Contents lists available at SciVerse ScienceDirect
Hormones and Behavior
j ourna l homepage: www.e lsev ie r .com/ locate /yhbeh
Functional similarity in relation to the external environment between circadianbehavioral and melatonin rhythms in the subtropical Indian weaver bird
Jyoti Singh a, Sangeeta Rani b, Vinod Kumar a,⁎a DST-IRHPA Center for Excellence in Biological Rhythm Research, University of Delhi, Delhi 110007, Indiab DST-IRHPA Center for Excellence in Biological Rhythm Research, University of Lucknow, Lucknow 226007, India
The present study investigated whether the circadian oscillators controlling rhythms in activity behaviorand melatonin secretion shared similar functional relationship with the external environment. We simul-taneously measured the effects of varying illuminations on rhythms of movement and melatonin levels inIndian weaver birds under synchronized (experiment 1) and freerunning (experiment 2) light conditions.In experiment 1, weaverbirds were exposed to 12 h light: 12 h darkness (12L:12D; L=20 lx, D=0.1 lx)for 2.5 weeks. Then, the illumination of the dark period was sequentially enhanced to 1-, 5-, 10-, 20-and 100 lx at the intervals of about 2 to 4 weeks. In experiment 2, weaver birds similarly exposed for2.5 weeks to 12L:12D (L=100 lx; D=0.1 lx) were released in constant dim light (LLdim, 0.1 lx) for6 weeks. Thereafter, LLdim illumination was sequentially enhanced to 1-, 3- and 5 lx at the intervals ofabout 2 weeks. Whereas the activity of singly housed individuals was continuously recorded, the plasmamelatonin levels were measured at two time of the day, once in each light condition. The circadian out-puts in activity and melatonin were phase coupled with an inverse phase relationship: melatonin levelswere low during the active phase (light period) and high during the inactive phase (dark period). Thisphase relationship continued in both the synchronized and freerunning states as long as circadian activityand melatonin oscillators subjectively interpreted synchronously the daily light environment, based on il-lumination intensity and/or photophase contrast, as the times of day and night. There were dissociationsbetween the response of the activity rhythms and melatonin rhythms in light conditions when the con-trast between day and night was much reduced (20:10 lx) or became equal. We suggest that circadian os-cillators governing activity behavior and melatonin secretion in weaverbirds are phase coupled, but theyseem to independently respond to environmental cues. This would probably explain the varying degree towhich the involvement of pineal/melatonin in regulation of circadian behaviors has been found amongdifferent birds.
The circadian pacemaker system in birds is comprised of at leastthree anatomically separated components that lie in the pinealorgan, anterior hypothalamus and retinae of the eyes (Cassone andMenaker, 1984; Gwinner et al., 1997; Kumar et al., 2004). All thesecircadian components contain self sustained clocks, which functionindependently (Gwinner et al., 1997; Kumar and Singh, 2006;Kumar et al., 2004). Among these, the pineal clock is the most widelyinvestigated. The known circadian output from pineal is the melato-nin (Karaganis et al., 2008; Zatz et al., 1988), which has low levelsduring the day and high at night, irrespective of the fact the species
).
rights reserved.
is diurnal, crepuscular or nocturnal (Brandstätter et al., 2000, 2001;Kumar and Follett, 1993; Zatz et al., 1988).
The pineal and melatonin seem to be involved in the circadian or-ganization of activity behavior (Kumar, 2001; Menaker andZimmermann, 1976). The removal of the pineal (pinealectomy)leads to arrhythmic activity behavior in songbirds, including thehouse sparrow (Passer domesticus, Gaston, 1971), white crownedsparrow (Zonotrichia leucophrys gambelii, Gaston, 1971; Gaston andMenaker, 1968), white throated sparrow, (Zonotrichia albicollis,McMillan, 1972), java sparrow (Padda oryzivora, Ebihara andKawamura, 1981), house finch (Carpodacus mexicanus, Fuchs, 1983),Indianweaver bird (Ploceus philippinus,Rani et al., 2005) and redhead-ed bunting (Emberiza bruniceps, Singh et al., unpublished obs.). Also,the transplantation of pineal from another individual restores rhyth-micity in the pinealectomized house sparrow (Zimmerman andMenaker, 1979). Further, melatonin given in injections, infusions ordrinking water restores circadian rhythmicity in the arrhythmic
528 J. Singh et al. / Hormones and Behavior 61 (2012) 527–534
birds (Cassone et al., 2008; Chabot and Menaker, 1992; Gwinner,1989; Turek et al., 1976; Singh et al., unpublished obs.).
However, the avian pineal is not equivalent of the mammaliansuprachiasmatic nucleus (SCN), the removal of which causes immedi-ate loss of circadian rhythms in activity and many other functions inall species of mammals which have been investigated (Golombekand Rosentein, 2010; Klein et al., 1991). The role of pineal in avian cir-cadian organization is quite variable (Kumar, 2001). For example, thepinealectomy does not affect circadian activity rhythms in pigeon(Columba livia; Ebihara et al., 1984), Japanese quail (Coturnix coturnixjaponica; Simpson and Follett, 1981) and chicken (Gallus domesticus;McBride, 1973). Among songbirds, the pinealectomy less severely af-fects the circadian activity rhythms in the European starling (Sturnusvulgaris; Gwinner, 1978). In those songbird species, in which the pi-nealectomy results in the abolition or drastic disruption of the circa-dian activity rhythms (see above), the effects are not theimmediate, but gradual; the loss of circadian rhythmicity occurs atleast 4–5 cycles after the pinealectomy (Gaston and Menaker, 1968;Gwinner et al., 1997; Kumar, 2001; Kumar et al., 2004; Rani et al.,2005). Also, pineal transplant restores the phase and not the periodof the donor's circadian rhythms (Zimmerman and Menaker, 1979)and pinealectomized birds can be entrained to light–dark cycles(Takahashi and Menaker, 1982). Further, the restored circadianrhythms in the pineal transplanted or melatonin treated individualsare loosely synchronized (see Cassone et al., 2008; Gwinner et al.,1997; Zimmerman and Menaker, 1979). Therefore, there is at leastone other circadian oscillator of the avian circadian pacemaker sys-tem in interaction with the pineal generates the circadian rhythmsin activity (Kumar et al., 2004; Takahashi and Menaker, 1982). Inthe absence of this interaction in pinealectomized individuals, thereis gradual loss of the circadian rhythmicity. The circadian melatoninoutput from the pineal seems to serve as input to the avian circadianpacemaker system, and provides phase to the circadian behaviors, in-cluding the activity rhythm (Kumar et al., 2004).
