Figures
Abstract
Lunar cycle-associated physiology has been found in a wide variety of organisms. Recent study has revealed that mRNA levels of Cryptochrome (Cry), one of the circadian clock genes, were significantly higher on a full moon night than on a new moon night in coral, implying the involvement of a photoreception system in the lunar-synchronized spawning. To better establish the generalities surrounding such a mechanism and explore the underlying molecular mechanism, we focused on the relationship between lunar phase, Cry gene expression, and the spawning behavior in a lunar-synchronized spawner, the goldlined spinefoot (Siganus guttatus), and we identified two kinds of Cry genes in this animal. Their mRNA levels showed lunar cycle-dependent expression in the medial part of the brain (mesencephalon and diencephalon) peaking at the first quarter moon. Since this lunar phase coincided with the reproductive phase of the goldlined spinefoot, Cry gene expression was considered a state variable in the lunar phase recognition system. Based on the expression profiles of SgCrys together with the moonlight's pattern of timing and duration during its nightly lunar cycle, we have further speculated on a model of lunar phase recognition for reproductive control in the goldlined spinefoot, which integrates both moonlight and circadian signals in a manner similar to photoperiodic response.
Citation: Fukushiro M, Takeuchi T, Takeuchi Y, Hur S-P, Sugama N, Takemura A, et al. (2011) Lunar Phase-Dependent Expression of Cryptochrome and a Photoperiodic Mechanism for Lunar Phase-Recognition in a Reef Fish, Goldlined Spinefoot. PLoS ONE 6(12): e28643. https://doi.org/10.1371/journal.pone.0028643
Editor: Shin Yamazaki, Vanderbilt University, United States of America
Received: September 12, 2011; Accepted: November 11, 2011; Published: December 7, 2011
Copyright: © 2011 Fukushiro et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was partially supported by the Grants-in-Aid from the Ministry of Education, Culture, Sports, Science and Technology (MEXT), and the Japanese Society for the Promotion of Science (JSPS) of Japan awarded to A.T. and T.O. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. No additional external funding was received for this study.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Synchronous reproductive behaviour is a phenomenon that occurs when all individuals within a population reproduce at a specific phase set by certain periodic changes in the environment. This is an important adaptive strategy in organisms that enables higher reproductive efficiency, decreased predation risk, and the prevention of interspecies mating [1]. In addition to daily and annual cyclic changes in the terrestrial and aquatic environments, lunar cycle-dependent changes are used as cues for synchronous behavior in some species. The lunar- or semilunar-synchronized reproduction has been seen in a wide variety of living organisms in multiple phyla [2]. Although lunar- or semilunar-synchronized spawning behaviors have been demonstrated in corals [3], [4], marine insects [5], certain groupers [6], spinefoots [7], [8] and others, molecular mechanism(s) underlying this behavior remain elusive. Several factors combined have been suggested to determine the reproductive timing in corals. For example, the month of spawning is set by cycles in solar radiation and local weather pattern, the date of spawning is set by the lunar or tidal cycle, and the hour and minute of spawning is set by sunset time [9]. Responsiveness to the amount of moonlight may also be a factor as shown by the coral mRNA levels of Cryptochrome (Cry) which were significantly higher on full moon nights than on new moon nights [10]. Another well-studied organism is a marine midge Clunio marinus which shows (semi)lunar rhythms in reproductive behavior even in laboratory condition, and the rhythms are likely controlled by a combination of circadian clocks, moonlight, tidal fluctuations and temperature [5].
CRY proteins are highly related to (6–4) photolyases, enzymes repairing UV-induced DNA damage with photon energy [11], [12], and contain FAD chromophore [13], [14]. The photosensory properties of cryptochromes (CRYs) have been investigated in a plant Arabidopsis thaliana and a fruit fly Drosophila melanogaster, and they are shown to play important roles in photomorphogenesis and circadian photoreception, respectively [15], [16], [17]. In Drosophila circadian photoreception, illumination triggers interaction between dCRY and dTIMLESS [18], [19], both of which are important factors for circadian clock oscillation and resetting. This interaction results in the degradation of both factors, and consequently another critical factor dPER (Period) becomes destabilized [20]. These events are crucial for phase-shift of the Drosophila circadian clock. In vertebrates, CRY1 and CRY2 are considered to lack the photosensitivity but instead constitute negative feedback loops in the circadian clock system together with PERIOD, CLOCK, and BMAL [21], [22]. The moonlight-associated upregulation of Cry mRNA in coral [10] does not necessarily establish coral CRY as a moonlight-sensitive photoreceptor, but it does imply the involvement of CRY in lunar-synchronized spawning. Because the corals spawn during a full moon, it is difficult to infer whether Cry mRNA levels are directly associated with a putative “circalunar clock” that contains oscillatory factors called state variables determining the phase of the oscillation or whether they are responsive only to the intensity and/or the duration of moonlight. In terms of other lunar-responding animals, there have yet to be any molecular observations of a potential gene involved in the lunar-dependent spawning.
