 h. A hypothesis has been put forward that even during the winter time the actual photoperiod may be as long as 16 h, as humans usually spend 8 hrday in darkness during their nighttime sleep and otherwise are exposed to artificial and ambient natural light for 16 h w27x. Therefore, no difference in the melatonin signal duration between the summer and winter time can be expected.
)
Corresponding author. Fax: q42 Ž2. 471-9517.
Indeed, when human subjects experience a lighting regime with just 10 h of light and 14 h of darkness ŽLD 10:14 h. with forced inactivity during the dark period, the melatonin signal duration is extended by about 2 h as compared with the pattern under the artificial and natural light–dark regime ŽLD 16:8. w25,27x. All human subjects studied so far were students and employees who were not exposed to the whole long natural photoperiod during the summer time. Intensity of light under the natural summer photoperiod may be by more than 2 orders higher than intensity under an artificial lighting regime. As the extent of the photic entrainment of the human pacemaker controlling circadian rhythms depends on light intensity w3x, it may be expected that the actual ambient summer photoperiod might shape the melatonin signal more markedly than the combined artificial and natural 16 h photoperiod, the so-called ‘‘short night’’ regime w27x. The present study was undertaken to find out whether exposure to the whole natural summer photoperiod would shorten the melatonin signal and reset the melatonin rhythm relative to the winter ‘‘LD 16:8’’ pattern. Another circadian rhythm controlled by the same pacemaker as the rhythmic melatonin production, namely the rhythm in cortisol secretion w14x, was followed as well, in order to compare phase shifts of both the circadian rhythms. To be able to study the rhythms under field conditions in the country, a non-invasive sampling method was used, i.e., melatonin and cortisol were assayed in the
0006-8993r97r$17.00 Copyright q 1997 Elsevier Science B.V. All rights reserved. PII S 0 0 0 6 - 8 9 9 3 Ž 9 7 . 0 0 3 5 8 - 2
D. VondrasoÕa ˇ ´ et al.r Brain Research 759 (1997) 166–170
saliva. Salivary melatonin and cortisol levels have been reported to be highly correlated with serum levels w15,19,20,24x. Eight high-school students, 4 boys and 4 girls aged 15 years, participated in the study. All the volunteers were high melatonin secretors, with peak values higher than 25 pg of melatoninrml of saliva. The subjects gave informed consent after the nature of the study was fully explained. Both the winter and the summer parts of the study were performed during the school year. The students were synchronized to a similar schedule: they slept daily from 22.00 to 23.00 h until 07.00 h of the local time and were thus exposed to artificial and natural light for 15–16 hrday. During winter, the local time was the same as the standard time, whereas during summer the daylight saving time was introduced and hence the local time was 1 h ahead the standard time. All subsequent data are given in the standard time. Accordingly, before the start of the summer experiment the students slept daily from 21.00 to 22.00 until 06.00 of the standard time. The study was performed near Prague, at 508N latitude. The winter part of the study was performed from January 14 to January 15 ŽFig. 1.. Sunrise occurred at 07.54 h, sunset at 16.24 h; consequently, the ambient natural photoperiod lasted just 8.5 h. The students gathered in the afternoon of January 14 in the house of D.V. Since 16.00 h, they were sampled in 1–2 h intervals until 10.00 h of the next day when they left. They slept in darkness from 22.00 h until 07.00 h; otherwise, they experienced light of an intensity lower than 100 lux. The summer part of the study was performed from June 24 until June 27. At that time, the sunrise occurred at 03.52 h and sunset at 20.13 h; consequently, the natural photoperiod lasted 16.3 h. The students gathered in the afternoon of June 24 in a cabin in the country. From June 24 until June 27, they slept from 21.00 h until 04.00 in darkness and were outdoors until
Fig. 1. Experimental paradigm. Black bars represent sleep in darkness, cross-hatched bars intervals when individual subjects started to sleep, hatched bars light periods of intensity lower than 100 lux and open bars bright light periods. Open triangles indicate times of sunset and sunrise, respectively, arrows the sampling times and double lines times when the outdoor light intensity fell bellow 2000 lux in the evening, and rose above 2000 lux in the morning, respectively.