Therefore, a key question to answer is whether the circadian oscil-lators governing activity behavior and melatonin secretion in song-birds share similar functional relationship with the externalenvironment, e.g. the light environment. If yes, then the circadianoutputs from these oscillators will exhibit parallels under imposedlight conditions, including those conditions that will enforce the dis-ruption and abolition of the circadian rhythmicity (Aschoff, 1981;Aschoff and von Goetz, 1989; Binkley et al., 1972; Gwinner et al.,1987; Kumar, 2001; Kumar et al., 1992, 2000, 2007a). Alternatively,if the circadian oscillators governing activity behavior and melatoninsecretion independently respond to environmental cues, weaverbirdswill show dissociations between the response of the activity rhythmsand melatonin rhythms under the weak light zeitgeber conditions, i.e.when the contrast between light and dark periods is much reduced,and under the constant light conditions (LL) at certain intensitylevels.
Hence, the goal of the present study was to investigate similarityin functional relationship of circadian rhythms of activity and melato-nin secretion with the external environment between in a photoperi-odic species, the Indian weaverbird (Ploceus philippinus) which whenpinealectomized exhibits loss of circadian rhythms in activity (Rani etal., 2005). To achieve this goal, we employed an experimental para-digm that has previously been employed to probe the subjective in-terpretation of the day and night, based on light intensity andphotophase contrasts, and the persistence of circadian rhythms inJapanese quail (Meyer and Millam, 1991; Meyer et al., 1988) andblackheaded bunting (Emberiza melanocephala, Kumar et al., 1992).We subjected weaverbirds to 12 h light: 12 h darkness (12L:12D) orto constant dim light (LLdim) with changing illuminations at about2 week intervals. The activity of each bird was continuously recorded,and plasma melatonin levels were measured at two times of the day,once in each light condition.
Materials and methods
Animals and housing
Adult male Indian weaver birds (Ploceus philippinus) were cap-tured and kept in an outdoor aviary until used in this study. Theywere put individually in activity cages (size=60×35×45 cm) thatwere placed inside temperature regulated (24±2 °C) lightproofwooden chambers (size=75×50×70 cm), illuminated by compactfluorescent lamps (CFL; 14 W, 230 V, Phillips, India) providing 12 hlight and 12 h darkness (12L:12D). The desired light intensity wasobtained by covering the CFL by the strips of a black paper sheet.The light intensity was measured at the floor of the cage. Food (grainsof Setaria italica and Oryza sativa) and water were available ad libi-tum, and replenished during the light phase.
Experiments
The experiments were carried out in accordance with the guide-lines of the Institutional Ethics Committee applied at the Departmentof Zoology, University of Lucknow, where the experiments were car-ried out. Beginning on 24 November 2007, two experiments com-pared the effects of light conditions on activity and plasmamelatonin rhythms. Experiment 1 examined the effects of steadilychanging illumination contrasts between the light and dark periodsin an equinox photoperiod. Birds (n=6) were initially exposed to12L:12D (20:0.1 lx). Since the illumination contrast determines thetimes of day and night (Kumar et al., 1992), the first 12 h periodwith high illumination (zeitgeber time, ZT, 0–12; ZT 0=light onset)was called day and the later 12 h with low illumination (ZT 12–24)was called the night. Beginning on day 18, the dark period illumina-tion was sequentially enhanced to 1-, 5-, 10-, 20- and 100 lx at the in-tervals of about 2 to 4 weeks. This resulted in 24 h day as constantlight condition (LL) during the 20:20 lx, and reversed the times ofday and night during the 20:100 lx exposure.
Experiment 2 (n=7) used a different protocol. Birds initially ex-posed to 12L:12D (100:0.1 lx) for about 3 weeks, were released inconstant dim light (LLdim, 0.1 lx) for 6 weeks. Thereafter, the LLdim il-lumination was sequentially enhanced to 1-, 3- and 5 lx at the inter-vals of about 2 or 2.5 weeks.
Measurement of activity and melatonin profiles
Activity of each individual was monitored using an infra-red sen-sor system that detected the movement of a bird within the cage. Itwas recorded in 5-min bins and stored on separate channels of acomputerized data-logging system using The Chronobiology Kit,from Stanford Software System, USA. Computerized activity records(actograms) were obtained for each bird, as described repeatedly(Malik et al., 2004; Singh et al., 2011).
To examine daily melatonin profile under different light condi-tions, blood samples were taken during the middle of the day (ZT0–12) and night (ZT 12–24) in each experiment at regular intervalsrepresenting each light condition. Thus, the timing of blood samplingwas centered on ZT 6 and ZT 18, as defined by the LD cycle in the be-ginning of the experiment. All birds were bled at both time pointswith an interval of 1 day. When birds received LL conditions, exceptLLdim at 0.1 lx, the times corresponding to ZT 0 and ZT 12 of the ini-tial LD condition were still taken as reference points for describing theactivity and melatonin measurements. However, in LLdim at 0.1 lx(experiment 2), when birds freeran with period (tau, τ) less than24 h (τb24 h), the onset of activity was considered correspondingto the ZT 0, and referred to as the circadian time zero (CT 0). Hence,CT 0–12 constituted the “subjective” day, and CT 12–24 as the “sub-jective” night. In this condition, blood sampling scattered over a peri-od of one week in between days 10 and 17 of the LLdim, was done atCT 6 and CT 18 of each individual.