Investigation of the relationship between lunar phase, Cry gene expression patterns, and the lunar-dependent spawning response in animal species other than corals could provide better understanding of a lunar phase-recognition mechanism as well as the generalities surrounding a lunar-associated change of Cry gene expression across animal species. In this study, we focused on the Cry genes of a lunar-synchronized spawner, the goldlined spinefoot (formerly referred to as the golden rabbitfish or orange-spotted spinefoot) (Siganus guttatus). The goldlined spinefoot is a reef fish with a lunar-dependent rhythmic spawning restricted to a species-specific lunar phase (around the first quarter moon) during spawning season [7], [8], [23], [24], [25]. Two Cry genes, SgCry1 and SgCry3, were identified as showing lunar cycle-dependent expression that peaked at the first quarter moon in the goldlined spinefoot brain, leading us to a concept that SgCry genes do not simply respond to the intensity of moonlight, but they are state variables in a lunar phase-recognition system.
Results
Cloning and phylogenetic analysis of SgCry1 and SgCry3
We identified two kinds of Cry genes in the juvenile goldlined spinefoot. Phylogenetic analysis (Figure 1) showed that they can be classified in vertebrate CRY1 and CRY3 groups, respectively, which led us to identify them as SgCRY1 (AB643455) and SgCRY3 (AB643456). In the Nighbor-Joining phylogenetic tree (Figure 1), Cry3 genes seem to form a monophyletic group with Cry2 genes, which strongly suggests that Cry3 genes are fish orthologs of Cry2 genes.
Goldlined spinefoot CRY sequences (SgCRY1 and SgCRY3; present study) and their related sequences obtained from NCBI Entrez database (accession nos. are shown in Table S1) were analyzed using the Neighbor-Joining method [44] and CLUSTAL W (http://clustalw.ddbj.nig.ac.jp/top-e.html). Bootstrap probabilities (p) are represented by closed circles on the nodes (p>95%) or values near the nodes.
Weekly changes in female GSI
In order to evaluate the mRNA expression profiles of SgCry mRNA expression during sexual maturation of the female goldlined spinefoot, we focused on medial part of the brain (mesencephalon and diencephalon) and the ovary since these sites are closely associated with reproduction in vertebrates. We collected the medial part of the brain and ovary from mature fish for approximately 2 months at 7- or 8-day intervals during and just before the first spawning. Concomitantly, we measured the gonadosomatic index (GSI, Figure 2) to confirm that the fish had sexually matured during the study period. GSI did not significantly change from May 25th to June 15th, 2009, but did increase to a peak on June 29th, 2009 (first quarter moon) and significantly decreased (P < 0.01) on July 7th, 2009 (full moon), indicating that June 29th, 2009 was the first of the two spawning phases during this reproductive season.
Gonadosomatic index (GSI) was calculated as described in Materials and Methods. Numbers of fish sampled are indicated above the bars. An arrow and an arrowhead show the spawning and the summer solstice, respectively. Indicated by seven schematic images are lunar phases from new moon (May 25th, left) to full moon (July 7th, right).
Daily variations of SgCry1 and SgCry3 mRNA expression
We investigated daily variations in SgCry1 and SgCry3 mRNA expression and lunar phase-dependent change in the expression pattern using quantitative RT-PCR and cDNA synthesized from the brain and ovary RNA (Figure 3, and Figure S1). Abundances of SgCry1 mRNA in the brain fluctuated with peaks at dawn (ZT0 [Zeitgeber time 0]) and gradual decreases during the daytime (Figure 3A). Although the averaged SgCry1 mRNA expression levels diverged depending on lunar phase (see below), the expression patterns were similar (Figure 3A). On the other hand, mRNA levels of SgCry3 were almost constant in the brain throughout the day and during all lunar phases (Figure 3B). In the ovary, daily variation was not observed for SgCry1 and SgCry3 mRNA expression (Figure 3C, D) except that weak daily variations were observed for SgCry1 and SgCry3 mRNA expression at June 1st and May25th, respectively (p<0.05, Figure S1).
