Circadian synchronization determines critical day length for seasonal responses

Physiology & Behavior 147 (2015) 282–290

Contents lists available at ScienceDirect

Physiology & Behavior journal homepage: www.elsevier.com/locate/phb

Circadian synchronization determines critical day length for seasonal responses Gaurav Majumdar a, Amit Kumar Trivedi a, Neelu Jain Gupta b, Vinod Kumar a,⁎ a b

DST-IRHPA Center for Excellence in Biological Rhythms Research, IndoUS Center for Biological Timing, Department of Zoology,University of Delhi, Delhi 110 007, India Department of Zoology, MMH College Ghaziabad 201009, India

H I G H L I G H T S • • • • •

T-photocycles entrain circadian rhythm of photoinducibility (CRP) in buntings. Entrainment to T-cycles shifts photoinducible phase and alters critical day length. CRP-entrainment based critical day length (CD) involves changes at genetic levels. EYA3 is involved in the photoperiod perception, rather than in the induction. Photoperiod effects on CD involve thyroid hormone responsive genes.

a r t i c l e

i n f o

Article history: Received 31 January 2015 Received in revised form 2 May 2015 Accepted 4 May 2015 Available online 7 May 2015 Keywords: Bunting Critical day length Photoperiodic response Photoinducible phase T-cycle

a b s t r a c t A photoperiodic species initiates fat deposition (in migrants) and gonadal recrudescence in response to a specific duration of natural daylight, called critical day length (CD), when light extends in the inductive phase of the endogenous circadian rhythm of photoinducibility (CRP). The molecular basis of species-specificCD, determined by the entrainment of the CRP, has been poorly understood. To investigate this, we measured expression levels of genes implicated in the photoperiod-induced changes in reproduction (EYA3, TSH beta, DIO2, DIO3, GNRH and GNIH) and metabolism (SIRT1, HMGCR, FASN and PPAR alpha) in photosensitive redheaded buntings subjected to light–dark cycles of varying period lengths (T-photocycles). Buntings were exposed to six T22, T24 or T26 photocycles, with 1 h additional light at night falling at different phases of the entrained CRP (T2211L = 6L:4D:1L:11D; T2411L = 6L:4D:1L:13D,T2412L = 6L:5D:1L:12D, T2413L = 6L:6D:1L:11D; T2612L = 6L:5D:1L:14D). Photoinduction at genetic and phenotypic levels in T2412L and T2413L, not T2411L, groups confirmed CD being close to 12 h in buntings under T24. Compared to T24, exposure to T22 advanced CD by 1 h, as evidenced by photoinduction in the T2211L, not T226L, group. Similarly, CD appeared to be delayed under T26, with no photoinduction in the T2612L group. Further, to show that induction of response under a Tphotocycle was because of the interaction of inductive phase of the CRP with 1 h during the dark period in each cycle, not with the 6 h main light periods falling 2 h earlier each successive 24 h day in a T22 paradigm, a group of buntings was exposed to 6L:16D (T226L), to which they did not respond. The mRNA expression of genes, particularly TSH beta, DIO2, DIO3 and PPAR alpha, was significantly correlated with changes in reproductive and metabolic phenotypes. These results suggest CRP-entrainment based genetic regulation of the CD, and extend the idea that synchronization with environment is a critical measure in a seasonal species for its temporal adaptation in the wild. © 2015 Elsevier Inc. All rights reserved.

1. Introduction Day length (photoperiod) is used to time reproduction and associated events to the best time of year in many vertebrates, including birds. To avoid resource competition, many species time gonadal recrudescence in response to a specific duration of the natural photoperiod. ⁎ Corresponding author. E-mail address: [email protected], [email protected] (V. Kumar).

http://dx.doi.org/10.1016/j.physbeh.2015.05.005 0031-9384/© 2015 Elsevier Inc. All rights reserved.

As a measure of this, the minimum light period in a 24 h environment that will induce a response in half of the test population has been defined as the critical day length, CD, which can be both species- and response-specific[1]. For example, CD for the induction of gonadal growth is close to 11.5 h in Japanese quail, Coturnix coturnix japonica [2], 12 h in blackheaded bunting, Emberiza melanocephala[3] and redheaded bunting, Emberizabruniceps [4] and 12.5 h in the golden hamster, Mesocritecus auratus[5]. Also, 11.5 h and 11.75 h photoperiods induce half-maximal testis growth and full, albeit at a slower rate, fat