167
20.00 in the evening and again from 04.15 h in the morning; the intensity of the indoor light between 20.00 h and 21.00 h and between 04.00 h and 04.15 h, respectively, was lower than 100 lux. In the morning, the intensity of the outdoor light exceeded 2000 lux at 04.18 h, 04.15 h and 04.40 h on June 25, 26 and 27, respectively. In the evening, the intensity fell below 2000 lux at 19.45 h, 19.45 h and 19.54 h on June 24, 25 and 26, respectively. The sampling started at 17.00 h on June 26 and continued until 09.00 h on June 27; the subjects had just 2 full days to adjust to the long natural photoperiod. The highest ambient light intensity during the study was 104 000 lux. Saliva samples Ž3–5 ml. were collected without stimulation. The subjects were asked not to eat and drink during the preceding 30 min, to rince their mouth with water immediately before spitting directly into a plastic tube and to avoid coffee and usage of tooth paste throughout the whole sampling period. The samples were stored at y208C until required for assay, when they were defrosted, kept at room temperature for 30 min, vortexed and centrifuged at 3000 = g for 10 min at 48C; the supernatant was retained. Melatonin was measured by a direct radioimmunoassay w6x, validated for use in saliva w19,20x. w 3 HxMelatonin, specific activity 3.15 Tbqrmmol, was purchased from the Radiochemical Centre ŽAmersham, UK.. The antiserum, batch GrS 704-8483, was kindly provided by Dr. J. Arendt via Stockgrand Ltd., Department of Biochemistry, University of Surrey. Samples of 25 and 100 pg melatoninrml had interassay coefficients of variation of 13 and 9%, respectively. The limit of assay detection was 6 pgrml for 500 m l saliva assayed. The value of 0 pgrml was arbitrarily assigned to all baseline daytimes valued which were below the level of detection. Samples were assayed in duplicate. For construction of the calibration curve, an afternoon saliva of the corresponding subject sampled between 14.00 h and 17.00 h was used; no melatonin was ever found in any afternoon saliva sample assayed separately against the charcoal-stripped saliva standard curve. Cortisol was measured by a direct radioimmunoassay, validated for use in saliva w24x. Iodinated cortisol Žw 125 Ixiodohistaminyl-carboxymethyloxiiminocortisol. and rabbit antiserum against cortisol-3-CMObovine serum albumin were prepared by the Institute of Endocrinology and Biochemica ŽPrague., respectively. Diluted saliva samples, equivalent to 25 m l of saliva were incubated with 100 m l of antiserum Ždiluted 1:4000– 1:16 000. and labeled cortisol Žapproximately 15 000 c.p.m.rtube.. The measuring range of the method was 7–500 pg cortisol per tube. The intra-assay variability was 7–12% for the given range. Samples were assayed in duplicate and read against the calibration curve based on cortisol ŽSigma. w2x. Two phase markers of the melatonin rhythm were followed, namely the time of the evening melatonin rise and the time of morning melatonin decline. The rise was defined as the phase of melatonin rhythm at which the
168
D. VondrasoÕa ˇ ´ et al.r Brain Research 759 (1997) 166–170
salivary melatonin concentration reached a value of 12 pgrml during its evening increase. Similarly, the decline was defined as the phase of the melatonin rhythm at which the concentration reached the 12 pg level during its morning decrease. Alternatively, the times of the rise and decline were read at the level of 40% of the maximum nocturnal melatonin concentration for each subject. The time interval between the evening rise and the morning decline represents the duration of the melatonin signal. For the cortisol rhythm, just one phase marker was used, namely the time when cortisol during its morning rise reached the concentration of 3 ngrml of saliva. Alternatively, the time of the rise was read at the level of 50% of the maximum morning concentration. The paired t-test was employed for analysis of the difference between the winter and summer melatonin and cortisol rhythm characteristics, with a s 0.05 required for significance. Data for the winter and summer study are summarized in Table 1 and Fig. 2. The evening melatonin rise occurred at about the same time in summer as in winter. The morning melatonin decline in summer was, however, phase advanced by about 2 h as compared with the winter pattern. Consequently, the summer duration of the nocturnal melatonin signal of about 6 h was by 2 h shorter than the winter duration of about 8 h. The summer morning melatonin decline from the maximum level at 01.00 h occurred already at 03.00 h Ž P - 0.001., almost 1 h before sunrise, and was thus not triggered by the morning light. The morning cortisol rise in summer, similarly as the melatonin decline, was also phase advanced by about 2 h as compared with the winter pattern. The pacemaker controlling the morning melatonin decline and the cortisol rise was thus phase advanced by 2 h in summer as compared with the winter phase. A 1 h phase advance might be due to the effect of daylight saving time w26x. A further phase advance was rather due to the early morning bright light exposure than to earlier awakenings during the summer study, as the photoperiod appears to be a much more stronger entraining agent than the sleep-wake
Fig. 2. Salivary melatonin ŽA. and cortisol ŽB. rhythms in winter Žclosed circles. and summer Žopen circles.. Each point represents the mean" S.E.M. from 8 Žmelatonin. or 7 Žcortisol. subjects. Full Žwinter. and dashed Žsummer. arrows indicate times of sunset and sunrise, respectively.