529J. Singh et al. / Hormones and Behavior 61 (2012) 527–534
Samples of about 100–125 μl of blood were collected in heparin-ized capillary tubes by puncturing the wing vein and immediatelycentrifuged. The plasma was stored at −20 °C until assayed for mel-atonin. The ELISA of plasma melatonin was done using a specific kitfor melatonin (product no. RE54021) from IBL International GmbH,Hamburg, Germany. This assay has been validated and used for themeasurement of plasma melatonin in several studies (Lahiri et al.,2004; Terzieva et al., 2009). However, we did not perform validationtests of this assay for our species. Briefly, this involved the extractionof samples along with standards and controls using methanol. Theextracted volume was dried, re-dissolved in double distilled water.Then, the 50 μl of this solution was incubated with 50 μl melatoninanti-serum (rabbit, polyclonal) for 20 h at 4 °C. Subsequently, 150 μlof freshly prepared enzyme conjugate (anti-biotin antibodies raisedin goat and conjugated to alkaline phosphatase in Tris buffer) wasadded and solution was incubated for 2 h at room temperature withcontinuous shaking at 500 rpm. To this, 200 μl of freshly prepared p-
20:0
.1 lu
x20
:1 lu
x20
:5 lu
x20
:10
lux
20:2
0 lu
x20
:100
lux
A
B
C
E
F
D
Day
s / L
ight
con
diti
ons
0 12 0 12 0
0
10
20
30
40
50
60
70
80
90
100
Zeitgebertime
0 12 0 12 0
Fig. 1. (Experiment 1). Double plotted activity records of a representative individual (left hn=5) interposed with melatonin levels (bars, mean±SE; n=5 or 6, except in 20:100 in werbirds exposed to T=24 h (LD 12:12, 20:0.1 lx) with sequential change in illumination of tand 100 lx. In the right hand panel, the shaded areas indicate illumination contrasts betweenerror if exceeds the limit of the symbol. Asterisks on the actogram and bar indicate the timethe right side of the actogram indicate the part of the actogram used to calculate the activity
nitrophenyl phosphate (PNPP) substrate solution was added to eachwell, and incubation continued for another 40 min. Finally, 50 μl ofPNPP stop solution was added to each well, and optical density wasmeasured at 450 nm, using 650 nm wavelength as the reference.The lower detection limit of the assay was 1.6 pg/ml. The intra- andinter-assay variations were 7.4% and 12.8%, respectively.
Data presentation and analysis
Actograms were plotted for each bird. The period of the circadianrhythm in activity (tau, τ) was calculated by Chi square periodogramanalysis using The Chronobiology Kit program. Also, activity counts athourly intervals were calculated for a selected duration in the exper-iment. They were averaged for a number of days, and from this, thedaily activity profile was plotted as mean±SE. This is presented inthe right hand panels of Figs. 1 and 2, to better illustrate the responseof the endogenous clock system to experimental conditions. Along
0
200
400
0
80
160
24020 lux 0.1 luxA
* a
0
200
400
0
80
160
24020 lux 10 luxD
*
a
0
200
400
0
80
160
24020 lux 1 luxB
*
a
0
200
400
0
80
160
24020 lux 20 luxE
0
200
400
0
80
160
24020 lux 5 luxC
*
a
0 6 12 18 24
0
200
400
0
80
160
24020 lux 100 luxF
*
a
Act
ivit
y / h
our
Pla
sma
mel
aton
in (
pg/m
l)
0 12 24
Zeitgebertime
and panel) and daily activity profile (hollow circles linked with a solid line, mean±SE,hich n=4) during the middle of the day and night (right hand panel) of Indian weav-he second half of the 24 h day (ZT 12–24, initially described as night) to 1-, 5-, 10-, 20-,two 12 h halves of the 24 h day. Vertical lines on the point symbol indicates the standardof blood sampling and the significance of difference (Pb0.05), respectively. Brackets onprofile and period of circadian activity rhythm.
Zeitgeber time Zeitgeber time
0
10
20
30
40
50
60
70
80
90
100
Day
s / L
ight
con
diti
ons
A
B
C
E
D
100:
0.1
lux
LL
1 lu
xL
L 0
.1 lu
xL
L 3
lux
LL
5 lu
x
0 12 0 12 0
0 12 0 12 0
0
200
400
600
0
70
140
210
280
* a
0 6 12 18 24
0
50
100
150
0
70
140
210
280
a
0
15
30
45
0
70
140
210
280
* a
0
50
100
150
0
70
140
210
280
* a
0
50
100
150
0
70
140
210
280A
ctiv
ity
/ h
our
Pla
sma
mel
aton
in (
pg
/ml)
A
B
C
E
D
0 12 24
100 0.1 lux
0.1 lux
1 lux
3 lux
5 lux
Fig. 2. (Experiment 2). Double plotted activity records of a representative individual (left hand panel) and daily activity profile (hollow circles linked with a solid line, mean±SE,n=7) interposed with melatonin levels (bars, mean±SE; n=6 or 7) during the middle of the day and night (right hand panel) of Indian weaverbirds initially exposed to 12L:12D(100:0.1 lx) for about 3 weeks. Then, they were released in constant dim light (LLdim, 0.1 lx), and after six weeks the light intensity of LLdim was sequentially increased to 1-, 3- and5 lx at the intervals of about 2.5 and 2 weeks, respectively. Asterisks on the actogram and bar indicate the time of blood sampling and the significance of difference (Pb0.05), respec-tively. Brackets on the right side of the actogram indicate the part of the actogram used to calculate the activity profile and period of circadian activity rhythm.
530 J. Singh et al. / Hormones and Behavior 61 (2012) 527–534
with the activity, the mean±SE melatonin levels are plotted in eachcondition, showing the relationship between the behavioral and mel-atonin rhythms.
The data were analyzed in two ways. One, we used the Student'spaired t-test to show differences between two values as the functionof time (e.g. day and night) in the same light conditions. Secondly,we employed one-way analysis of variance for repeated measures(1-way ANOVA RM) followed by Newman–Keuls multiple compari-son post-hoc test to determine significant changes in a measurementamong different light conditions in the experiment. For example, 1-way ANOVA RM determined significant changes in the activity pro-file over 24 h, and in the activity measurements (duration, period)and melatonin levels among the light conditions in experiment A.Pb0.05 was considered significant. All statistical analyses were per-formed using GraphPad prism software program (version 5.0,GraphPad Software Inc. San Diego, USA). For experiment 1, themean±SE activity profile is plotted for five birds since due to tech-nical reasons the data from one individual was incomplete. Also,plasma samples from one or two individuals at one or two time
points in certain light conditions could not be collected for melato-nin measurements.