Each tissue (brain [mesencephalon and diencephalon]: n = 4, ovary: n = 3) was collected at ZT0, ZT6, ZT12, ZT18 from new moon to full moon over one and half lunar cycle. Each Cry mRNA level was calculated as a value relative to that of the Sgβ-actin gene. Error bars represent ± SD. Lunar phases are indicated by schematic moon images. Solid lines show the first lunar cycle, and broken lines show second lunar cycle. (A) Daily changes in the SgCry1 mRNA levels in the brain. p<0.01, Two-way ANOVA. (B) Daily changes in the SgCry3 mRNA levels in the brain. (C, D) Daily changes in the SgCry1 and SgCry3 mRNA levels in the ovary. The graphs are separated into single panels for each lunar phase in Figure S1.
Lunar variations of SgCry1 and SgCry3 mRNA expression
We reanalyzed the results of quantitative RT-PCR from the perspective of possible lunar phase-dependent variation in SgCry1 and SgCry3 mRNA levels (Figure 4). Both SgCry1 and SgCry3 mRNA expression in the brain showed significant lunar-dependency (p<0.01; Figure 4A, B). Expression was highest on June 1st (first quarter moon), showed decrease on June 8th (full moon), and the lowest expression around June 8th and June 15th (last quarter moon). There were also fluctuations in both expression levels with peaks on June 29th (first quarter moon). In contrast to the clear lunar phase-dependency in the brain, levels in the ovary did not show much significant variation (Figure 4C, D).
Each Cry mRNA was calculated as a value relative to that of the Sgβ-actin gene. For each lunar phase, mRNA levels at ZT0, ZT6, ZT12 and ZT18 are averaged except for SgCry1 mRNA levels in the brain (mesencephalon and diencephalon). Since SgCry1 showed diurnal variation in the brain, values were normalized to reduce the deviation due to the daily rhythmic changes as follows; values at each daily time point (ZT0, ZT6, ZT12, and ZT18) were divided by the mean values of all the data for the same time point (0.0340, 0.0242, 0.0109, and 0.0141 for ZT0, ZT6, ZT12, and ZT18, respectively). Error bars represent ± SD. Lunar phases are indicated by schematic moon images. (A) Lunar changes in the SgCry1 mRNA levels in the brain. p<0.01, One-way ANOVA. Post hoc test; p<0.05, 5/25 vs 6/8, 6/15; p<0.01, 6/1 vs 6/8, 6/15, 6/22, 6/29, 7/7). (B) Lunar changes in the SgCry3 mRNA levels in the brain. p<0.01, One-way ANOVA. Post hoc test; p<0.05, 6/15 vs 6/22, 7/7; p<0.01, 5/25 vs 6/8, 6/15; p<0.01, 6/1 vs 6/8, 6/15, 6/22, 6/29, 7/7; p<0.01, 6/15 vs 6/29. (C, D) SgCry mRNA levels in the ovary plotted against lunar phase.
Discussion
In this study, we identified two cryptochrome genes in the goldlined spinefoot, SgCry1 and SgCry3, and investigated their mRNA levels in the brain (mesencephalon and diencephalon) and ovary over the two months comprising the reproductive season in this fish species. SgCry1 mRNA levels showed diurnal variation with a peak at dawn (ZT0, Figure 3A), indicating that SgCry1 mRNA expression was under circadian clock control and/or environmental light signals. In contrast to the brain, SgCry1 mRNA levels in the ovary did not show strong diurnal fluctuation (Figure 3C, and Figure S1). This finding is analogous to our previous observation that mRNA levels of Xenopus tropicalis Cry1 and Cry2 did not significantly differ between midday and midnight in the ovary [26]. CRYs in the ovary might be involved in physiological mechanisms other than circadian function.