cycle alone w7x. As the morning melatonin decline occurred spontaneously before exposure to bright ambient light, it was actually due to the phase advancing effect and not to the suppressive effect of the morning bright light w17x. In the evening, the situation was less clear. The melatonin rise occurred at about the same time in summer as in winter, around 21.30 h or little later. In summer, bright light in the evening until 20.00 h might have a phase
Table 1 Melatonin rise, decline and signal duration and cortisol rise in winter and summer
Time of winter melatonin rise Ž8. Time of summer melatonin rise Ž8. Time of winter melatonin decline Ž8. Time of summer melatonin decline Ž8. Winter melatonin signal duration Žh. Ž8. Summer melatonin signal duration Žh. Ž8. Time of winter cortisol rise Ž7. Time of summer cortisol rise Ž7.
A
B
21 h 48 min " 30 min 21 h 39 min " 12 min 6 h 18 min " 18 min 3 h 54 min " 12 min 8.1 " 0.6 6.2 " 0.4 a 5 h 0 min " 12 min 2 h 30 min " 12 min
21 h 54 min " 24 min 21 h 36 min " 12 min 6 h 00 min " 12 min 3 h 42 min " 6 min b 8.1 " 0.5 6.1 " 0.2 a 4 h 42 min " 12 min 2 h 42 min " 18 min b
b
b
Standard times of the evening melatonin rise, of the morning melatonin decline and cortisol rise and duration of the nocturnal melatonin signal were determined from individual winter and summer salivary melatonin and cortisol rhythm profiles Ždata not shown.. The values for melatonin were read either at the level of 12 pgrml saliva ŽA. or at the level of 40% of the maximum nocturnal concentration ŽB.. The values for cortisol were read either at the level of 3 ngrml saliva ŽA. or at the 50% of the maximum morning concentration ŽB.. Data are expressed as means" S.E.M.; figures in parentheses indicate the number of subjects. a P - 0.01, b P - 0.001, respectively, paired t-test, comparing values for winter with those for summer.
D. VondrasoÕa ˇ ´ et al.r Brain Research 759 (1997) 166–170
delaying effect on the melatonin rise and hence on an underlying evening component of a complex pacemaker controlling the rise w11x. The delaying effect could have been counterbalanced by a phase advancing effect of the early morning bright light on the morning pacemaker component controlling the melatonin decline via interaction of both components. Alternatively, the evening bright light might suppress the beginning of the nocturnal melatonin production w17x. Whatever is the case, the summer melatonin signal was markedly reduced as compared with the winter duration, due to the advanced melatonin decline and no change in the phase of the rise. Though in summer a phase advance of the morning melatonin decline relative to the winter pattern was reported in previous studies, no difference was found in the melatonin signal duration, due to a similar phase advance of the evening melatonin rise; in these studies, however, subjects were not exposed to the actual natural summer photoperiod from sunrise until sunset w4,10,12,13x. Also, our study did not confirm a summer–winter difference in the melatonin rhythm amplitude found in human pineal glands obtained at autopsy w8x. The winter melatonin pattern might not, however, reflect the actual short winter days. The photoperiod combined of artificial and natural light lasted 15–16 h. When such a photoperiod is shortened to 10 h and subjects are inactive during the remaining 14 h in darkness, the melatonin signal is extended by about 2 h as compared with the 16 h photoperiod signal w25,27x. It is therefore plausible to expect that difference between the summer and winter melatonin signal duration in humans exposed to actual summer and winter days at 508N latitude might be as long as 4 h. Such an exposure might have been normal, e.g. in the Middle Ages for our ancestors working in the field. In summer, they had to work from sunrise until sunset due to their unefficient and primitive tools whereas in winter they could sleep for a longer time w22x. These medieval people might be more affected by seasonal cycles and consequently be more photoperiodic than the