Circadian patterns and photoperiodic modulation of clock gene expression and neuroendocrine hormone secretion in the marine teleost Larimichthys crocea

ABSTRACT The light/dark cycle, known as the photoperiod, plays a crucial role in influencing various physiological activities in fish, such as growth, feeding and reproduction. However, the underlying mechanisms of this influence are not fully understood. This study focuses on exploring the impact of different light regimes (LD: 12 h of light and 12 h of darkness; LL: 24 h of light and 0 h of darkness; DD: 0 h of light and 24 h of darkness) on the expression of clock genes (LcClocka, LcClockb, LcBmal, LcPer1, LcPer2) and the secretion of hormones (melatonin, GnRH, NPY) in the large yellow croaker, Larimichthys crocea. Real-time quantitative PCR (RT-qPCR) and enzyme-linked immunosorbent assays were utilized to assess how photoperiod variations affect clock gene expression and hormone secretion. The results indicate that changes in photoperiod can disrupt the rhythmic patterns of clock genes, leading to phase shifts and decreased expression. Particularly under LL conditions, the pineal LcClocka, LcBmal and LcPer1 genes lose their rhythmicity, while LcClockb and LcPer2 genes exhibit phase shifts, highlighting the importance of dark phase entrainment for maintaining rhythmicity. Additionally, altered photoperiod affects the neuroendocrine system of L. crocea. In comparison to the LD condition, LL and DD treatments showed a phase delay of GnRH secretion and an acceleration of NPY synthesis. These findings provide valuable insights into the regulatory patterns of circadian rhythms in fish and may contribute to optimizing the light environment in the L. crocea farming industry.