Reanalysis of SgCry mRNA expression in relation to lunar variation revealed that the expression of both Crys has strong lunar phase-dependency in the brain but not in the ovary (Figure 4A, B). SgCry1 mRNA expression demonstrated diurnal and lunar variation, while SgCry3 mRNA expression demonstrated only lunar phase-dependent variation. These SgCry mRNA variations were highest at the first quarter moon and then decreased before the full moon. Such rhythmic changes were observed in not only the lunar cycle in relation to spawning (June 29th) but also the preceding cycle (May 25th – June 15th), indicating that these changes are not simply associated with spawning. Because we reared the fish in this study under natural photoperiod and moonlight without exposure to the tidal cycle, which has an approximately two-week periodicity, it is highly probable that the environmental cue leading to the perception of lunar phase is not the tide but rather moonlight. In fact, some authors have previously revealed that the goldlined spinefoot is responsive to full moonlight such that there are affects on nocturnal melatonin [27] and Period2 mRNA levels in the pineal gland [28]. In addition to retinal photoreceptors, pineal or encephalic photoreceptors such as exorhodopsin [29], pinopsin [30], parapinopsin [31] and encephalopsin [32] may exist in the fish, and they may be coordinately involved in the system for discriminating strong solar light (∼1×105 Lux) and weak moonlight (less than 1 Lux).
SgCRY1 and SgCRY2 are classified into vertebrate CRY1 and CRY2 groups, respectively (Figure 1). CRYs in the CRY1 and CRY2 groups are called “mammalian type” CRYs [33] and considered to be non-photoreceptive [34] in contrast to photoreceptive “insect type” CRYs (insect CRYs, Figure 1). However, recent studies suggest the photoreceptive nature of the mammalian-type CRYs [35], [36], and therefore SgCRYs might also be candidate photoreceptors for moonlight perception in the goldlined spinefoot, as has been suggested for coral [10]. This idea supports the present result that SgCry3 mRNA levels were highest at the first quarter moon.
Although a mechanism for the detection of moon phase has yet to be identified, the present results imply a general mechanism relevant to a lunar-associated change of Cry gene expression across animal species. The simplest possible mechanism for the detection of moon phase would utilize maximal moonlight intensity and/or the length of a light period that exceeds a certain light intensity. Such a mechanism, however, would be useful for detecting the full moonlight, as in the case of corals, but may be inappropriate for detecting the quarter moons. Therefore, alternative mechanism is likely used at least in the goldlined spinefoot. Another possible mechanism that could explain the recognition of specific lunar phase is a mechanism similar to that for photoperiodic reproductive control, which involves the circadian clock and photoreception. In fact, photoperiodic influences of the moon on the regulation of (semi)lunar-rhythm has been postulated and experimentally explored in a marine midge C. marinus [5].
In birds, there are photoreceptors in the deep brain that detect environmental light information [37], [38], which is then transmitted to the mediobasal hypothalamus (MBH), followed by the regulation of gonadal maturation. In the MBH of long-day responsive quails, day length is measured according to the light conditions surrounding a specific phase in the early night called the “photo-inducible phase” [39], [40]. Previous studies have shown that photoperiodism also induces gonadal maturation in fish such as masu salmon (Oncorhynchus masou, [41]), rainbow trout (O. mykiss, [42]), and gilthead seabream (Sparus aurata, [43]). Moonlight is seen to periodically change both in intensity and duration (timing of moonrise and moonset) in accordance with the moon phase (Figure 5A). When comparing the moonlight patterns (Figure 5A) to the lunar-dependent change in SgCry mRNA expression (Figure 4A, B), we noted a decrease in SgCry mRNA levels from the first quarter moon to full moon that paralleled the increase in light intensity during ZT18–ZT21. Therefore, we are speculating that the mRNA expression of SgCry is controlled by the light conditions during ZT18–ZT21. That is, SgCry mRNA expression is repressed in moon phases with moonlight irradiation during ZT18–ZT21 (phase II and III, Figure 5A, B) and activated in moon phases without moonlight irradiation during ZT18–ZT21 (phase IV and I, Figure 5A, B). If so, the lunar phase-dependent gene expression could be regulated by a moonlight signal given during the “photo-repressive phase”, a specific circadian time similar to the photo-inducible phase in photoperiodism.
(A) Schematic of solar light phase in Okinawa, moonlight phase, and moonlight intensity during the experiments. A putative photo-repressive phase is indicated by white frame. (B) Lunar limit cycle model served by SgCry as a state-variable. SgCry mRNA level decreases from first lunar quarter to last lunar quarter when moonlight intensity is relatively high and exposed time of moonlight is relatively long (Phase II, Phase III. SgCry mRNA levels increase from last lunar quarter to first lunar quarter (Phase IV–Phase I).