present civilized society w21x. However, even nowadays humans can experience the true long summer days, e.g. during camping. The summer duration of the melatonin signal of about 6 h corresponds well with the reported duration in subjects exposed during the winter time to a 16 h bright light skeleton photoperiod, with 3 h bright light pulse in the evening and again in the morning w5x. Also, the 2 h summer phase advance of the morning melatonin decline and cortisol rise is in agreement with a similar phase advance found in a Finnish study at 608N Žlatitude; as only urban residents participated in the study, difference between the summer and winter melatonin signal duration was not significant w16x. The 2 day exposure to the natural summer photoperiod appears to be sufficient for adjustment to summer days, as the morning melatonin decline occurred spontaneously before sunrise. In animals, adjustment to a transition from a short to a long photoperiod occurs almost immediately, whereas adjustment to a transi-
169
tion from a long to a short photoperiod is gradual only and may take more days or even weeks w11x. Animal studies also indicate that not just the melatonin rhythm, but the controlling circadian pacemaking system itself located in the suprachiasmatic nucleus ŽSCN. of the hypothalamus is photoperiod dependent w23x. Even human studies suggest that the summer and winter SCNs may differ w9x. In conclusion, the present data show that exposure to an actual natural summer photoperiod from sunrise until sunset phase advances the circadian oscillator controlling the morning melatonin decline and cortisol rise and shortens the melatonin signal duration as compared with the winter pattern. Thus true summer days evoke a long photoperiod response even in humans.
Acknowledgements The authors are grateful to Mr. M. Klicpera for his excellent technical assistence. The work was supported by Grant Agency of the Academy of Sciences of the Czech Republic Grant A 7011604r1996 through contribution of PRO.MED.CS a.s.
References w1x J. Beck-Friis, D. Van Rosen, B.F. Kjellman, J.G. Ljunggren, L. Wetterberg, Melatonin in relation to body measures, sex, age, season and the use of drugs in patients with major affective disorders and healthy subjects, Psychoneuroendocrinology 10 Ž1985. 173–183. w2x M. Bickova, Comparison of isotope ˇ ´ R. Hampl, Z. Putz, L. Starka, ´ and non-isotope variants of direct cortisol immunoassay, in: S. Goro ŽEd.., Advances in Steroid Analysis, Academiai Kiado, ´ ´ Budapest, 1988, pp. 101–106. w3x D.B. Boivin, J.F. Duffy, R.E. Kronauer, C.A. Czeisler, Dose–response relationships for resetting of human circadian clock by light, Nature 379 Ž1996. 540–542. w4x J.J. Broadway, J. Arendt, S. Folkard, Bright light phase shifts the human melatonin rhythm during the Antarctic winter, Neurosci. Lett. 29 Ž1987. 185–189. w5x M. Buresova, ˇ ´ M. Dvorakova, ˇ´ ´ P. Zvolsky, ´ H. Illnerova, ´ Human circadian rhythm in serum melatonin in short winter days and in simulated artificial long days, Neurosci. Lett. 136 Ž1992. 173–176. w6x S. Frazer, P. Cowen, M. Franklin, Direct radioimmunoassay for melatonin in plasma, Clin. Chem. 29 Ž1983. 396–401. w7x T.M. Hoban, A.J. Lewy, R.L. Sack, C.M. Singer, The effects of shifting sleep two hours within a fixed photoperiod, J. Neural Transm. 85 Ž1991. 61–68. w8x M.A. Hofman, D.J. Skene, D.F. Swaab, Effect of photoperiod on the melatonin and 5-methoxytryptophol rhythms in the human pineal gland, Brain Res. 671 Ž1995. 254–260. w9x M.A. Hofman, D. Swaab, Seasonal changes in the suprachiasmatic nucleus of man, Neurosci. Lett. 139 Ž1992. 257–260. w10x K.S. Honma, S. Honma, M. Koshaka, N. Fukuda, Seasonal variation in the human circadian rhythm: dissociation between sleep and temperature rhythm, Am. J. Physiol. 262 Ž1992. R885–R891. w11x H. Illnerova, ´ The suprachiasmatic nucleus and the rhytmic pineal melatonin production, in: D.C. Klein, R.J. Moore, S.M. Reppert ŽEds.., Suprachiasmatic Nucleus: the Mind’s Clock, Oxford University Press, New York, 1991, pp. 197–216.