Results
Experiment 1 (Table 1, Fig. 1)
Activity rhythmsInitially, the activity pattern in all birds followed the LD cycle
(20:0.1 lx), with activity during the day and rest at night (Fig. 1A). In-crease in night illumination by 10-folds (i.e. 20:1 lx) did not alter thispattern, except in the last 2 h of the dark phase when birds were pro-gressively active (Fig. 1B). This trend was continued on further en-hancement of night illumination to 5 lx (20:5 lx). In the secondweek of 20:5 lx condition, day activity was attenuated (Pb0.05,paired t-test, as compared to preceding 20:1 lx condition), and nightactivity was enhanced in the last 4 h (Fig. 1C). The pattern was almostsimilar until the sixth week of 20:5 lx, when the night illuminationwas raised to 10 lx. Under 20:10 lx, activity levels were elevated
Table 1Effect of illumination on period and amount of daily activity in the Indian weaver bird under LD condition (12L:12D).
Light condition (12:12 in lux) Period of activity rhythm(τ, h; mean±SE)
531J. Singh et al. / Hormones and Behavior 61 (2012) 527–534
both in the day and night, but with a much greater individual varia-tion during the daytime (Fig. 1D). In all the above conditions, therewas a significant daily rhythm in the activity profile as revealed by1-way RM ANOVA (20:0.1 lx, F23,92=11.07, Pb0.0001; 20:1 lx,F23,92=11.33, P b0.0001; 20:5 lx, F23,92=7.31, Pb0.0001; 20:10 lx,F23,92=3.29, Pb0.0001). After the day and night illuminations wereequaled (20:20 lx, i.e. LL), birds lost rhythm in their activity behavior(Fig. 1E; F23,92=1.34, Pb0.1662, 1-way RM ANOVA). On a further 5-fold increase in light illumination of the second half (ZT 12–24;20:100 lx), birds exhibited phase reversal in their activity behavior.Birds were at rest during the first 12 h and active during the last12 h, with a significant daily rhythm in the activity behavior(Fig. 1F; F23,92=9.15, pb0.0001).
The period of the activity rhythm, and the amount of activity duringthe entire 24 h day or during the first (ZT 0–12) and second (ZT 12–24)halves of the day in different light conditions (Figs. 1A–F) are shown inTable 1. Apart from the 20:20 lx condition, birds were synchronized inall light conditions (τ=24 h; Figs. 1D and E). In 20:10 lx, however, 1/5 birds was arrhythmic. In all synchronized states (20:0.1, 20:1, 20:5,or 20:10 lx), the activity in the 12 h period with higher light intensitywas significantly greater than the activity in the 12 h period withlower light intensity (Pb0.05; paired t-test). However, the mean totalactivity per day (24 h) was not significantly different among differentlight conditions (Table 1; F3,12=1.446, p=0.2782). Also, there was asignificant difference in the amount of activity at night (ZT 12–24),but not during the day, among the four synchronized states with similaractivity phases (20:0.1, 20:1, 20:5 and 20:10 lx; day: F3,12=1.010,P=0.4222; night: F3,12=1.010, P=0.0004; 1-way RM ANOVA,Figs. 1A–D). A similar result was found when the second 12 h period(ZT 12–24) had higher light illumination, 20:100 lx (F23,92=9.149,Pb0.0001; 1-way RM ANOVA; Fig. 1F).
Plasma melatonin levelsPlasma melatonin followed the typical daily pattern in each light
condition with low levels during the day and high levels at night(Fig. 1, bars in the right hand panel). This pattern was consistent,found in all birds, and corresponded to the activity profile in the diur-nal weaver birds. In all synchronized states (20:0.1, 20:1, 20:5 and20:10 lx), melatonin levels were significantly high (Pb0.05; pairedt-test) during the night (ZT 18) than during the day (ZT 6). Whenlight illuminations of the LD phases were equaled (20:20 lx), theday and night time melatonin levels were not significantly different,
Table 2Effect of illumination on period and amount of daily activity in the Indian weaver bird und
Light condition (LD, LLdim, in lux) Period of activity rhythm(τ, h; mean±SE)
albeit “night” values were still relatively high (day=24.19±7.3 pg/ml; night=44.19±9.1 pg/ml). The subsequent drastic change inthe light illumination (20:100 lx) reversed melatonin profiles(Fig. 1F). Interestingly, however, total circadian (daily) output of mel-atonin were not significantly different among different light condi-tions (F5,55=1.698; P=0.1505, 1-way RM ANOVA).
Experiment 2 (Table 2, Fig. 2)
There was a significant rhythm in activity in 12L:12D(F23,138=30.57, Pb0.0001, 1-way RM ANOVA) and LLdim at 0.1 lx(F23,138=4.076, Pb0.0001, 1-way RM ANOVA) conditions. Duringthe first about 2.5 weeks of 12L:12D (100:0.1 lx) birds were fully syn-chronized (τ=24.0±0.0 h; Fig. 2A). Thereafter, birds freeran for thenext 6 weeks under LLdim (0.1 lx) with change in periods. During thefirst 2 weeks of LLdim, they freeran with τ=23.48±0.23 h, and there-after τ lengthened to 23.80±0.16 h (Fig. 2B). On ten-fold increase inlight intensity, 4/7 birds freeran with τ=23.9±0.15 h and 3/7 birdswere arrhythmic in the third week of exposure to LLdim at 1 lx(Table 2, Fig. 2C). The mean activity profile during the last week ofLLdim at 1.0 lx did not show a significant rhythm (F23,138=1.488,P=0.0843, 1-way RM ANOVA). At 3- and 5 lx of LLdim, respectively,5/7 and 6/7 birds were arrhythmic, and the remaining two(τ=23.5 h, τ=24.0 h) and one bird (τ=24.6 h) birds were stillrhythmic (Table 2, Fig. 2). Again, the mean activity profile did notshow a significant rhythm (3 lx LLdim: F23,138=1.333, P=0.1571;5 lx LLdim: F23,138=1.90, P=0.1126; 1-way RM ANOVA). The pres-ence and absence of rhythmicity was further confirmed by comparingthe amount of activity between the first (ZT 0–12) and second (ZT12–24) 12 h periods of the day. In 12L:12D and LLdim at 0.1 or 1 lx, ac-tivity in the first 12 h period (ZT/CT 0–12) was significantly greaterthan the activity in the second 12 h period (ZT/CT 12–24; Pb0.05,paired t-test; cf. Figs. 2A–C). Such differences in activity betweentwo halves of the day was lacking in the LLdim at 3- (P=0.3711)and 5- (P=0.2411) lux intensity (Table 2, Figs. 2D and E).