Introduction
The majority of organisms in the natural world demonstrate inherent circadian rhythms, which have primarily evolved to optimize their ability to adapt to cyclic environmental changes influenced by the Earth's light-dark cycle (Bloch et al. 2013).These rhythms are regulated by the circadian clock system, a highly conserved physiological process that enables organisms to anticipate environmental rhythmic changes and modulate their behavior and physiological activities accordingly (Emerson et al. 2008;Pierce et al. 2008).The circadian clock operates continuously for 24 h and depends on an intricate system of clock-related proteins that are arranged in a transcription-translation feedback loop (Partch et al. 2013).This loop includes both positive and negative regulatory elements.Examples of positive regulatory elements are genes like circadian locomotor output cycles kaput (Clock) and brain and muscle Arntlike protein-1 (Bmal1), which encode transcriptional activators such as CLOCK and BMAL1.When CLOCK and BMAL1 heterodimerize, it triggers the activation of negative regulatory elements such as the period circadian clock (Per) and cryptochrome circadian regulator (Cry) genes, both of which incorporate E-box enhancer sequences (Ishikawa et al. 2002).Subsequently, products of Per and Cry genes accumulate and dimerize to form complexes that interact with CLOCK and BMAL1 proteins, inhibiting their own transcription and regulating biological processes (Mohawk et al. 2012) Extensive research has delved into the molecular composition of circadian clock systems in a wide array of species, spanning mammals, teleost, insects, crustaceans and fungi (Dunlap 1999;Dvornyk et al. 2003;Loudon 2012), with potential variations across species due to gene duplication and loss (Pegoraro and Tauber 2011;Yuan et al. 2007).For instance, mouse express negative regulatory elements like Per1, Per2, Per3, Cry1 and Cry2 genes (Ko and Takahashi 2006), while zebrafish express Per1a, Per1b,Per2,Per3,Cry1a,Cry1b,Cry2a,Cry2b,Cry3,Cry4 and Cry5 genes (Moshe et al. 2014;Pierce et al. 2008;Ying et al. 2013).
The circadian clock system primarily operates the genes and proteins at the molecular level, implementing its function to keeping the rhythmic homeostasis.On a macroscopic scale, the regulation of biological behavioral rhythms is achieved through the coordination of functional clocks found in different tissues within an organism.In mammals, the suprachiasmatic nucleus (SCN) located in the hypothalamus acts as the central core clock, receiving external light information through the retinohypothalamic tract (Touitou 2016).Then, it coordinates the rhythms from the master to peripheral clocks with the humoral factors through various neuronal pathways (Dibner et al. 2010).In teleost, both the retina and pineal function as core clocks, directly responding to changes in the light-dark cycle and secreting melatonin to regulate downstream physiological behaviors (Ekström and Meissl 1997;Gupta et al. 2016).Although the secretion of melatonin to regulate circadian rhythms is a consistent feature of vertebrate physiology, the mechanisms governing circadian regulation among different species is still in need of being further clarified, especially in lower vertebrates (Saha et al. 2019).Fish possess light-sensitive sensors in deep brain regions, which play a crucial role in regulating circadian rhythms and seasonal reproduction (Fernandes et al. 2013;Yee Hang et al. 2016).It continues to display numerous rhythms even after undergoing pinealectomy (Sanchez et al. 2000).Thus, the presence of a central clock in fish has not been definitively confirmed.These findings indicate that the circadian system in teleost functions as a network of clocks that can operate independently while interacting to coordinate various rhythms (Cahill 2002;Matthew and Corsi 2002;Vatine et al. 2011).
Previous research consistently underscores the pivotal role of light as a critical environmental factor in aquatic settings, exerting influence across various life stages of fish (Bonvini et al. 2016;Casey et al. 2020;Wu et al. 2019).Photoperiod, in particular, serves as a vital source of rhythmic information for individual organisms (Kumar et al. 2010).Research has clarified that growth, food intake and digestion are intricately intertwined with specific behavioral rhythms, with the secretion of melatonin by the pineal considered a pivotal regulatory factor in governing these processes (Kuz'mina 2020).In mammals, photoperiod assumes a significant role in regulating diverse physiological processes, including growth, development and sexual maturation (Collier et al. 2006).Similarly, photoperiod exerts a profound impact on various physiological activities in fish, offering the possibility being applied to enhance the economic value of aquatic products (Bani et al. 2009;Liu et al. 2015).For example, European sea bass (Dicentrarchus labrax) juveniles have exhibited greater body length when reared under a 12 h light: 12 h dark photoperiod regimen compared to 24 h light and 24 h dark photoperiod conditions (Villamizar et al. 2009).However, the mortality rate of juvenile Atlantic cod (Gadus morhua) reared under a 24 h light photoperiod was lower than those reared under 18 h light: 6 h dark and 12 h light: 12 h dark photoperiod condition (Puvanendran and Brown 2002).Research on the sapphire devil indicates that long photoperiod conditions can induce ovarian development outside of the nonbreeding season (Bapary et al. 2009).These findings underscore the intimate relationship between the light environment and behavioral rhythms in fish, exerting profound effects on growth, feeding, reproduction and digestion.
Larimichthys crocea, commonly known as the yellow croaker, is an economically important fish species in East China Sea.To our knowledge, the circadian changes of clock components in L. crocea have not been characterized so far.Additionally, there is a significant lack of research on the regulatory mechanisms of photoperiod conditions, a critical factor influencing circadian rhythms in teleost.So, this study aimed to investigate circadian variations in clock genes and related neuroendocrinal hormones in specific tissues of L. crocea, as well as the modulation of these patterns by photoperiod.The expression patterns of clock genes (LcClocka,LcClockb,LcBmal,LcPer1,LcPer2) in the pineal, hypothalamus and pituitary have been extensively measured, with a particular emphasis on assessing their circadian variations and their responses to changes in photoperiod.Furthermore, the presence of reproductive and feeding regulatory hormones, such as GnRH and NPY, has been detected to unravel the interconnections between photoperiod, circadian rhythms and their impacts on the physiological processes involving the endocrine system in L. crocea.These findings provide valuable insights into how the circadian oscillator in L. crocea responds to light, forming a basis for improving aquaculture lighting conditions to boost the breeding industry.

Larimichthys crocea clock genes
We conducted multiple sequence alignments of clock gene protein sequences of L. crocea and representative species using GENEDOC software.For phylogenetic tree construction, Maximum-Likelihood phylogenetic reconstruction of the alignment was conducted using MEGAX with a TN substitution model and with 1000 Ultrafast bootstrap replications.The resulting tree was visualized using ItolTree v6 (https://itol.embl.de/).Protein structure domain identification, annotation and structural analysis were performed using SMART (Simple Modular Architecture Research Tool) (http:// smart.embl.de).