G. Majumdar et al. / Physiology & Behavior 147 (2015) 282–290

deposition and weight gain in the blackheaded bunting and redheaded bunting, respectively [3,4]. A photoperiod longer than CD acts as a long day; buntings fully fatten, gain weight and reproductively mature their testes in about 4-week exposure to a 13 h photoperiod [3,4]. A photoperiodic response is triggered when light period extends in the photoinducible phase (ϕi) lying in the second half of the endogenous circadian rhythm of photoinducibility (CRP; [6,7]).In a synchronized 24 h light–dark (LD) environment, the ϕi begins around midpoint of the daily cycle, i.e. about 12 h after lights on. This explains why a species with a CD of ~ 12 interprets 11- and 12 h light per day as non-inductive and inductive photoperiods, respectively. Thus, synchronization of CRP determines the position of ϕi in an LD cycle [6,7,8]. This can be tested by exposure to LD cycles with varying period lengths (T) falling within the range of circadian entrainment (e.g. T = 21–27 h; T-photocycle). In this model, the entrainment of CRP under T-photocycles correspondingly alters the position of ϕi, hence the duration of CD for the photoperiodic induction. This has been demonstrated in a previous study on blackheaded buntings [9]. Buntings interpreted 11 h light as inductive and non-inductive photoperiods in T22 and T24 photocycles, respectively. Similarly, buntings interpreted 12 h light as inductive and non-inductive photoperiods in T24 and T26 photocycles, respectively; a 13 h photoperiod was inductive in a T26 photocycle [9]. Using a skeleton paradigm, in which a short 1 h light period was introduced in the dark period of nonstimulatoryT-photocycles (T22, 6L:16D; T24, 6L:18D; T26, 6L:20D), the ϕi was shown to begin 10-, 11- and 12 h after the onset of main 6 h light period under T22, T24 and T26 photocycles, respectively [9]. It is poorly understood how the entrainment of CRP affects molecular events underlying induction of seasonal responses. Identification of molecules in photoperiodic induction has now provided an opportunity to investigate this. Under the control of eye absent 3 (EYA3) and thyroid stimulating hormone-beta subunit (TSH beta) from the pars tuberalis (PT), there is a rapid switching between type 2 and 3 iodothyronine deiodinase (DIO2 and DIO3) transcription in the tanycytes (ependymal cells) lining the third ventricle [10,11]. This occurs at the transcriptional levels as early as hour 14 (EYA3 and TSH beta; 1st wave of induction genes) and hour 18 (DIO2 and DIO3; 2nd wave of induction genes) of the first long day itself; i.e. when light impinges on to the ϕi [11,12,13]. As a result, T4 is converted into active T3 and in turn, GnRH (gonadotropin releasing hormone) released from the hypothalamus initiates gonadotropin secretion by the pars distalis [11,12,13,14]. In recent years, another hypothalamic dodecapeptide, GnIH (gonadotropin inhibiting hormone) has been shown regulating the synthesis and release of GnRH in response to change in the photoperiod environment [15]. Also, photoperiod-induced change in the mRNA expression of EYA3, TSH beta, DIO2 and DIO3 genes has been found around hours 15 and 19 after exposure to a single 13 or 16 h light period in night-migratory redheaded buntings [16]. A link between genes involved in the circadian timing and photoperiodic induction has been suggested. For example, BMAL1/CLOCK constituting positive element of the circadian clock feedback loop induces TSH beta mRNA expression, and PERIOD1, a core component of the negative element, reduces BMAL1/CLOCK-induced TSH beta levels in mice [17]. The presence of three E-box elements in EYA3 promoter sensitive to BMAL1/CLOCK further supports association between circadian clock and photoperiodism. In mammals, circadian system mediated effects on photoperiod-induced gene expressions involve melatonin, a circadian clock driven output from the pineal gland [12,17]. Melatonin phase synchronizes gene expression and regulates EYA3 expression in the PT[11]. Intriguingly, however, circadian clock governed melatonin secretion is not involved in the regulation of gonadal cycles in photoperiodic birds [18]. The present study asked a specific question, as to whether synchronization of the CRP to external photoperiod environment would determine CD by corresponding changes in the timing of transcription of

283

genes in systems associated with the reproduction and metabolism. If yes, then the exposure to shorter and longer T cycles would advance and delay, respectively, the timing of the transcriptional activation of photoperiod response genes, hence CD, compared to that in the T24 cycle. We investigated this, by examining gene expression in strongly circadian photosensitive redheaded buntings entrained to shorter and longer T-photocycles in a skeleton paradigm. Buntings were exposed to T22, T24 and T26 cycles, with a 6 h main photophase and 1 h additional light introduced during the darkness such that it lasts before or into the ϕi. Hypothalamic expression of six genes measured in the system associated with reproduction included EYA3 and TSH beta – 1st wave of induction genes, DIO2 and DIO3 – 2nd wave of induction genes, and GNRH and GNIH – downstream response genes [11]. As a measure of metabolic states, mRNA expression of genes involved in the glucose (SIRT1, silent mating type information regulation 2 homologue type 1) and lipid (3-hydroxy-3-methyl-glutaryl-CoA reductase, HMGCR or HMG-COA, fatty acid synthase, FASN, peroxisome proliferator-activated receptor alpha, PPAR alpha) metabolism was measured in the hypothalamus and liver. The prediction was that a 1 h light pulse would trigger a transcriptional activation response only if it interacted with the ϕi, beginning around midpoint in an entrained CRP. 2. Material and methods The redheaded bunting is a Palearctic-Indian latitudinal nightmigratory songbird, which overwinters in India (~ 25°N). It arrives in India in late September/October and begins to return to its breeding grounds in west Asia and east Europe (~ 40°N; north-west spring migration) generally in early April [19]. The present study was done on male buntings procured from the overwintering flocks in the first week of March 2013, at least three weeks before they generally initiate spring migration. After acclimation to natural day length and temperature conditions (NDL, sunrise to sunset = ~ 11.5 h) in an outdoor aviary for a week, birds were maintained indoors on a short photoperiod (8 h light:16 h darkness, 8L:16D) and constant temperature (24 ± 2 °C) conditions for two weeks, until the beginning of the experiment. The experiment was performed and the procedures including the ones for the sampling of tissues for mRNA expression measurements, described below, were adopted as per the approval of the Institutional Animal Ethics Committee at the University of Delhi, India. 2.1. Experiment We used an experimental design that was tested in a previous study on blackheaded buntings (E. melanocephala) that measured fat deposition and body mass gain, and testis recrudescence as a result of exposure to T-photocycles[9]. A total of 28 birds were used in this study. They were first exposed to 6L:18D (L = ~ 250 lx; D = 0 lx) for two weeks, and then housed in activity cages (n = 2 or 3) placed individually in photoperiodic chambers providing 6L:18D (L = ~ 250 lx; D = 0 lx) and constant temperature (24 ± 2 °C). After a week, they were randomly distributed in six groups, each of 4 or 5 birds, and exposed for 6 cycles of light–dark cycles (days) with varying period lengths (T), as schematically represented in Fig. 1. Groups 1 and 2 were exposed to a 22 h day (T = 22 or T22) with a 6 h light period (6L:16D, T226L), but group 2 received an additional 1 h light period interposed at zeitgeber time, ZT, 10 (ZT 0 = light onset) in the 16 h dark period (6L:4D:1L:11D, T2211L). Likewise, groups 3–5 were exposed to a 24 h day (6L:18D, T24) with an additional 1 h light period interposed at ZT10 (6L:4D:1L:13D, T2411L), ZT11(6L:5D:1L:12D, T2412L) or ZT12 (6L:6D:1L:11D, T2413L). A sixth group was exposed to a 26 h day (T26) with 1 h light pulse given at ZT11 (6L:5D:1L:14D, T2612L). Thus, we did not compensate light illumination period for the difference in ‘T’, given a short period of exposure (6 cycles) and longer light pulse duration (1 h). Here, a 6L:16D (T226L) group was included to confirm that