The lunar phase-dependent expression of SgCrys, which has a cycle advanced by one fourth that of the lunar cycle, raises the possibility that SgCry could function as a state variable determining the lunar phase. In this system, SgCry expression and moonlight intensity (or another state variable reflecting the moonlight intensity) would form a hypothetical lunar limit cycle (Figure 5B), and the combination of these two factors might constitute a putative “circalunar clock” and enable the goldlined spinefoot to recognize any specific lunar phase as a point within this limit cycle.
In consideration of the expression patterns, SgCry genes are likely under a combined regulation of circadian clock- and moonlight-responsive element(s), or alternatively they are regulated by a novel cis-element, the activity of which changes with the peak seen around the first quarter moon (a lunar phase-responsive element). Further analyses of the transcriptional regulatory mechanisms of SgCry1 and SgCry3 genes in the brain would shed light on the precise mechanism(s) for the lunar phase-recognition system that serves the oocyte maturation processes and decides the timing of spawning through induced synthesis or secretion of GnRH in the brain.
Materials and Methods
Experimental Fish
Experimental animal care was conducted under permission from the Committee for Animal Experimentation of the School of Science and Engineering at Waseda University (permission # 09A03). Animal experiments including sampling in the field were conducted under permission from Sesoko Station Tropical Biosphere Research Center at University of the Ryukyus (permission # 090522–090707). Juvenile goldlined spinefoot fish (0.08–0.15 g) were originally collected by using a fish net at low tide around the new moon period from a mangrove swamp of the Manna River in Okinawa, Japan. The fish were reared in outdoor tanks (capacity: 10 metric tons) with aerated running seawater for three or four years under natural photoperiod, natural moonlight, and water temperature conditions at Sesoko Station, Tropical Biosphere Research Center, University of the Ryukyus, Japan. There are roofs and frames over the tanks to protect them from rain and wind, but the sunlight and moonlight reach to the tanks. There is no artificial light possibly reaching the fish inside the tanks. The feeding of commercial pellets (EP1, Marubeni Nisshin, Tokyo, Japan) took place daily at 10:00 h. Approximately 3-year-old or 4-year-old fish with body weights ranging from 200 to 700 g were collected by using a fish net weekly from May 25th to July 7th, 2009 and used in the present experiments. During the experiment, sunrise and sunset occurred at approx. 6:00 and 19:00, respectively. The fish were taken from the tanks at random and anesthetized with iced seawater in a bucket. The fish were brought to the laboratory within a few minutes under natural light condition (in daytime sampling [06:00–18:00]) or with covering the bucket to keep the fish in the dark (in nighttime sampling [24:00]). The brain (medial part of the brain, mesencephalon and diencephalon) (n = 4) and ovary (n = 3) were collected from the fish at 06:00 (∼ZT0 [Zeitgeber time 0]), 12:00 (∼ZT6), and 18:00 (∼ZT12) under fluorescent light or at 24:00 (∼ZT18) under dim red light (<0.1 mW/cm2,>640 nm). In the sampling at 24:00, the eyes of each fish were covered with aluminum foil during decapitation to minimize an effect of dim red light. Samples were kept in RNAlater (Ambion) at 4°C overnight and then stored at −80°C until RNA extraction. During the sampling, body mass and ovarian mass were recorded separately. Gonadosomatic index (GSI) was calculated using the following formula: GSI = (ovarian mass/body mass)×100.
Complementary DNA Cloning of SgCry1 and SgCry3 Genes Encoding Full-Length Coding Sequences
Total RNA was extracted from juvenile goldlined spinefoot using TRIzol reagent (Invitrogen). First-strand cDNAs were synthesized with SuperScript III reverse transcriptase (Invitrogen) using KSII(dT)21 primer (GAGGTCGACGGTATCGATAAGC(T)21). The SgCry1 and SgCry3 cDNA fragments were amplified using a pair of degenerate primers (5′-GTCCTSGAYCCBTGGTTYG-3′ and 5′-GTCATDATNGCRTCDATCC-3′) that were designed based on the conserved regions of Cry genes from some species (Acropora millepora, Gasterosteus aculeatus, Oryzias latipes, Takifugu rubripes, Xenopus tropicalis, and Gallus gallus). Complementary DNAs including the 5′ and 3′ UTRs were obtained using RACE. cDNAs that included the entire coding sequence for SgCry1 and SgCry3 were obtained by PCR with primers for their 5′ and 3′ UTRs using PfuUltra (Stratagene). The amplified fragments were inserted into the pENTR/D-TOPO vector (Invitrogen), and the inserts of at least three independent clones for each gene were sequenced. Finally, SgCry1 and SgCry3 cDNAs for the CDS without presumed PCR errors were isolated.