Acclimation of experimental L. crocea
Five hundred fish (average body weight: 100 ± 20 g) were procured from local suppliers in Zhoushan, China, in May.The fish were initially acclimated for 30 d under LD (12 h of light and 12 h of darkness) light conditions (Figure S4), during which the water temperature was maintained at a constant 22 ± 0.5°C, with dissolved oxygen levels of 8.0 ± 0.3 mg/L and salinity ranging from 28‰ to 30‰.During acclimatization, fish were randomly and evenly distributed into six 500 L breeding barrels and were fed commercial dry fish pellets ad libitum at 09:00 h.Weekly water changes were performed to uphold water quality standards.The utilized breeding containers featured a proprietary selfcirculation and microfiltration system developed by Hai Xing Intelligent Equipment Co., Ltd.(Shandong, China).Prior to initiating the study, we ensured compliance with the relevant policies regarding animal experiments and animal welfare.All procedures conducted in this study were approved by the Institutional Animal Care and Use Committee of Zhejiang Ocean University.

Tissue distribution of L. crocea clock gene
To identify potential key tissues regulating circadian rhythms, the tissue distribution of clock genes in the L. crocea was analyzed using RT-qPCR.After anesthetizing the fish with MS-222 (Sigma-Aldrich, USA; 100 mg/L), the examined tissues included the pineal, hypothalamus, pituitary, heart, liver, kidney, intestine and head kidney, all dissected at 10:00 h on the last day of the acclimation phase (Figure S4).Biological replicates were obtained by dissecting the mentioned tissues from four fish.After dissection, the tissues were promptly placed in freeze-storage tubes and stored in a liquid nitrogen.
Briefly, total RNA was extracted from the tissues as described above using the RNA Fast200 Extreme Extraction Kit (Fei Jie Bio Company, China) as per the manufacturer's instructions.RNA concentration and quality were evaluated through UV absorbance measurements using a NanoDrop 1000 (Thermo Scientific, USA).RNA samples with OD 260 nm/OD 280 nm ratios between 1.8 and 2.1 were deemed suitable for subsequent RT-qPCR analysis.Reverse transcription was performed using a 10 µl system, containing 6 µl of total RNA mixed with 1 µl of oligo (dT) primer.The mixture was subjected to heat shock at 70°C for 15 min, followed by 2 min on ice.Then, 2 µl of 5 × M-MLV reaction buffer, 0.5 µl of dNTP mix, 0.25 µl of M-MLV reverse transcriptase enzyme and 0.25 µl of reverse transcriptase inhibitors (all from Takara, Japan) were added to initiate the reverse transcription reaction.The reaction included incubation at 42°C for 1 h, followed by heating to 70°C for 15 min and finally cooling to 12°C to complete the process.The reaction was carried out in a PCR apparatus (Heal Force, China).
Five clock genes expression level were quantified using RT-qPCR following the standard protocol.In a 10 μl reaction mixture, contained 5 μl TB Green Premix Ex Taq TM (Takara, Japan) served as the RT-qPCR master mix, with 0.4 μl of forward and reverse gene-specific primers (Table 1, Sangon Biotech, China), 3.2 μl of DEPC water (Beyotime, China) and 1 μl of c DNA template.Gene-specific primers were designed based on total cDNA sequences of LcClocka, LcClockb, LcBmal, LcPer1 and LcPer2 genes (entry numbers: Table S1), reference gene primers for LcActin and LcTubulin were also designed according to their cDNA sequences (accession numbers: XM_010733633.3,XM_027288870.1).Quant Studio TM Real-Time PCR System (Thermo Fisher, USA) was used for RT-qPCR reactions.Melting curve analysis confirmed gene amplification specificity and intensity.The 2 −ΔΔCt method was employed to calculate relative expression of clock genes, normalized by the mean Ct values of each gene across all samples within a run.

Light regimes and rearing of L. crocea
After 30 d of acclimation, the fish were divided into three groups (LD: 12 h of light and 12 h of darkness; LL: continuous light; DD: continuous darkness) for photoperiodic treatment.All fishes were maintained under their respective light conditions for 14 d (Figure S4).The temperature was kept at 22 ± 0.5°C, dissolved oxygen levels at 8.0 ± 0.3 mg/L and salinity within the range of 28‰ ~ 30‰.The lights were switched on at 06:00 h and switched off at 18:00 h, using white fluorescent bulbs with an intensity of 200 ± 20 lx at the water surface.Photoperiods were accurately maintained via an intelligent breeding system (Hai Xing Intelligent Technology, China).