284

G. Majumdar et al. / Physiology & Behavior 147 (2015) 282–290

Fig. 1. Experimental design. Photosensitive redheaded buntings entrained to short days (6 h light:18 h darkness; 6L:18D) were exposed to 22–26 h T-photocycles with or without 1 h additional light period introduce during the dark period lasting up to or ±1 h around the midpoint of the cycle, as follows. T = 22 h: 6L:16D (T226L) and 6L:4D:1L:11D (T2211L); T = 24 h: 6L:4D:1L:13D (T2411L), 6L:5D:1L:12D (T2412L) and 6L:6D:1L:11D (T2413L); T = 26 h: 6L:5D:1L:14D (T2612L). After six T-photocycles, the measurements on phenotypes and mRNA expressions were taken at hour 19 relative to lights of the respective light cycle (i.e. zeitgeber time, ZT, 19).

induction of response occurs under a T-photocycle because of the interaction of ϕi with 1 h and not the 6 h main light periods falling 2 h earlier each successive 24 h day in a T22 paradigm. We did not include a 6L:18D or 6L:20D group because that was not necessary for following reasons. There is a 2 h daily delay in 6 h main light period each successive 24 h day in a T26 paradigm, and hence logically at the end of the six cycle 12 h delay in the onset of 6 h light in T26 would be at the same time as with 12 h advance in the T22; this would not make a difference in terms of physiological effects, if any, between T22 and T26 photocycles. A short photoperiod in any case is non-stimulatory for redheaded buntings, and therefore an additional 6L:18D group would not have contributed to the present experimental design [9]. Food and water were provided ad libitum, and were replenished only during the main 6 h light phase. The activity behavior was monitored continuously. Observations on body mass, fat deposition and testis size were taken on days 0 and 7 of the experiment. Birds were sampled in the seventh cycle in between 18.5 and 19.5 h with reference to the light onset (zeitgeber time, ZT, 19). The sampling time was chosen in view of our recent study showing photoperiod induced mRNA expression of genes linked with reproduction and metabolism in redheaded buntings sampled at similar times following a single day of long day photostimulation [15,16]. 2.2. Measurement of activity behavior and change in phenotypes Each activity cage housed 2 or 3 birds, and was mounted with an infrared sensor, which continuously detected general activity of the cage inmates. The activity from each cage was recorded in 5min bins in a separate channel of the computerized data-logging system, as previously described in [20]. The Chronobiology Kit Stanford Software System was used for collecting and analyzing the activity data. Activity records (actograms) were obtained as double plots, wherein successive days were plotted sideways and underneath, to give a better visual resolution of the activity pattern. Finally,

activity emanated from a cage housing two individuals was plotted in a modulo tau format (i.e. corresponding to a ‘T’ photocycle) to show the response of activity behavior to a T22, T24 or T26 skeleton photocycle. Changes in the fat deposition, weight gain, and testis size, were taken as indices of the T-photocycles induced phenotype effects. The fat deposition was subjectively assessed and scored on a scale of 0–5 (0 = no visible subcutaneous fat, 5 = heavy fat deposits all over; [21]) independently by two persons. Body mass was recorded by weighing birds on a top pan balance to an accuracy of 0.1 g. Similarly, the dimensions of the left testis were measured to an accuracy of 0.5 mm (at the beginning birds were unilaterally laparotomized under local anesthesia), and from this testis volume (TV) was calculated using the formula 4/3πab2, where a and b denote half of the long and short axes, respectively [21,22]. Histology of testis examined the effects on reproductive maturation stage of testes, as per method recently used by [23]. Briefly, testes were removed and fixed overnight in 4% paraformaldehyde. Then, they were cryoprotected by passing through 10-, 20and 30% sucrose solution and sectioned at 20 μm on a Leica 1850 Cryostat. Four testis sections from each bird were processed for H–E (Hematoxylin–Eosin) stained differentiation.Images at 10 × were photographed and analyzed for the measurement (mean ± SE) of germinal epithelium thickness (μm), diameter of seminiferous tubules (μm) and dimension of intertubular spaces (area, μm2) using Motic Images 2.0 software (Motic Inc., Hong Kong). 2.3. Measurement of mRNA expression by the real-timePCR (qPCR) The method of sampling and tissue was the same as described in earlier publications [14,16,24]. Briefly, brain and liver were quickly taken out from birds decapitated in surgical room in the background of very dim illumination (b 1.0 lx) emanating red LED sensors; this illumination does not affect gene expression in buntings (unpublished obs.). Tissues were stored in RNA later (Ambion Inc., Cat no. AM7020). The hypothalamus and a portion (3 mg) of liver tissue, each time from the same area of a liver lobe, were incised out from each bird, homogenized in Tri reagent (Ambion Inc., Cat no. AM9738) and processed for RNA isolation, as per the manufacturer's protocol. 1 μg of total RNA was used to prepare the single strand cDNA (Fermentas, Cat no. K1641) used for the quantification of mRNA expression level of the gene of interest. For qPCR measurements, we used gene specific primers (Suppl. Table 1), many of which have been used in our earlier studies in this species [14,16]. Briefly, using previously standardized PCR conditions described [14,16], qPCR was performed by Applied Biosystems Step One Plus real-timePCR system, with a 15 μl reaction volume consisting 1 μl of cDNA (10ng/μl), 1 μl each of forward and reverse gene specific primers (500–1000 nM), 7.5 μl of SYBR Green I master (1 ×, ABI 4387669) and 4.5 μl nuclease free water. β-Actin was used as an internal control (reference) gene. The fold change in relative gene expression levels was calculated using the formula 2−(ΔΔCt)[25], as used in our recent publications [14,16,24]. Briefly, the fluorescence exceeding background levels gave the cycle threshold (C t ), and from this the ΔC t (Ct [target gene] − Ct [reference gene]) was calculated. Then, Ct values were normalized against the Ct value of a sample (calibrator) obtained from the cDNA of pooled mix samples from all six groups. Finally, ΔΔCt values obtained were plotted as negative power to 2. 2.4. Statistics Paired Student t-test compared initial and final values from the same group. One-way analysis of variance (one-way ANOVA) with Tukey's post hoc test determined significance in difference in the testis size and gene expression levels among six experimental groups at the end of the experiment. Significance was taken at p b 0.05. We