Quantitative RT-PCR Analysis
Total RNA was extracted from the tissues using TRIzol reagent (Invitrogen). Residual genomic DNA in the total RNA sample was eliminated by DNase I treatment (RNase-free recombinant DNase I, TaKaRa BIO). Quantitative RT-PCR analyses were performed using StepOnePlus (Applied Biosystems) along with a high capacity cDNA reverse transcription kit (Applied Biosystems). Each reaction included 1 µg of total RNA as a template. The primers for quantitative RT-PCR are shown in Table 1. In order to establish a consistently transcribed gene for reference, we isolated and sequenced partial cDNA from SgrpL13A (AB643457), SgPGK (AB643458), SgEF1α (AB643459), Sgβ-actin (AB643460), and Sgβ2M (AB643461) of the juvenile fish cDNA. We examined the performance of one or more pairs of the primer for each gene by amplification of the cDNA in the presence or absence of RTase, electrophoresis of the amplified fragments. Because we did not obtain primer sets giving specific amplification and a single band for two of the five genes (SgrpL13A and Sgβ2M), the other three genes with sufficient performance (SgPGK, SgEF1α, Sgβ-actin) were used in the further analysis. Among these genes, we selected Sgβ-actin as a reference control gene in the following measurements since the threshold cycles (Ct) for Sgβ-actin were relatively unchanged over the sampling time (Figure S2).
Supporting Information
Figure S1.
SgCry mRNA levels and their daily variation in the brain and ovary.Graphs shown in Fugure 3 were separated into single panels for each lunar phase. A hash mark denotes a data set in which one of the triplicate samples was missing due to a mistake in the experiment.
https://doi.org/10.1371/journal.pone.0028643.s001
(EPS)
Figure S2.
Threshold cylcles for control genes in quantitative RT-PCR analysis.Distribution of threshold cycles for three control genes (SgPGK, SgEF1α, Sgβ-actin) in the quantitative RT-PCR analysis (shown in Figure 3 and Figure 4) were shown by box plots. Outliers were omitted. Based on these plots and standard deviations of threshold cycles for SgPGK (brain 0.55; ovary, 1.23), SgEF1α (brain 0.46; ovary, 1.58), and Sgβ-actin (brain, 0.41; ovary, 1.32), we selected Sgβ-actin as a reference control gene.
https://doi.org/10.1371/journal.pone.0028643.s002
(EPS)
Table S1.
Accession nos. of amino acid sequences used for phylogenetic analysis.
https://doi.org/10.1371/journal.pone.0028643.s003
(DOC)
Author Contributions
Conceived and designed the experiments: MF AT TO. Performed the experiments: MF TT YT S-PH NS AT YK KO TO. Analyzed the data: MF TO. Contributed reagents/materials/analysis tools: AT TO. Wrote the paper: MF TO
References
- 1. Domeier ML, Colin PL (1997) Tropical reef fish spawning aggregations: Defined and reviewed. Bulletin of Marine Science 60: 698–726.
- 2.
Neumann D (1976) Entrainment of a semilunar rhythms. In: DeCoursey PJ, editor. Biological Rhythms in the Marine Environment. Los Angeles: University of South Carolina Press. pp. 115–127.
- 3. Harrison PL, Babcock RC, Bull GD, Oliver JK, Wallace CC, et al. (1984) Mass spawning in tropical reef corals. Science 223: 1186–1189.
- 4. Jokiel PL, Ito RY, Liu PM (1985) Night irradiance and synchronization of lunar release of planula larvae in the reef coral Pocillopora damicornis. Marine Biology 88: 167–174.
- 5. Kaiser TS, Neumann D, Heckel DG (2011) Timing the tides: genetic control of diurnal and lunar emergence times is correlated in the marine midge Clunio marinus. BMC Genet 12: 49.