Sampling and gene expression analysis
Following 14 d of exposure to different photoperiods, according to the tissue distribution results for L. crocea clock genes (Figure 2), the pineal, hypothalamus and pituitary tissues with higher mRNA levels were chosen for sampling to examine the diurnal expression patterns of clock genes in the L. crocea (Figure S4).
Organ sampling occurred at 6 separate time points: 02:00 (zeitgeber time 2, ZT2), 06:00 (ZT6), 10:00 (ZT10), 14:00 (ZT14), 18:00 (ZT18) and 22:00 (ZT22), with nocturnal collections under subdued red light to minimize light-induced effects.At each sampling time point, we sampled six individuals from each treatment group.After anesthetizing the fish with MS-222 (Sigma-Aldrich, USA; 100 mg/L), we initially performed tail amputation to collect blood, which was then stored in 1.5 ml tubes (Axygen, USA) pre-filled with heparin sodium (Solarbio, China; 10 mg/ml).Subsequently, after centrifugation at 4°C and 1000 g for 10 min, the obtained supernatant was transferred to cryotubes and stored in liquid nitrogen tank for subsequent hormone detection.Then, the corresponding tissues (pineal, hypothalamic and pituitary) were dissected and placed in cryotubes and subsequently stored in liquid nitrogen tank for RT-qPCR scores of clock gene expression profiles.Concurrently, eyeballs were also dissected and stored in a liquid nitrogen tank, solely for subsequent melatonin-level detection (Figure S4).The subsequent RNA extraction, reverse transcription and RT-qPCR conditions followed the same as those described above for the tissue distribution assay.

Melatonin, GnRH, NPY, circadian secretory rhythm analysis in plasma and eye
Measurement of melatonin, GnRH and NPY concentrations in the plasma and melatonin concentration in the eye were assessed using competitive inhibition Enzyme-linked Immunosorbent Assay (ELISA) Kits for melatonin, GnRH and NPY from Jiang Lai Biological Company (Shanghai, China).As per the manufacturer's instructions, plasma samples were thawed completely and directly subjected to test; the eyes were homogenized using a weight-to-volume ratio of 1 g: 9 ml PBS buffer, ensuring thorough homogenization.Afterward, the homogenate was centrifuged at a speed of 2000-3000 g for 20 min and the resulting supernatant was collected and divided into 1.5 ml tubes (Axygen, USA) for subsequent analysis.

Statistical analysis
Data are presented as mean ± standard deviation (SD).One-way ANOVA analysis was employed using Prism 8.0 software to assess mRNA-level differences among the six daily time points under LD, LL and DD conditions.Statistical significance was defined as p < 0.05, with distinct letters indicating significant differences in figures.To examine the influence of photoperiod changes on downstream hormone secretion (melatonin, GnRH, NPY), plasma hormone concentration data at six daily time points were subjected to One-way ANOVA analysis, with distinct letters also indicating significant differences in figures.Significant differences among three photoperiod treatment groups at each time point also assessed through One-way ANOVA with asterisks denoting their significance (*p < 0.05, **p < 0.01, ***p < 0.0001).
For assessment of daily and circadian rhythmicity in clock gene expression, cosinor analysis was performed using an online tool (https://cosinor.online/app/cosinor. php).The calculation model was f = M + Acos(ωt + φ) where M (mesor) is the median of the fitted cosine curve, A, ω and φ represent amplitude, angular frequency and peak of the cosine fitting (acrophase), respectively.The variable t spans 24 h and f represents gene expression at each 4-h interval.Significance for daily or circadian rhythmicity was determined at p < 0.05.