There was a significant effect of the light condition on the amountof activity per day, 24 h (ZT/CT 0–24; F4,24=15.06, Pb0.0001, 1-wayRM ANOVA), or per 12 h (first 12 h, ZT/CT 0–12: F4,24=32.46,Pb0.0001; second 12 h, ZT/CT 12–24: F4,24=9.32, Pb0.0001, 1-wayRM ANOVA). In particular, the activity in 12L:12D was significantlyhigher than in other conditions (Pb0.05, Newman–Keuls post hoctest). In fact, birds entrained to 12L:12D significantly lost their
532 J. Singh et al. / Hormones and Behavior 61 (2012) 527–534
activity when released in 0.1 lx LLdim (Pb0.05, paired t-test), probablydue to the masking by the low light intensity. Also, there was no dif-ference in the activity per day (24 h) or activity per 12 h each day inLLdim among 1-, 3- and 5-lux light conditions (Figs. 2C–E).
Plasma melatonin levelsAs in experiment 1, weaverbirds had a typical diurnal melatonin
profile under 12L:12D (Fig. 2A, bars in the right hand panel), withlow melatonin levels during the day and high levels at night (Pb0.05;paired t-test). When released in LLdim (0.1 lx), daily melatonin patternwas continuedwith significantly lowmelatonin levels during the activeperiod (subjective day) and high levels at night (subjective night)(Pb0.05, paired t-test; Fig. 2B). A ten-fold increase in illumination ofthe LLdim (1 lx) did not alter the plasma melatonin pattern (Fig. 2C);the subjective night levels were significantly higher than the subjectivedaytime levels (Pb0.05; paired t-test). The difference between the sub-jective night and day melatonin levels was considerably reduced underLLdim at 3 lx, and the subjective night levels were no longer significant-ly higher than the subjective day levels (P=0.2007, paired t-test;Fig. 2D). The subjective night and day melatonin levels were almostsimilar under LLdim at 5 lx (Fig. 2E; P=0.6739, paired t-test). Interest-ingly, the total melatonin output was not significantly different amongthe five light conditions imposed during the experiment (F4,48=1.949,P=0.1175, 1-way RM ANOVA).
Discussion
Weaverbirds exhibited an activity pattern typical of a diurnal spe-cies, i.e. hopping during the day and rest at night, as long as they weresynchronized to 12L:12D (experiment 1; Fig. 1). As soon as the illumi-nation contrast between light and dark periods was reduced belowthe entrainment limit, the daily rhythms broke loose and the activitywas scattered throughout 24 h, i.e., birds became arrhythmic (cf.Figs. 1A–E). A 20:10 lx light environment in the experiment 1appeared to be near the threshold of the entrainment limit that de-fines the probable day and night times of the 24 h day, since 1/5bird was clearly arrhythmic under this condition (Table 1; Fig. 1D).At equal 20 lx illuminations of the light and dark periods, all birdswere arrhythmic (Table 1; Fig. 1E).
A similar daily rhythm in melatonin secretion is also suggestedby the low and high plasma melatonin levels in each light conditionas long as weaverbirds could differentiate the times of day andnight in the 24 day (Figs. 1 and 2; bars in the right hand panel).The melatonin levels were low during the “perceived” day (activephase) and high at “perceived” night (inactive phase) of the dailyor circadian cycle (Figs. 1 and 2, right hand panels) under boththe LD and LL conditions. This suggested that circadian functionsof the activity and melatonin oscillators were phase coupled. Thatis, activity and melatonin profiles had an inverse phase relation-ship: activity (day) = low melatonin levels; rest (night) = highmelatonin levels (cf. Figs. 1A–F, right hand panel). A similar rela-tionship between activity and plasma melatonin levels has beenreported in several other birds (Hendel and Turek, 1978; Kumaret al., 2000, 2007a; Oshima et al., 1987; Yamada et al., 1988).
Experiment 2 showed that circadian rhythms were disrupted atrelatively lower light intensities in the freerunning states (LLdim;
Figs 2D and E, left hand panels). For example, a 3- or 5-lux light inten-sity in LL paradigm (experiment 2) resulted in abolition of the circa-dian rhythms in activity and melatonin secretion (Figs. 2D and E).But, 5 lx combined with 20 lx in the 12:12 paradigm was interpretedas the night (experiment 1, Fig. 1) and resulted in synchronized circa-dian rhythms of activity and high melatonin levels at night. This sug-gests that the effects of the light intensity on circadian system are notabsolute. Synchronization of circadian rhythms with an LD environ-ment therefore appears to be dependent upon the subjective inter-pretation of the phases of illuminations, based on the photophase
contrast rather than on the absolute light intensity. This is consistentwith conclusions from two early studies under 12L:12D with varyingcombinations of light intensities of the light and dark periods on ac-tivity rhythms and photoperiodic responses in the Japanese quail(Meyer et al., 1988) and blackheaded bunting (Kumar et al., 1992).A study on Japanese quail, in which plasma melatonin levels weremeasured in birds exposed to varying amplitudes of the 12L:12Dcycle (L=2000 or 1500 lx, D=0 or 2 lx; L=2 lx, D=0.2 lx), alsoshowed that the melatonin levels in 12 h dim light were dependentupon the relative light intensity of the accompanying 12 h periodsin the 24 h day (Meyer and Millam, 1991). The effects of relativechanges in light intensity rather than the light intensity per se onmelatonin levels has been reported in few other studies (Kumar etal., 2007b; Lynch et al., 1981; Reiter et al., 1983).