G. Majumdar et al. / Physiology & Behavior 147 (2015) 282–290

285

Table 1 Summary of responses in photosensitive redheaded buntings (Emberiza bruniceps) exposed to 6 cycles of different T-photocycles (+, response; ++ and +++, enhanced [magnitude] response; blank cell — no response). Response

Measurement

Parameters

T-photocycle (Group) T226L (1)

Phenotype

Gross

Testis histology

Reproduction genes

1st wave genes 2nd wave genes Downstream genes

Metabolism

Lipid

Glucose

Body mass Fat depots Testis volume Germinal epithelium Tubular diameter Intertubular area EYA3 TSH beta DIO2 DIO3 GNRH GNIH HMGCR PPARα FASN SIRT1

T2211L (2)

+ + + + + + +

T2411L (3)

T2412L (4)

T2413L (5)

+ +

++ ++ ++

+ + ++ ++ +++ ++ ++ ++

++

+

++ ++

+ ++

+ +

+++ +++ +

T2612L (6)

+

++ +

T226L (6L:16D); T2211L (6L:4D:1L:11D); T2411L (6L:4D:1L:13D); T2412L (6L:5D:1L:12D); T2413L (6L:6D:1L:11D); T2612L (6L:5D:1L:14D).

calculated Pearson's correlation coefficient (r) to show relationships between photoperiod-induced phenotypes and genetic variables.All analyses were done using GraphPad prism software (USA) version 5.0.

3. Results Table 1 provides a summary of relative changes in the phenotype and gene expression in bunting after exposure to 6 cycles of different

Fig. 2. Phenotype responses (mean ± SE) of photosensitive redheaded buntings exposed to six T-photocycles in a skeleton paradigm. a) Change in body mass, b) fat deposition score, c) and testis size, d–f): changes in testis histology, measured in the germinal epithelium thickness (d), diameter of the seminiferous tubules (e) and intertubular space area (f). Asterisk (*) on a bar indicates a significant difference from its initial measurement (paired t-test). Similar and different alphabet letters on bars in c, d, and e figures indicate no-difference and significant difference, respectively, between groups at the end of T-photocycle treatments, shown on X-axis, as determined by Tukey's post hoc test. Significance was taken at p b 0.05.

286

G. Majumdar et al. / Physiology & Behavior 147 (2015) 282–290

T-photocycles.There were differential responses in the phenotype, reproduction and metabolism genes among the T-photocycles. 3.1. Activity behavior and phenotype responses The activity recordings during the initial week of 6L:18D (Suppl. Fig. 1a) and following 6 cycles of T-photocyles showed that circadian activity rhythm was synchronized with the 6 h main light period, regardless of the period length of LD cycles (Suppl. Fig.1b–g). There was also an activity component (perhaps masked) during the 1 h light period introduced at different hours, relative to lights-on, during the dark period of an LD cycle (Suppl. Fig. 1c–g). Body mass was not changed, except a reduction in the T2612L group (p b 0.05; paired t-test;Fig. 2a) at end of T-photocycle exposures. Similarly, fat deposition was visible in T2412L and T2413L(p b 0.05; paired ttest), but not in the other groups (Fig. 2b). Also, testis recrudescence was initiated in T2211L, T2412L, and T2413L(p b 0.05, beginning vs. end; paired t-test), but not in the T226L, T2411L, and T2612L groups (Fig. 2c). Histology of testes also reflected this (Fig. 3a–f). There were with significantly larger seminiferous tubules in birds exposed to T2211L, T2412L or T2413L photocycles, compared to those exposed to T226L, T2411L, or T2612L photocycles (p b 0.05, Tukey's test, cf. Figs. 2c, e; 3a–f). The same was true of other two histological parameters, but with a group difference. For example, there were significantly thicker germinal epithelium and greater intertubular spaces in T2211L and T2413L, but not in the T2412L in which testis recrudescence was initiated (cf. Figs. 2c, d and f; Fig. 3b, d, e). 3.2. Gene expressions 3.2.1. Genes linked with photoperiodic control of reproduction EYA3 mRNA expression significantly varied among T-photocycles (F5,22 = 6.486, p = 0.0008, one-way ANOVA), with significantly higher levels in T2211L, T2411L, T2412L and T2413L groups than in the T226L and T2612L groups (p b 0.05, Tukey's test; Fig 4a). Similarly, TSH beta and