- 6. Colin PL, Shapiro DY, Weiler D (1987) Aspects of the reproduction of two groupers, Epinephelus guttatus and E. striatus in the West Indies. Bulletin of Marine Science 40: 220–230.
- 7. Rahman MD, Takemura A, Takano K (2000) Correlation between plasma steroid hormones and vitellogenin profiles and lunar periodicity in the female golden rabbitfish, Siganus guttatus (Bloch).Comp Biochem Physiol B Biochem Mol Biol. 127: 113–122.
- 8. Takemura A, Rahman MS, Nakamura S, Park YJ, Takano K (2004) Lunar cycles and reproductive activity in reef fishes with particular attention to rabbitfishes. Fish and Fisheries 5: 317–328.
- 9. Vize PD, Embesi JA, Nickell M, Brown DP, Hagman DK (2005) Tight temporal consistency of coral mass spawning at the Flower Garden Banks, Gulf of Mexico, from 1997-2003. Gulf of Mexico Science 23: 107–114.
- 10. Levy O, Appelbaum L, Leggat W, Gothlif Y, Hayward DC, et al. (2007) Light-responsive cryptochromes from a simple multicellular animal, the coral Acropora millepora. Science 318: 467–470.
- 11. Todo T, Ryo H, Yamamoto K, Toh H, Inui T, et al. (1996) Similarity among the Drosophila (6-4)photolyase, a human photolyase homolog, and the DNA photolyase-blue-light photoreceptor family. Science 272: 109–112.
- 12. Hsu DS, Zhao X, Zhao S, Kazantsev A, Wang RP, et al. (1996) Putative human blue-light photoreceptors hCRY1 and hCRY2 are flavoproteins. Biochemistry 35: 13871–13877.
- 13. Kanai S, Kikuno R, Toh H, Ryo H, Todo T (1997) Molecular evolution of the photolyase-blue-light photoreceptor family. J Mol Evol 45: 535–548.
- 14. Ozturk N, Selby CP, Song SH, Ye R, Tan C, et al. (2009) Comparative photochemistry of animal type 1 and type 4 cryptochromes. Biochemistry 48: 8585–8593.
- 15. Li QH, Yang HQ (2007) Cryptochrome signaling in plants. Photochem Photobiol 83: 94–101.
- 16. Emery P, So WV, Kaneko M, Hall JC, Rosbash M (1998) CRY, a Drosophila clock and light-regulated cryptochrome, is a major contributor to circadian rhythm resetting and photosensitivity. Cell 95: 669–679.
- 17. Stanewsky R, Kaneko M, Emery P, Beretta B, Wager-Smith K, et al. (1998) The cryb mutation identifies cryptochrome as a circadian photoreceptor in Drosophila. Cell 95: 681–692.
- 18. Ceriani MF, Darlington TK, Staknis D, Mas P, Petti AA, et al. (1999) Light-dependent sequestration of TIMELESS by CRYPTOCHROME. Science 285: 553–556.
- 19. Rosato E, Codd V, Mazzotta G, Piccin A, Zordan M, et al. (2001) Light-dependent interaction between Drosophila CRY and the clock protein PER mediated by the carboxy terminus of CRY. Curr Biol 11: 909–917.
- 20. Busza A, Emery-Le M, Rosbash M, Emery P (2004) Roles of the two Drosophila CRYPTOCHROME structural domains in circadian photoreception. Science 304: 1503–1506.
- 21. Liu AC, Lewis WG, Kay SA (2007) Mammalian circadian signaling networks and therapeutic targets. Nat Chem Biol 3: 630–639.
- 22. Ukai-Tadenuma M, Yamada RG, Xu H, Ripperger JA, Liu AC, et al. (2011) Delay in feedback repression by cryptochrome 1 is required for circadian clock function. Cell 144: 268–281.
- 23. Hoque MM, Takemura A, Matsuyama M, Matsuura S, Takano K (1999) Lunar spawning in Siganus canaliculatus. J Fish Biol 55: 1213–1222.
- 24. Harahap AP, Takemura A, Nakamura S, Rahman MS, Takano K (2001) Histological evidence of lunar-synchronized ovarian development and spawning in the spiny rabbitfish Siganus spinus (Linnaeus) around the Ryukyus. Fisheries Science 67: 888–893.