Sequences and phylogenetics of Larimichthys crocea clock genes
PCR was used to confirm the presence of five clock genes (LcClocka, LcClockb, LcBmal, LcPer1 and LcPer2) in L. crocea.These genes are designated with specific sequences in the Gene-Bank database (entry numbers: Table S1).Detailed amino acid sequence alignments and phylogenetic analyses for these genes are displayed in Figure S1-S3 and Figure 1.domains and a PAS-related C-terminal basic region (PAC) domain (Figure 1b, S1).To explore their evolutionary relationships, we aligned their DNA sequences with Clock and Bmal genes from mammals, teleost and constructed a phylogenetic tree.The analysis revealed that LcClocka clusters with Clocka genes from teleost such as Morone saxatilis and Danio rerio (Figure 1a).Similarly, LcClockb clusters with the Clockb family from marine teleost such as Takifugu rubripes, Perca fluviatilis, etc (Figure 1a).
The LcBmal gene ORF encompasses 1917 bp, encoding a polypeptide of 638 amino acids.In addition, the LcBmal gene has the same functional domain as LcClocka and LcClockb genes, including a typical helix-ring-helix (HLH) domain, two PAS domains and a PAS-associated C-terminal fundamental region (PAC) domain (Figure 1b, S2).LcBmal gene also clusters with Bmal1 family of Danio rerio and other fish species, distinct from the mammalian cluster (Figure 1a).These findings underscore that the circadian clock of marine teleost is evolutionarily closer to that of mammals while being distinct from the evolutionary branch of circadian clocks in amphibian.
The ORF of LcPer1 gene spans 4185 bp, yielding a polypeptide of 1394 amino acids.Similarly, the ORF of LcPer2 gene spans 4521 bp, encoding a 1506amino acid polypeptide.Their protein sequences exhibit highly similar domains to human and mouse Per proteins, including two PAS domains and a PAS-related C-terminal basic region (PAC) domain (Figure 1b, S3).Phylogenetic analysis demonstrates that LcPer1 and LcPer2 genes cluster with Per family from Morone saxatilis and Danio rerio (Figure 1a), highlighting their close evolutionary relationship.

Tissue distribution of clock genes in L. crocea
As depicted in Figure 2, all five clock genes were detectable in diverse tissues of L. crocea, with varying levels of expression abundance.Results revealed that LcClocka, LcClockb, LcBmal and LcPer2 genes exhibit higher expression in the pineal and pituitary.However, there is no significant difference in LcPer1 gene expression across different tissues.In addition to the pineal and pituitary, LcPer1 gene also demonstrates high expression levels in liver, heart and intestinal tissues.Furthermore, LcBmal displayed relatively high expression in the liver, while LcPer2 showed relatively high expression in the heart.Meanwhile, the expression of five clock genes in the head kidney and kidney were relatively low.
Under LL condition, only LcClockb and LcPer2 genes exhibited daily rhythm and their expression gradually increased from ZT10, reached the acrophase at ZT20:11 and ZT22:47 (Table 2).Notably, a phase delay of LcClockb, LcPer2 expression patterns were observed when compared to the LD condition.However, the expression patterns of the other three clock genes did not exhibit significant circadian variation (Figure 3).
In the DD group, five clock genes kept significant circadian rhythms, with levels increasing from ZT6 and reached the acrophase at ZT14:12, ZT15:22, ZT16:38, ZT16:52, ZT15:08, respectively.Compared to the LD group, with only a decrease in peak expression levels for the LcClocka, LcClockb and LcPer1 genes.(Figure 3 and Table 2).
The above results show that alterations in photoperiod had a notable impact on the expression patterns of LcClocka, LcBmal and LcPer1 genes, resulting in the decrease compared to the 12 h light: 12 h dark photoperiod at ZT14 (Figure 3, A, C and D).Conversely, the expression of the LcClockb gene remained largely unaffected by changes in photoperiod.Notably, LL conditions had a more significant effect on the rhythmic expression of clock genes compared to the DD condition, primarily characterized by the loss of rhythmicity in LcClocka, LcBmal and LcPer1 gene expression following LL treatment, while significant rhythmicity persisted under DD conditions.

Temporal patterns of five core clock genes in the hypothalamus under different light regimes
In Figure 4, under LD conditions, LcClocka, LcClockb and LcPer1 genes exhibited significant circadian variation.Cosinor analysis confirmed significant circadian rhythms for LcClocka, LcClockb and LcPer1 genes (Table 2), but without LcBmal and LcPer2 (Figure 4).
Under LL conditions, LcClockb, LcBmal, LcPer1 and LcPer2 genes all exhibited significant expression rhythms (Table 2).LcClockb, LcBmal and LcPer2 gene expression increased from ZT2 and reached the acrophase at ZT11:49, ZT7:40, ZT9:12, respectively.While LcPer1 gene expression started to increase at ZT18, reached the acrophase at ZT5:59.Compared to the LD condition, the peak expression of the LcClocka gene significantly decreased, while the peak expression of LcClockb, LcBmal and LcPer2 genes relatively increased.It is noteworthy that there were shifts in the increasing phase of expression for all five clock genes (Figure 4b, c, e).
These results show that photoperiodic changes notably affect the daily expression patterns of five clock genes.Specifically, under LL and DD conditions, the expression patterns of LcClockb, LcBmal and LcPer2 genes displayed stronger circadian expression amplitudes compared to the LD condition (Figure 4).