DIO2 mRNA expressions with significant differences among the Tphotocycles(TSH beta: F5,22 = 6.349, p = 0.0009;DIO2: F5,22 = 5.607, p = 0.0018; one-way ANOVA) were significantly higher in T2211L, T2412L and T2413L than the T226L, T2411L and T2612L groups (p b 0.05, Tukey's test; Fig 4b, c). Conversely, DIO3 expression levels, with a significant difference among the treatment cycles (F5,22 = 34.22, p b 0.0001; one-way ANOVA), were significantly higher in T226L and T2612L than the other T-photocycles(p b 0.05, Tukey's test; Fig. 4d). Downstream reproductive genes, GNRH and GNIH, also showed significant differences in expressions among the T-photocycles(GNRH: F5,22 = 8.469, p = 0.0001;GNIH: F5,22 = 4.251, p = 0.0074, one-way ANOVA; Fig. 4e,f). GNRH mRNA levels were high in T2412L and T2413L groups, as compared to T2411L and/or 2612L groups (p b 0.05; Fig. 4e). In the other three groups, GNRH levels were either intermediate (T2211L, T2411L) or very low (T2612L) after six T-photocycles(Fig. 4e). However, GNRH mRNA expression was unexpectedly high in the T226L group (Fig. 4e). GNIH expression pattern with high levels in the T226L group was almost similar to that of the DIO3, except low GNIH levels in the T2612L group (cf. Fig. 4d, f). 3.2.2. Genes linked with metabolic regulation HMGCR and PPAR alpha mRNA expressions were significantly varied among the T-photocycles in the hypothalamus (HMGCR: F5,22 = 12.13, p b 0.0001;PPAR alpha: F5,22 = 9.956, p b0.0001; one-wayANOVA, Fig. 5a, c), but not in the liver (HMGCR: F5,22 = 0.8075, p = 0.5566; liver — F5,22 =2.221, p=0.0884; one-wayANOVA,Fig. 5b, d). Hypothalamic HMGCR mRNA levels were significantly higher in T2211L and T2412L than the T2411L, T226L and T2612L groups (p b 0.05, Tukey's test; Fig. 5a). HMGCR mRNA expression was also higher in T2413L, as compared to the T2612L(Fig. 5a). Similarly, PPAR alpha mRNA expression levels were significantly higher in T2412L and T2413L than the other but T2211L photocycles (p b 0.05, Tukey's test; Fig. 5d); the lowest hypothalamic PPAR alpha expression was found in T2612L birds (Fig. 5c). However, FASN and SIRT1 gene expressions were not significantly altered both in the hypothalamus (FASN: F5,22 = 0.5150, p = 0.7620; SIRT1:

Fig. 3. Representative microphotographs of Hematoxylin–Eosin (H–E) stained testis sections from six groups of redheaded buntings at the end of exposure to T-photocycles, as indicated at the top left corner of each image and described in the Supplementary Fig. 1 (a — T226L (6L:16D); b — T2211L (6L:4D:1L:11D); c — T2411L (6L:4D:1L:13D); d — T2412L (6L:5D:1L:12D);e — T2413L (6L:6D:1L:11D); T2612L— (6L:5D:1L:14D). Abbreviations: GE, germinal epithelium; IS, interstitial space; ST, seminiferous tubule; SPG, spermatogonial; TD, tubular diameter. Broken lines or covered space represent measurement IS and TD, respectively. Note scale bar = 500 μm.

G. Majumdar et al. / Physiology & Behavior 147 (2015) 282–290

287

Fig. 4. Hypothalamic mRNA expressions (2−ΔΔct value; mean ± SE) of EYA3 (a), TSH beta (2), DIO2 (c), DIO3 (d), GNRH (e) and GNIH (f) genes at hour 19 after light on (zeitgeber time, ZT, 19) in redheaded buntings on day 7 of exposure to six T-photocycle treatments, shown on the X-axis, as per details given in Fig. 1. Similar and different alphabet letters on bars indicate nodifference and significant difference, respectively, at the end of the experiment between different treatment groups, as determined by Tukey's post hoc test. Significance was taken at p b 0.05.

F5,22 = 1.490, p = 0.2334; one-wayANOVA,Fig.5f–h) and liver tissues (FASN: F5,22 = 0.3600, p = 0.8703; SIRT1: F5,22 = 0.6345, p = 0.6756; one-wayANOVA,Fig. 5f–h). 3.2.3. Photoperiod induced phenotypes and gene expressions: correlation At the end of the six-day experiment, there was a significant positive correlation of body mass on testis size (r = 0.4087, p = 0.0308; Fig. 6a) and hypothalamic PPAR alpha mRNA expression levels (r = 0.4665; p = 0.0123; Fig. 6b). Further, there were significant positive correlations of testis size on TSH beta(r = 0.5086, p = 0.0057; Fig. 6c) and DIO2(r = 0.3974, p = 0.0362; Fig. 6d) mRNA expression levels, and of TSH beta on DIO2 expression levels (r = 0.6135, p = 0.0005; Fig. 6f). Similarly, there were significant negative correlations of testis size (r=−0.4293, p= 0.0226; Fig. 6d) and TSH beta(r =− 0.4800, p =0.0097; Fig. 6f) on DIO3 expression levels. However, testis size was not correlated with EYA3, GNRH and GNIH gene expressions in this experiment (Fig. 6c, e). 4. Discussion The entrainment of CRP determines CD by alteration in the timing of the transcription of genes responsible for photoperiodic induction of a seasonal response in redheaded buntings. This is evidenced by (i) the interpretation of 1 h light pulse at hours 10 and 12 in T24 as a short and long day, respectively (cf. Figs. 2, 4, 5), and (ii) decrease and an increase in CD duration under T22 and T26, respectively, compared to that