- 25. Rahman MS, Takemura A, Nakamura S, Takano K (2003) Rhythmic changes in testicular activity with lunar cycle in the forktail rabbitfish. J Fish Biol 62: 495–499.
- 26. Kubo Y, Takeuchi T, Okano K, Okano T (2010) Cryptochrome genes are highly expressed in the ovary of the African clawed frog, Xenopus tropicalis. PLoS One 5: e9273.
- 27. Takemura A, Susilo ES, Rahman MD, Morita M (2004) Perception and possible utilization of moonlight intensity for reproductive activities in a lunar-synchronized spawner, the golden rabbitfish.Journal of Experimental Zoology, Part A: Comparative Experimental Biology 301A: 844–851.
- 28. Sugama N, Park JG, Park YJ, Takeuchi Y, Kim SJ, et al. (2008) Moonlight affects nocturnal Period2 transcript levels in the pineal gland of the reef fish Siganus guttatus. J Pineal Res 45: 133–141.
- 29. Mano H, Kojima D, Fukada Y (1999) Exo-rhodopsin: a novel rhodopsin expressed in the zebrafish pineal gland. Brain Res Mol Brain Res 73: 110–118.
- 30. Okano T, Yoshizawa T, Fukada Y (1994) Pinopsin is a chicken pineal photoreceptive molecule. Nature 372: 94–97.
- 31. Blackshaw S, Snyder SH (1997) Parapinopsin, a novel catfish opsin localized to the parapineal organ, defines a new gene family. Journal of Neuroscience 17: 8083–8092.
- 32. Blackshaw S, Snyder SH (1999) Encephalopsin: a novel mammalian extraretinal opsin discretely localized in the brain. J Neurosci 19: 3681–3690.
- 33. Rubin EB, Shemesh Y, Cohen M, Elgavish S, Robertson HM, et al. (2006) Molecular and phylogenetic analyses reveal mammalian-like clockwork in the honey bee (Apis mellifera) and shed new light on the molecular evolution of the circadian clock. Genome Res 16: 1352–1365.
- 34. Griffin EA Jr, Staknis D, Weitz CJ (1999) Light-independent role of CRY1 and CRY2 in the mammalian circadian clock. Science 286: 768–771.
- 35. Tu DC, Batten ML, Palczewski K, Van Gelder RN (2004) Nonvisual photoreception in the chick iris. Science 306: 129–131.
- 36. Foley LE, Gegear RJ, Reppert SM (2011) Human cryptochrome exhibits light-dependent magnetosensitivity. Nat Commun 2: 356.
- 37. Wada Y, Okano T, Adachi A, Ebihara S, Fukada Y (1998) Identification of rhodopsin in the pigeon deep brain. FEBS Lett 424: 53–56.
- 38. Nakane Y, Ikegami K, Ono H, Yamamoto N, Yoshida S, et al. (2010) A mammalian neural tissue opsin (Opsin 5) is a deep brain photoreceptor in birds. Proc Natl Acad Sci U S A 107: 15264–15268.
- 39. Yasuo S, Watanabe M, Tsukada A, Takagi T, Iigo M, et al. (2004) Photoinducible phase-specific light induction of Cry1 gene in the pars tuberalis of Japanese quail. Endocrinology 145: 1612–1616.
- 40. Yoshimura T (2006) Molecular mechanism of the photoperiodic response of gonads in birds and mammals. Comp Biochem Physiol A Mol Integr Physiol 144: 345–350.
- 41. Amano M, Hyodo S, Kitamura S, Ikuta K, Suzuki Y, et al. (1995) Short photoperiod accelerates preoptic and ventral telencephalic salmon GnRH synthesis and precocious maturation in underyearling male masu salmon. Gen Comp Endocrinol 99: 22–27.
- 42. Choi S, Lee CH, Park W, Kim DJ, Sohn YC (2010) Effects of shortened photoperiod on gonadotropin-releasing hormone, gonadotropin, and vitellogenin gene expression associated with ovarian maturation in rainbow trout. Zoolog Sci 27: 24–32.
- 43. Meseguer C, Ramos J, Bayarri MJ, Oliveira C, Sanchez-Vazquez FJ (2008) Light synchronization of the daily spawning rhythms of gilthead sea bream (Sparus aurata L) kept under different photoperiod and after shifting the LD cycle. Chronobiol Int 25: 666–679.
- 44. Saitou N, Nei M (1987) The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol Biol Evol 4: 406–425.