170
D. VondrasoÕa ˇ ´ et al.r Brain Research 759 (1997) 166–170
w12x H. Illnerova, The circadian rhythm in ´ P. Zvolsky, ´ J. Vanecek, ˇˇ plasma melatonin concentration of the urbanized man, the effect of summer and winter time, Brain Res. 328 Ž1985. 186–189. w13x D.J. Kennaway, P. Royles, Circadian rhythms of 6-sulphatoxymelatonin, cortisol and electrolyte excretion at the summer and winter solstices in normal men and women, Acta Endocrinol. ŽCopenh.. 113 Ž1986. 450–456. w14x D.C. Klein, R.J. Moore, S.M. Reppert ŽEds.., Suprachiasmatic Nucleus: the Mind’s Clock, Oxford University Press, New York, 1991. w15x M.-L. Laakso, T. Porkka-Heiskanen, A. Alila, D. Stenberg, D. Johansson, Correlation between salivary and serum melatonin: dependence on serum melatonin levels, J. Pineal Res. 9 Ž1990. 39–50. w16x M.-L. Laakso, T. Porkka-Heiskanen, A. Alila, D. Stenberg, D. Johansson, Twenty-four-hour rhythms in relation to the natural photoperiod: a field study in humans, J. Biol. Rhythms 9 Ž1994. 283–293. w17x A.J. Lewy, T.A. Wehr, F.K. Goodwin, D.A. Newsome, S.P. Markey, Light suppresses melatonin production in humans, Science 210 Ž1980. 1267–1269. w18x I. Makkison, J. Arendt, Melatonin secretion in humans on two different arctic bases Ž688 and 758S., J. Interdisc. Cycle Res. 22 Ž1991. 149–150. w19x A. Miles, D. Philbrick, S.F. Tidmarsh, D.M. Shaw, Direct radioimmunoassay for melatonin in saliva, Clin. Chem. 31 Ž1985. 1412– 1413.
w20x R. Nowak, I.C. McMillen, J. Redman, R.V. Short, The correlation between serum and salivary melatonin concentrations and urinary 6-hydroxymelatonin sulphate excretion rates: two non-invasive techniques for monitoring human circadian rhythmicity, Clin. Endocrinol. ŽCopenh.. 27 Ž1987. 445–452. w21x T. Roenneberg, J. Aschoff, Annual rhythm of human reproduction II. Environmental correlations, J. Biol. Rhythms 5 Ž1990. 217–239. w22x Z. Smetanka, Z., Legenda o Ostojovi, Mlada´ Fronta, Prague, 1992, ´ 290 pp. w23x A. Sumova, ´ Y. Travnıckova, ´ ´ˇ ´ R. Peters, W.J. Schwartz, H. Illnerova, ´ The rat suprachiasmatic nucleus is a clock for all seasons, Proc. Natl. Acad. Sci. USA 92 Ž1995. 7754–7758. w24x T. Umeda, R. Hiramatsu, T. Iwaoka, T. Shimada, F. Miura, T. Sato, Use of saliva for monitoring unbound free cortisol levels in serum, Clin. Chim. Acta 110 Ž1981. 245–253. w25x T.A. Wehr, The durations of human melatonin secretion and sleep respond to changes in daylength Žphotoperiod., J. Clin. Endocrinol. Metab. 73 Ž1991. 1276–1280. w26x T.A. Wehr, H.A. Giesen, D.E. Moul, E.H. Turner, P.J. Schwartz, Suppression of men’s responses to seasonal changes in day length by modern artificial lighting, Am. J. Physiol. 269 Ž1995. R173–R178. w27x T.A. Wehr, D.E. Moul, G. Barbato, H.A. Giesen, J.A. Seidel, C. Barker, C. Bender, Conservation of photoperiod-response mechanisms in humans, Am. J. Physiol. 265 Ž1993. R846–R857.