Temporal patterns of five core clock genes in the pituitary under different light regimes
As shown in Figure 5, cosinor analysis confirmed the absence of significant circadian rhythms for five clock genes under LD conditions (Table 2).Their transcription profiles revealed an initial increase upon entering the light phase (ZT6), reached the acrophase at ZT13:09, ZT12:01, ZT15:32, ZT11:28, ZT14:33, respectively, and followed by a subsequent decline.Interestingly, a distinct bimodal pattern was observed with their increased expression at the beginning of the dark phase (ZT18 to ZT6).
Following exposure to LL and DD photoperiodic conditions, the rhythmic expression of some genes was improved.After LL treatment, the LcClockb gene displayed daily rhythm (Table 2).Under DD conditions, both LcPer1 and LcPer2 genes displayed circadian expression rhythms (Table 2).Notably, all five clock genes showed the decreased expression between ZT6 to ZT14 and an increase between ZT14 to ZT18.A phase delay of five clock genes were found in the LL and DD group relative to the LD group (Figure 5).
The results show that both LL and DD conditions had a similar impact on clock gene expression in the pituitary.There was no significant difference on the peak expression levels but rather shifts in the phase of all five clock genes in this tissue.

Temporal patterns of melatonin levels in plasma and eye of L. crocea after exposure to different photoperiod
As shown in Figure 6a, b, plasma melatonin exhibited significant circadian oscillation under the three photoperiod conditions, with low levels during the day and high levels in the night.Cosinor analysis confirmed the significant circadian rhythms of melatonin in plasma in all the photoperiod condition (Table 3).Compared to the LD condition, the LL group melatonin suppressed by about 17% at ZT22 (One-way ANOVA, p < 0.0001), while the DD condition melatonin increased by approximately 22.5% at ZT2 (One-way ANOVA, p < 0.0001).
Melatonin in the eye also exhibited significant circadian oscillation under the three conditions, with low levels during the day and high levels in night (Figure 6c, d).
Cosinor analysis results indicated that the circadian variation of melatonin in the eye maintained a significant circadian rhythm (Table 3).Furthermore, the acrophase was delayed in the LL group (+1.9 h) and DD group (+0.8 h) compared to the LD group (Table 3).Compared to the LD group, Melatonin in LL group was suppressed by 11.5% at ZT14 and 14.8% at ZT2 (Figure 6c, One-way ANOVA, p = 0.0001, 0.0229), whereas melatonin in the DD group was enhanced by 28.16% at ZT6 (Figure 6d, One-way ANOVA, p < 0.0001).

Temporal patterns of gonadotropin releasing hormone (GnRH) and neuropeptide Y (NPY) levels in plasma of L. crocea after exposure to different photoperiod
As depicted in Figure 7, the temporal patterns of plasma GnRH displayed pronounced circadian oscillations, with elevated levels during the daytime and diminished levels during the nighttime when L. crocea were exposed to LD, LL, or DD conditions.This trend stands in contrast to the pattern observed for melatonin.Cosinor analysis confirmed the significant circadian rhythm in the circadian variation of GnRH in plasma under the three photoperiodic conditions, with LL and DD groups exhibiting delayed acrophases compared to LD (+0.47 h and +0.92  7a, b).Comparatively, the LL condition increased GnRH levels by 29.3% at ZT14 and 81.7% at ZT22 (One-way ANOVA, p < 0.0001), while the DD condition elevated GnRH levels by 17.3% at ZT14 and 55.7% at ZT22 compared to LD (One-way ANOVA, p < 0.0001).
The temporal patterns of plasma NPY also exhibited significant circadian oscillation, with lower levels during the day and higher levels in the night when fish were exposed to LD, LL, or DD (Figure 7c,d).Similar to the daily secretion pattern of melatonin, cosinor analysis confirmed the significant circadian rhythm in the circadian variation of NPY levels in plasma under the three photoperiodic conditions (Table 4).Compared to the LD condition, the LL condition enhanced NPY levels by 10.5% at ZT14 and 36.3% at ZT22 (One-way ANOVA, p < 0.0001), while the DD group enhanced NPY levels by 18.5% at ZT18 and 24.2% at ZT22 (One-way ANOVA, p = 0.0001, 0.0024).