It is known that exposure to bright continuous light abolishes cir-cadian (daily) rhythms in activity and melatonin in several verte-brates (Arendt, 1995) including birds (Binkley, 1989; Kumar andFollett, 1993; Siguenza et al., 1988; Yamada et al., 1988), irrespectivewhether they are diurnal or nocturnal A desynchronized activityrhythm and dampened melatonin rhythm (or arrhythmic activityand complete absence of melatonin) have been found during sum-mers in species living in the polar environment (Griffiths et al.,1979; Miché et al., 1991; Reierth and Stokkan, 1998; Stokkan et al.,1994). Further, the study of Reierth et al. (1999) on Arctic Svalbardptarmigan (Lagopus mutus hyperboreus) showed that the melatoninproduction was not completely abolished during summers. Rather,plasma melatonin levels slightly rose around midnight in some indi-viduals, although changes in daily melatonin levels were not signifi-cant (Reierth et al., 1999). The present results indicate a comparablesituation under LD 20:20 lx and LLdim at 3- and 5 lx, in which both,the rhythms in activity and melatonin secretion, were desynchro-nized (Figs. 1E and 2D and E). We would argue that such responsesunder LL have evolutionary implications, since in an aperiodic envi-ronment individual would not necessarily require a temporally orga-nized behavior. The development of arrhythmicity in such situationscould in fact be an advantageous feature of the internal timing systemwhich acts to temporally organize daily and seasonal events in a peri-odic environment. Further, in spite of the total loss in the daily pat-tern, total melatonin output, measured as sum of the levels at twotimes of the day, did not significantly differ among the light condi-tions in both experiments (experiment 1: F5,55=1.698; P=0.1505;experiment 2: F4,48=1.949, P=0.1175, 1-way RM ANOVA). Thistends to suggest that the external environment has major effects onthe synchronization of circadian melatonin rhythms, and not on theproduction of melatonin.
The rhythms in activity behavior and melatonin secretionappeared to follow each other in their responses to varying intensitiesof light and dark periods under LD cycles and of LL (cf. Figs. 1 and 2).There was the corresponding pattern and timing of the two circadianfunctions to light conditions, as is evidenced by the following obser-vations. (i) The night illumination levels similarly affected the activityand melatonin levels under 12L:12D; brighter nights had higher ac-tivity and low melatonin levels than the dimmer nights (Table 1;Figs. 1A–D). (ii) Both, the activity and melatonin rhythms were lostunder 20:20 lx in the experiment 1 (Table 1, Fig. 1E) and LLdim at 3-and 5-lux in the experiment 2 (Table 2, Figs. 2D and E). (iii) Therewas a parallel phase reversal in both, the activity pattern and melato-nin levels under 20:100 (read as 100:20) lx condition (experiment 1,Fig. 1F). Hence, circadian oscillators governing activity behavior andmelatonin secretion in weaverbirds appear phase related, and possi-bly have similar endogenous periods. This is consistent with previousfindings on European starlings showing close phase relationships be-tween the circadian oscillators governing behavioral rhythms andmelatonin secretion (Kumar et al., 2000, 2007a). However, wewould like to caution that the above observations and conclusionsare based on the measurement of melatonin using an assay the
533J. Singh et al. / Hormones and Behavior 61 (2012) 527–534
validity of which is yet to be determined for our species. Nonetheless,we are quite confident of our results on melatonin levels, which wereconsistent in all the samples including several typical daytime andnighttime melatonin blood samples.
A few observations of the current study however indicate that theactivity and melatonin oscillators in weaverbirds might not be causallylinked as one would suggest from their phase relationships observed inseveral lighting conditions, discussed above. For example, in spite ofarrhythmicity in activity under certain experimental conditions, birdshad differences in the melatonin levels measured at two times of theday (Figs. 1E and 2D). Also, 1/5 birds that was arrhythmic in activityunder 20:10 lx (experiment 1) exhibited clear day–night difference inits melatonin levels (day=14.3 pg/ml; night=98.7 pg/ml). Further,in 20:20 lx, all birds were arrhythmic in activity but melatonin levelsduring the subjective day and night were still measured as 24.19±7.3and 44.19±9.1 pg/ml, respectively (Fig. 1E). Similarly, in the experi-ment 2, most birds were arrhythmic (experiment 2, Fig. 2C) by theend of exposure to LLdim at 1 lx, but the subjective day and night mela-tonin levels were 44.53±8.8 and 99.70±10.68 pg/ml, respectively.Under LLdim of 3 lx, all birds were arrhythmic, and their day and nightmelatonin levels were 64.18±8.48 and 79.70±15.45 pg/ml, respec-tively. Also, in experiment 2, the light conditions significantly affectedthe total amount of activity but not the melatonin (cf. Figs. 2 A–E).Such differences in response to light conditions between the rhythmsof activity and melatonin secretion in the current experiments (Figs. 1and 2) are consistent with the reported spontaneous temporal dissoci-ation between circadian rhythms in the European starling (Ebihara andGwinner, 1992; Kumar et al., 2000).
In summary, the phase relationship between activity and melato-nin rhythms in Indian weaverbirds continued to exist in both, thesynchronized and freerunning states as long as the daily light envi-ronment, based on the illumination intensity and/or photophase con-trast, was subjectively interpreted by endogenous oscillatorssynchronously as the times of the day and night. We suggest that cir-cadian oscillators governing activity behavior and melatonin secre-tion in weaverbirds are phase coupled, but they seem toindependently respond to environmental cues. This would probablyexplain the varying degree to which the involvement of pinealclock/melatonin in regulation of circadian behaviors has been foundamong different birds (Kumar, 2001).