in the T24 photocycle, as evidenced by enhanced TSH beta and DIO2 mRNA expressions in T2211L and T2412L, not T2612L, groups (Fig. 4b, c). Also, photoperiod-induced transcription of genes in systems associated with reproduction paralleled with the testicular phenotype. Along with the testis growth initiation, there was an increase in the seminiferous tubule diameter and germinal epithelium thickness, indicators of the spermatogonial proliferation, and intertubular space for Leydig cell histogenesis for testosterone biosynthesis. We suggest the involvement of EYA3 in the photoperiod perception, rather than in the photoperiodic induction of gonadal recrudescence [14], as evidenced by T-cycle dependent EYA3 expressions in the present experiments. EYA3 mRNA expressions were low in T226L and T2612L, high in T2413L, and were at intermediate levels in the T2211L, T2411L and T2412L groups (Fig. 4a). How is EYA3 involved in the perception of photoperiod length in birds is unclear at the present. However, we tend to offer a few plausible explanations, although purely speculative at this time. First, deep brain photoreceptor pigments may act as connecting molecules. There are significant positive and negative correlations of RHODOPSIN and NEUROPSIN on EYA3 expression in bunting's hypothalamus in response to increasing daily photoperiods [16]. Secondly, the duration of melatonin secretion, as dictated by photoperiod, may affect hypothalamic EYA3 expression, as reported in the Soay sheep [11]. This interpretation invites caution since the duration of night melatonin decodes photoperiod length [26,27], but is not involved in the photoperiodic induction of gonadal recrudescence in songbirds [18].

288

G. Majumdar et al. / Physiology & Behavior 147 (2015) 282–290

Fig. 5. Hypothalamic (left panel) and liver (right panel) mRNA expression (2−ΔΔct value; mean ± SE) of HMGCR (a, b), PPAR alpha (c, d), FASN (e, f) and SIRT1 (g, h) genes at hour 19 after light on (zeitgeber time, ZT, 19) in redheaded buntings on day 7 of exposure to six T-photocycle treatments, shown on the X-axis, as per details given in Fig. 1. Similar and different alphabet letters on bars indicate no-difference and significant difference, respectively, at the end of the experiment between different treatment groups, as determined by Tukey's post hoc test. Significance was taken at p b 0.05.

There appears the role of thyroid hormone responsive genes in mediating the circadian entrainment effects on CD for photoperiodic induction in buntings (cf. Figs. 2 and 4). Positive and negative correlations of TSH beta on DIO2 and DIO3 expressions, respectively, in parallel with changes in testicular size, support this. Although relatively less distinct, the downstream reproductive genes (GNRH and GNIH) followed similar expression patterns, with a few exceptions as noted below (Fig. 4e, f). Thus, in general, present results are consistent with the reported genetic regulation of long-day effects on reproductive axis in photoperiodic birds and mammals [10,11]. However, few results particularly on GNRH and GnIH are at variance from our initial predictions for this study. First, hypothalamic GNRH mRNA levels were high in the non-stimulatory T226L group, possibly

indicating the role of GNRH in maintenance of the photosensitive state in reproductively immature birds (Fig. 4e), as suggested in European starlings (Sturnus vulgaris; [28,29]). Secondly, GNIH mRNA expressions were low in the non-stimulatory T2411L and T2612L groups (Fig. 4f). Finally, inconsistent with predicted reciprocal relationship, there were low mRNA expressions of GNRH and GnIH in T2411L and T2612L (cf. Fig. 4e, f), and of DIO2 and DIO3 in the T2411L group (cf. Fig. 4c, d). We do not know precise reasons for these inconsistencies, but tend to provide a few purely speculative explanations. GNRH and GNIH probably act as independent neural control systems, and respond differentially to a skeleton photoperiod, in which 1 h light interacts with almost median point (±1 h) of the entrained CRP, at which the stimulatory and nonstimulatory effects may be distinguished. Also, there could possibly be

G. Majumdar et al. / Physiology & Behavior 147 (2015) 282–290

289

Fig. 6. Correlations as determined by Pearson's correlation coefficient between responses in redheaded buntings at the end of T-photocycle treatments. Graphs ‘a’ and ‘b’ show positive correlation of body mass on testis size and PPAR alpha gene expressions, respectively. Similarly, graph plots c–e show correlation of testis size on (c) TSH beta (closed circle, solid line; positive correlation) and EYA3 (open circle, broken line; no correlation), (d) DIO2 (closed circle, solid line; positive correlation) and DIO3 (open circle, broken line; negative correlation), and (e) GNRH (closed circle, solid line; no correlation) and GNIH (open circle, broken line; no correlation) gene expressions. Graph ‘f’ shows correlation of TSH beta on DIO2 (closed circle, solid line; positive correlation) and DIO3 (open circle, broken line; negative correlation). A solid or broken line in scatter plots denotes linear regression.