Discussion
Similar to mammals, fish exhibit physiological and behavioral processes closely associated with circadian rhythms.These rhythms are orchestrated by a central autoregulatory feedback loop that involves clock genes, including Clock, Bmal, Per, and Cry (Vatine et al. 2011;Zhdanova and Reebs 2012).In mammals, key components like the SCN orchestrating the phase synchronization of the circadian rhythms across various somatic peripheral tissues (Welsh et al. 2010).However, the circadian clocks observed in aquatic animals exhibit distinct deviations from those inherent in mammals.Numerous studies suggest that fish possess a relatively decentralized circadian timing system, comprising both the pineal and retina as central oscillators, along with peripheral clocks (Cavallari et al. 2011;Davies et al. 2015).The results indicate that both the pineal and pituitary tissues exhibit high levels of clock gene expression, suggesting a tissue-specific oscillation of clock genes in the L. crocea.The pituitary, as a vital endocrine organ, indicating a close connection between the circadian rhythm system and the neuroendocrine system in the L. crocea.This aligns with previous research emphasizing the interconnection between endocrine and rhythmic processes in fish (Isorna et al. 2016).Furthermore, our results demonstrated the expression of clock genes in the remaining six tissues, supporting the notion that the circadian timing system in fish tends  toward decentralization (Tamai et al. 2005).Consequently, we selected the pineal, pituitary and the significant intermediate endocrine organ, the hypothalamus, as samples to investigate the impact of photoperiod on the temporal expression of clock genes in L. crocea.Additionally, the diurnal secretion patterns of downstream-related hormones (Melatonin, GnRH, NPY) were explored and their altered responses to varying photoperiods were revealed.These works provide new insights into the mechanisms governing circadian rhythmic control and establishes a foundation for enhancing aquaculture lighting conditions to promote the growth of this commercially significant marine fish.
Our results show that under LD condition, all five clock genes in pineal exhibited circadian rhythms, highlighting its role as a core rhythm oscillator (Falcon et al. 2006).L. crocea and Japanese medaka showed similar Clock and Bmal expression patterns, both displaying a gradual increase in expression levels during the light phase and a subsequent decrease during the dark phase.However, the Per1 gene in Japanese medaka display reduced expression during the light phase and increased  expression during the dark phase (Onoue et al. 2019), contrasting with the pattern observed in L. crocea.
Likewise, the Per2 gene in European sea bass also exhibits a pattern of decreasing expression during the light phase and increasing expression during the dark phase, which is entirely different from that in L. crocea (Ma et al. 2023).Taken together, these findings collectively suggest that the clock genes expression patterns of L. corcea are different from other species in pineal.As for hypothalamus, LcClocka, LcClockb and LcPer1 genes exhibited significant rhythmic expression under LD condition, which indicating its crucial role in circadian regulation in L. crocea.This consistent with previous research on chickens, where the hypothalamus also serves as an important circadian oscillator (Jiang et al. 2017).Notably, under LD condition, the hypothalamus showed a delayed phase in clock gene expression compared to the pineal, indicating the priority of pineal to hypothalamus in photoreceptive activities.This finding supports the notion that, in teleost, following photosensitive activity in the retina or pineal, signals are conveyed to the hypothalamus, thereby influencing various physiological processes, such as growth, reproduction and metabolism (Falcón et al. 2010;Wu et al. 2020).In the pituitary, none of the five clock genes exhibited circadian rhythms under LD condition.However, their expression in pituitary exceeded that in other tissues, implying the significant role of pituitary in circadian rhythms regulation, the specific functional mechanisms of pituitary require further investigation.Numerous survival mechanisms among vertebrates, encompassing processes like reproduction, migration and hibernation, display remarkable adaptability in response to seasonal fluctuations (Ikegami et al. 2013;Ross et al. 2015).As is widely recognized, the transition of seasons is primarily characterized by alterations in photoperiod.Vertebrates possess a distinctive biological clock mechanism that enables them to acclimate to varying external photoperiodic conditions (Inagaki et al. 