Acknowledgments
A generous grant from the Department of Science and Technology,New Delhi, India under its scheme of IRHPA funding supported the re-search being reported here.
References
Arendt, J., 1995. Melatonin and the mammalian pineal gland, 1st edition. Chapman andHall, London, p. 331.
Aschoff, J., 1981. Handbook of behavioral neurobiology. Biological Rhythms, vol 4. Ple-num Press, New York, p. 545.
Aschoff, J., von Goetz, C., 1989. Masking of circadian activity rhythms in canaries bylight and dark. J. Biol. Rhythms 4, 29–38.
Binkley, S., 1989. The Clockwork Sparrow: Time, Clocks and Calendars in Biological Or-ganisms, 1st edition. Prentice Hall, USA, p. 272.
Binkley, S., Kluth, E., Menaker, M., 1972. Pineal and locomotor activity: levels and ar-rhythmia in sparrows. J. Comp. Physiol. 77, 163–169.
Brandstätter, R., Kumar, V., Abraham, U., Gwinner, E., 2000. Photoperiodic informationacquired and stored in vivo is retained in vitro by a circadian oscillator, the avianpineal gland. Proc. Natl. Acad. Sci. U. S. A. 97, 12324–12328.
Brandstätter, R., Kumar, V., Van't Hof, T.J., Gwinner, E., 2001. Seasonal variations of invivo and in vitro melatonin production in a passeriform bird, the house sparrow(Passer domesticus). J. Pineal Res. 31, 120–126.
Cassone, V.M., Menaker, M., 1984. Is the avian circadian system a neuroendocrine loop?J. Exp. Zool. 232, 539–549.
Cassone, V.M., Bartell, P.A., Earnest, B.J., Kumar, V., 2008. Duration of melatonin regu-lates seasonal changes in song control nuclei of the house sparrow, Passer domes-ticus: independence from gonads and circadian entrainment. J. Biol. Rhythms 23,49–58.
Chabot, C.C., Menaker, M., 1992. Effects of physiological cycles of infused melatonin oncircadian rhythmicity in pigeons. J. Comp. Physiol. 170 A, 615–622.
Ebihara, S., Gwinner, E., 1992. Different circadian pacemakers control feeding and loco-motor activity rhythms in European starlings. J. Comp. Physiol. 171, 63–67.
Ebihara, S., Kawamura, H., 1981. The role of the pineal organ and the suprachiasmaticnucleus in the control of circadian locomotor rhythms in the Java sparrow, Paddaoryzivora. J. Comp. Physiol. 141, 207–214.
Ebihara, S., Uchiyama, K., Oshima, I., 1984. Circadian organization in the pigeon, Colum-ba livia: the role of the pineal organ and the eye. J. Comp. Physiol. 154, 59–69.
Fuchs, J.L., 1983. Effects of pinealectomy and subsequent melatonin implants on activ-ity rhythms in the house finch (Carpodacus mexicanus). J. Comp. Physiol. 153,413–419.
Gaston, S., 1971. The influence of the pineal organ on the circadian activity rhythm inbirds. In: Menaker, M. (Ed.), Biochronometry. Natl. Acad. Sciences, WashingtonDC, pp. 541–549.
Gaston, S., Menaker, M., 1968. Pineal function: the biological clock in the sparrow. Sci-ence 160, 1125–1127.
Griffiths, D., Seamark, R.F., Bryden, M.M., 1979. Summer and winter cycles in plasmamelatonin levels in the elephant seal (Mirounga leonina). Aust. J. Biol. Sci. 32,581–586.
Gwinner, E., 1978. Effects of pinealectomy on circadian locomotor activity rhythms inEuropean starlings, Sturnus vulgaris. J. Comp. Physiol. 126, 123–129.
Gwinner, E., 1989. Melatonin in the circadian system of birds: model of internal reso-nance. In: Hiroshige, T., Honma, K. (Eds.), Circadian Clocks and Ecology. HokkaidoUniv. Press, Sapporo, Japan, pp. 127–145.
Gwinner, E., Subbaraj, R., Bluhm, C.K., Gerkema, M.P., 1987. Differential effects of pine-alectomy on circadian rhythms of feeding and perch-hopping in the European star-ling. J. Biol. Rhythms 2, 109–120.
Gwinner, E., Hau, M., Heigl, S., 1997. Melatonin: generation and modulation of aviancircadian systems. Brain Res. Bull. 44, 439–444.
Hendel, R.C., Turek, F.W., 1978. Suppression of locomotor activity in sparrows by treat-ment with melatonin. Physiol. Behav. 21, 275–278.
Karaganis, S.P., Kumar, V., Beremand, P.D., Bailey, M.J., Thomas, T.L., Cassone, V.M.,2008. Circadian genomics of the chick pineal gland in vitro. BMC Genomics 9,206.
Klein, D.C., Moore, R.Y., Reppert, S.M., 1991. Suprachiasmatic Nucleus: The Mind'sClock, 1 edition. Oxford University Press, USA, p. 230.
Kumar, V., 2001. Melatonin and circadian rhythmicity in birds. In: Dawson, A.,Chaturvedi, C.M. (Eds.), Avian Endocrinology. Narosa Publ. House, N. Delhi,pp. 93–112.
Kumar, V., Follett, B.K., 1993. The circadian nature of melatonin secretion in Japanesequail (Coturnix coturnix japonica). J. Pineal Res. 14, 192–200.
Kumar, V., Kumar, B.S., Singh, B.P., 1992. Photostimulation of blackheaded bunting:subjective interpretation of day and night depends upon both photophase contrastand light intensity. Physiol. Behav. 51, 1213–1217.
Kumar, V., Gwinner, E., Van't Hof, T.J., 2000. Circadian rhythms of melatonin in the Eu-ropean starling exposed to different lighting conditions: relationship with locomo-tor and feeding rhythms. J. Comp. Physiol. A 186, 205–215.
Kumar, V., Singh, B.P., 2006. The timekeeping system in birds. Proc. Indian Nat Sci.Academy B 5&6, 267–284.