significant individual differences in genetic response to an LD cycle, as has been reported for large individual variations in GNRH mRNA expression in photosensitive male European starlings [28]. Hypothalamic expressions of HMGCR and PPAR alpha genes (Fig 3b, d) involved in the metabolism of serum lipoprotein and lipids, respectively [30,31], consistent with circadian rhythm dependence of the regulation of metabolism in buntings, as has been reported in mammals [32,33]. Increased PPAR alpha expression may account for diet (hyperphagia) induced fat deposition and weight gain, perhaps involving cytogenesis in the median eminence [34]. However, mRNA expressions of SIRT1 and FASN genes, involved in the glucose metabolism and de novo lipogenesis, respectively [35,36], were unaltered regardless of the external Tphotocycles. This was consistent with the absence of change in metabolic phenotypes (fat deposition and body mass gain) in buntings (Fig. 2). Here, we would discount a small body fat depot recorded in T2412L and T2413L groups, which at score ~1 in a scale of 1–5 only indicated the initiation of fat deposition in these birds. At this time, we could possibly suggest that the rate of change in regulatory molecules underlying lipogenesis in buntings was faster under T2412L and T2413L than under T22 photocycle. Similarly, for tissue differences in expression patterns of HMGCR and PPAR alpha genes, we speculate that while ‘gene switches’ were activated in the hypothalamus, the duration of T-photocycle exposure was perhaps too short a period to effect changes in the liver (cf. Fig. 3). In summary, these results present evidence for a molecular basis of CD, as determined by the entrainment of the CRP. The mRNA

expression patterns of genes further show the involvement of thyroid hormone responsive genes in photoperiod-induced effects in redheaded buntings, evidenced by correlations of body mass and testes on mRNA expressions at the end of T-photocycles. Thus, present results may help explain species-specific response to external photoperiod changes, hence the seasonal timing, in photoperiodic species. In migratory songbirds, photoperiodic regulated events associated with spring migration and reproduction are initiated with a temporal difference to avoid the physiological conflict in the wild [37]. So, it was not surprising to find a difference in expression patterns between genes associated with reproduction and metabolism in migratory redheaded buntings. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.physbeh.2015.05.005. Authors' contributions VK conceived the idea; GM and AKT carried out the experiment; NJ contributed to the experiments. GM, VK and AKT analyzed data; VK and GM wrote paper. Funding Department of Biotechnology, New Delhi (BT/PR4984/MED/30/752/ 2012), Science and Engineering Research Board (IR/SO/LU-02/2005, SR/

290

G. Majumdar et al. / Physiology & Behavior 147 (2015) 282–290

SO/AS-50/2010) and DU-DST PURSE (Dean (R)/2012/1477) grants provided generous financial support. A UGC research award supported NJG, and GM received a CSIR Senior Research Fellowship. Conflict of interest The authors have no conflict of interests. References [1] V. Kumar, Photoperiodism in higher vertebrates: an adaptive strategy in temporal environment, Indian J. Exp. Biol. 35 (1997) 427–437. [2] B.K. Follett, S.L. Maung, Rate of testicular maturation, in relation to gonadotrophin and testosterone levels, in quail exposed to various artificial photoperiods and to natural daylengths, J. Endocrinol. 78 (1978) 267–280. [3] M. Misra, S. Rani, S. Singh, V. Kumar, Regulation of seasonality in the migratory male blackheaded bunting (Emberiza melanocephala), Reprod. Nutr. Dev. 44 (2004) 341–352. [4] S. Rani, S. Singh, M. Misra, S. Malik, B.P. Singh, V. Kumar, Daily light regulates seasonal responses in the migratory male redheaded bunting (Emberiza bruniceps), J Exp Zool A Comp. Exp. Biol. 303 (2005) 541–550. [5] J.A. Elliott, Circadian rhythms and photoperiodic time measurement in mammals, Fed. Proc. 35 (1976) 2339–2346. [6] V. Kumar, B.K. Follett, in: J.B.S. Haldane (Ed.),Comm Proc. Zool. Soc. 1993, pp. 217–227 Calcutta. [7] V. Kumar, J.C. Wingfield, A. Dawson, M. Ramenofsky, S. Rani, P. Bartell, Biological clocks and regulation of seasonal reproduction and migration in birds, Physiol. Biochem. Zool. 83 (2010) 827–835. [8] W.M. Hammer, J.T. Enright, Relationships between photoperiodism and circadian rhythms of activity in the house finch, J. Exp. Biol. 46 (1967) 43–61. [9] V. Kumar, B.S. Kumar, Entrainment of circadian system under variable photocycles (T-photocycles) alters the critical daylength for photoperiodic induction in blackheaded buntings, J. Exp. Zool. 273 (1995) 297–302. [10] T. Yoshimura, S. Yasuo, M. Watanabe, M. Iigo, T. Yamamura, K. Hirunagi, S. Ebihara, Light-induced hormone conversion of T4 to T3 regulates photoperiodic response of gonads in birds, Nature 426 (2003) 178–181. [11] N. Nakao, H. Ono, T. Yamamura, T. Anraku, T. Takagi, K. Higashi, S. Yasuo, Y. Katou, S. Kageyama, Y. Uno, T. Kasukawa, M. Iigo, P.J. Sharp, A. Iwasawa, Y. Suzuki, S. Sugano, T. Niimi, M. Mizutani, T. Namikawa, S. Ebihara, H.R. Ueda, T. Yoshimura, Thyrotrophin in the pars tuberalis triggers photoperiodic response, Nature 452 (2008) 317–322. [12] H. Dardente, C.A. Wyse, M.J. Birnie, S.M. Dupre, A.S. Loudon, G.A. Lincoln, D.G. Hazlerigg, A molecular switch for photoperiod responsiveness in mammals, Curr. Biol. 20 (2010) 2193–2198. [13] T.J. Stevenson, G.F. Ball, Disruption of neuropsin mRNA expression via RNA interference facilitates the photoinduced increase in thyrotropin-stimulating subunit beta in birds, Eur. J. Neurosci. 36 (2012) 2859–2865. [14] G. Majumdar, G. Yadav, S. Rani, V. Kumar, A photoperiodic molecular response in migratory redheaded bunting exposed to a single long day, Gen. Comp. Endocrinol. 204C (2014) 104–113. [15] K. Tsutsui, T. Ubuka, G.E. Bentley, L.J. Kriegsfeld, Review: regulatory mechanisms of gonadotropin-inhibitory hormone (GnIH) synthesis and release in photoperiodic animals, Front. Neurosci 7 (2013)http://dx.doi.org/10.3389/fnins.2013.00060. [16] G. Majumdar, S. Rani, V. Kumar, Hypothalamic gene switches control transitions between seasonal life history states in a night-migratory photoperiodic songbird, Mol. Cell. Endo. 399 (2015) 110–121.