Kumar, V., Van't Hof, T.J., Gwinner, E., 2007a. Circadian behavioral and melatoninrhythms in the European starling under light–dark cycles with steadily changingperiods: evidence for close mutual coupling. Horm. Behav. 52, 409–416.
Kumar, V., Rani, S., Malik, S., Trivedi, A.K., Schwabl, I., Helm, B., Gwinner, E., 2007b. Day-time light intensity affects seasonal timing via changes in the nocturnal melatoninlevels. Naturwissenschaften 94, 693–696.
Lahiri, D.K., Ge, Y.-W., Sharman, E.H., Bondy, St.C., 2004. Age-related changes in serummelatonin in mice: higher levels of combined melatonin and 6-hydroxymelatoninsulfate in the cerebral cortex than serum, heart, liver and kidney tissues. J. PinealRes. 36, 217–223.
Lynch, H.J., Rivest, R.W., Ronsheim, P.M., Wurtman, R.J., 1981. Light intensity and thecontrol of melatonin secretion in rats. Neuroendocrinology 33, 181–185.
Malik, S., Rani, S., Kumar, V., 2004. Wavelength dependency of light induced effects onphotoperiodic clock in the migratory blackheaded bunting (Emberiza melanoce-phala). Chronobiol. Int. 21, 367–384.
McBride, S., 1973. Pineal biochemical rhythms of the chicken (Gallus domesticus). Ph.D.dissertation, Pittsburgh, PA, University of Pittsburgh.
McMillan, J.P., 1972. Pinealectomy abolishes the circadian rhythm of migratory rest-lessness. J. Comp. Physiol. 79, 105–112.
Menaker, M., Zimmermann, N.H., 1976. Role of pineal in the circadian system of birds.Am. Zool. 16, 45–55.
Meyer, W.E., Millam, J.R., 1991. Plasma melatonin levels in Japanese quail exposed todim light are determined by subjective interpretation of day and night, not light in-tensity. Gen. Comp. Endocrinol. 82, 377–385.
Meyer, W.E., Millam, J.R., Bradley, F.A., 1988. Photostimulation of Japanese quail by dimlight depends upon photophase contrast, not light intensity. Biol. Reprod. 38,536–543.
Miché, F., Vivien-Roels, B., Pévet, P., Spehner, C., Robin, J.P., LeMaho, Y., 1991. Dailypattern of melatonin secretion in an Antarctic bird, the emperor penguin, Apte-nodytes forsteri: seasonal variations, effect of constant illumination and of ad-ministration of isoproterenol or propranolol. Gen. Comp. Endocrinol. 84,249–263.
534 J. Singh et al. / Hormones and Behavior 61 (2012) 527–534
Oshima, I., Yamada, H., Sato, K., Ebihara, S., 1987. The phase relationship between cir-cadian rhythms of locomotor activity and circulating melatonin in the pigeon (Co-lumba livia). Gen. Comp. Endocrinol. 67, 409–414.
Rani, S., Singh, S., Kumar, V., 2005. The pineal clock affects behavioural circadianrhythms but not photoperiodic induction in the Indian weaver bird (Ploceus philip-pinus). J. Ornithol. 146, 355–364.
Reierth, E., Stokkan, K.A., 1998. Activity rhythm in high-Arctic Svalbard ptarmigan(Lagopus mutus hyperboreus). Can. J. Zool. 76, 2031–2039.
Reierth, E., Van't Hof, T.J., Stokkan, K.-A., 1999. Seasonal and daily variations in plasmamelatonin in the high-Arctic Svalbard ptarmigan (Lagopus mutus hyperboreus). J.Biol. Rhythms 14, 314–319.
Reiter, R.J., Steinlechner, S., Richardson, B.A., King, T.S., 1983. Differential responseof pineal melatonin levels to light at night in laboratory raised and wild-captured 13-lined ground squirrels (Spermophilus tridecemlineatus). Life Sci.32, 2625–2629.
Siguenza, A.F., Recio, J.M., Sanchez, M., Agapito, M.T., 1988. Endogenous regulation ofthe pineal melatonin secretion in Gallus domesticus. J. Comp. Physiol. B 158,381–386.
Simpson, S.M., Follett, B.K., 1981. Pineal and hypothalamic pacemakers: their role inregulating circadian rhythmicity in the Japanese quail. J. Comp. Physiol. 144,381–389.
Singh, J., Budki, P., Rani, S., Kumar, V., 2011. Temperature alters the photoperiodicallycontrolled phenologies linked with migration and reproduction in a night-migratory songbird. Proc. R. Soc. B. doi:10.1098/rspb.2011.1062.
Stokkan, K.A., Tyler, N.J.C., Reiter, J.R., 1994. The pineal gland signals autumn to rein-deer (Rangifer tarandus tarandus) exposed to the continuous daylight of the Arcticsummer. Can. J. Zool. 72, 904–909.
Takahashi, J.S., Menaker, M., 1982. Entrainment of the circadian system of the housesparrow: a population of oscillators in pinealectomized birds. J. Comp. Physiol.146, 245–253.
Terzieva, D.D., Mateva, N.D., Vladimirova-Kitova, L.G., 2009. Melatonin reference limitsat 3:00 AM and 8:00 AM in healthy adults. Clin. Lab. 55, 359–361.
Turek, F.W., McMillan, J.P., Menaker, M., 1976. Melatonin: effects on the circadian loco-motor rhythm of sparrows. Science 194, 1441–1443.
Yamada, H., Oshima, I., Sato, K., Ebihara, S., 1988. Loss of the circadian rhythms of loco-motor activity, food intake and plasma melatonin concentration induced by con-stant bright light in the pigeon (Columba livia). J. Comp. Physiol. A 163, 459–463.
Zatz, M., Mullen, D.A., Moskal, J.R., 1988. Photoendocrine transduction in cultured chickpineal cells: effects of light, dark, and potassium on the melatonin rhythm. BrainRes. 438, 199–215.
Zimmerman, N.H., Menaker, M., 1979. The pineal gland: a pacemaker within the circa-dian system of the house sparrow. Proc. Natl. Acad. Sci. U. S. A. 76, 999–1003.