[17] C. Unfried, N. Ansari, S. Yasuo, H.W. Korf, G.C. Von, Impact of melatonin and molecular clockwork components on the expression of thyrotropin beta-chain (Tshb) and the Tsh receptor in the mouse pars tuberalis, Endocrinol 150 (2009) 4653–4662. [18] V. Kumar, A. Dawson, Melatoninand Circadian Rhythmicityin Birds, in: C.M. Chaturvedi (Ed.), Avian Endocrinology, Narosa Publishing House, New Delhi 2001, pp. 93–112. [19] A. Ali, S.D. Ripley, Handbook of the Birds of India and Pakistan, Oxford University Press, Bombay/London/New York, 1974. [20] S. Malik, S. Rani, V. Kumar, Wavelength dependency of light-induced effects on photoperiodic clock in the migratory blackheaded bunting (Emberiza melanocephala), Chronobiol. Int. 21 (2004) 367–384. [21] V. Kumar, B.S. Kumar, B.P. Singh, A. Sarkar, A common functional basis for the photoperiodic mechanism regulating reproductive and metabolic responses in the migratory redheaded bunting, Period. Biolog. 93 (1991) 169–174. [22] V. Kumar, S. Singh, M. Misra, S. Malik, Effects of duration and time of food availability on photoperiodic responses in the migratory male blackheaded bunting (Emberiza melanocephala), J. Exp. Biol. 204 (2001) 2843–2848. [23] A.S. Dixit, N.S. Singh, Environmental control of seasonal reproduction in the wild and captive Eurasian tree sparrow (Passer montanus) with respect to variations in gonadal mass, histology, and sex steroids, Can. J. Zool. 91 (2013) 302–312. [24] D. Singh, S. Rani, V. Kumar, Daily expression of six clock genes in central and peripheral tissues of a night-migratory songbird: evidence for tissue-specific circadian timing, Chronobiol. Int. 30 (2013) 1208–1217. [25] K.J. Livak, T.D. Schmittgen, Analysis of relative gene expression data using real-time quantitative PCR and the 2(− Delta Delta C(T)) method, Methods 25 (2001) 402–408. [26] V. Kumar, B.K. Follett, The circadian nature of melatonin secretion in japanese quail (coturnix coturnix japonica), J. Pineal Res. 14 (1993) 192–200. [27] R. Brandstatter, V. Kumar, T.J. Van't Hof, E.. Gwinner, Seasonal variations of in vivo and in vitro melatonin production in a passeriform bird, the house sparrow (Passer domesticus), J. Pineal Res. 31 (2001) 120–126. [28] T. Ubuka, P.A. Cadigan, A. Wang, J. Liu, G.E. Bentley, Identification of european starling GnRH-I precursor mRNA and its seasonal regulation, Gen. Comp. Endocrinol. 162 (2009) 301–306. [29] G.E. Bentley, S. Tucker, H. Chou, M. Hau, N. Perfito, Testicular growth and regression are not correlated with Dio2 expression in a wild male songbird, Sturnus vulgaris, exposed to natural changes in photoperiod, Endocrinology 154 (2013) 1813–1819. [30] J.L. Goldstein, M.S. Brown, Regulation of the mevalonate pathway, Nature 343 (1990) 425–430. [31] N. Latruffe, J. Vamecq, Peroxisome proliferators and peroxisome proliferator activated receptors (PPARs) as regulators of lipid metabolism, Biochimie 79 (1997) 81–94. [32] J. Acimovic, M. Fink, D. Pompon, I. Bjorkhem, J. Hirayama, P. Sassone-Corsi, M. Golicnik, D. Rozman, CREM modulates the circadian expression of CYP51, HMGCR and cholesterogenesis in the liver, Biochem. Biophys. Res. Commun. 376 (2008) 206–210. [33] L. Chen, G. Yang, PPARs integrate the mammalian clock and energy metabolism, PPAR Res. 2014 (2014) 653017. [34] D.A. Lee, J.L. Bedont, T. Pak, H. Wang, J. Song, A. Miranda-Angulo, V. Takiar, V. Charubhumi, F. Balordi, H. Takebayashi, S. Aja, E. Ford, G. Fishell, S. Blackshaw, Tanycytes of the hypothalamic median eminence form a diet-responsive neurogenic niche, Nat. Neurosci. 15 (2012) 700–702. [35] C. Sun, F. Zhang, X. Ge, T. Yan, X. Chen, X. Shi, Q. Zhai, SIRT1 improves insulin sensitivity under insulin-resistant conditions by repressing PTP1B, Cell Metab. 6 (2007) 307–319. [36] B. Schmid, J.F. Rippmann, M. Tadayyon, B.S. Hamilton, Inhibition of fatty acid synthase prevents preadipocyte differentiation, Biochem. Biophys. Res. Commun. 328 (2005) 1073–1082. [37] V. Kumar, S. Rani, B.P. Singh, Biological clock help reduce the physiological conflicts in avian migrants, J. Orninthol. 147 (2006) 281–286.