2007).In this study, we disrupted L. crocea circadian rhythm by adjusting the photoperiod, a method successfully used in rat models to investigate photoperiodic impact on the circadian clock and the neuroendocrine system (Ross et al. 2015).Cosine analysis indicated that under LL condition, LcClocka, LcBmal and LcPer1 gene expression rhythms in pineal were significantly disrupted (p > 0.05), LcClockb and LcPer2 genes still had expression rhythms but with weaker amplitudes compared to LD condition.Prior studies on the pineal clock genes Cry1, Bmal1, E4bp4 and Per3 in chicken have suggested that after 12 d of LL treatment, the amplitude of pineal clock gene expression weakens, which consistent with our findings (Turkowska et al. 2014).Nevertheless, under the condition of DD, L. crocea pineal clock gene expression rhythms persisted but with weaker amplitude.This finding is consistent with the research results for pineal clock genes (Bmal1, Bmal2, Clock, Per2) in chickens, as under DD photoperiod conditions, the expression rhythms of these genes were also unaffected (Okano et al. 2001;Yoshimura et al. 2000).Analysis from the pineal indicated that L. crocea clock genes appeared more sensitive to the disrupted photoperiod under LL than DD conditions, which may be related to the behavioral habits of L. crocea.This aligning with a study on razor clam, also show that the circadian rhythm of clock gene appeared to be more significantly influenced under LL than DD (Kong et al. 2023).So, modifying photoperiods can partly disrupt clock gene circadian rhythmicity, primarily causing phase delays and adjustments in daily expression patterns, especially for the constant light treatment in daily cycle.Additionally, some studies have also shown that long photoperiod and short photoperiod have effects on the rhythmic regulation of fish (Sanjita Devi et al. 2022).Therefore, we will continue to performed the systematic research combined with more clock family to better clarified the circadian patterns by photoperiodic modulation in L. crocea.
Previous research has highlighted the vital role of the pineal in melatonin synthesis and release, serving as both photoreceptors and endocrine tissue (Falcon et al. 2009).Melatonin, primarily secreted by the pineal, is recognized as a key molecule in the regulation of internal rhythms, effectively translating light signals into chemical signals (Sánchez-Vázquez et al. 2019).It's worthwhile to mention that in vertebrates, retina may also play a significant role in establishing plasma melatonin rhythms (Gem et al. 1986).Therefore, we investigated the influence of photoperiod on rhythmic melatonin secretion in both plasma and eye in L. crocea.Under the LD condition, melatonin levels in both tissues exhibited a same pattern: decreasing to low levels during the day and elevating to higher levels during the night.These findings consisted with observations in other teleost species, like ayu (Iigo et al. 2003), sea bream (Borja et al. 1996) and sea bass (Bayarri et al. 2004).Melatonin secretion rhythms remained stable even under disrupted daily photoperiod like LL and DD conditions, with variation primarily in peak level and had a certain inhibitory effect on melatonin secretion under LL condition.In the study of melatonin rhythms in Nile tilapia and African catfish, it was also observed that a significant secretion rhythm persists under DD conditions, with melatonin secretion being inhibited under LL conditions, consistent with the findings of our study (Martinez-Chavez et al. 2008).Research on trout and sturgeon has shown that aanat gene (the primary enzyme for melatonin synthesis) maintains its expression rhythm under LL and DD conditions, this outcome indirectly supports the findings of this study (Begay et al. 1998;Coon et al. 1999).Collectively, these data emphasize the ability of pineal to perceive and retain seasonal cues from the surrounding photoperiod.This adaptation likely aids in synchronizing melatonin-mediated physiological responses with the external environment circadian rhythm, essential for survival, especially in response to abrupt changes in the natural light cycle.
The connection between circadian rhythm and neuroendocrine regulation is an emerging research field, with numerous studies highlighting photoperiod significant impact on the reproduction and food intake of teleost (Biswas et al. 2005;Migaud et al. 2010).GnRH, a hypothalamic hormone crucial for vertebrate gonad development (Robertson et al. 2009), exhibited significantly increased secretion under LL condition compared to LD and DD treatment in our results, indicating extended light exposure likely promote L. crocea reproduction, in line with studies in damselfish (Takeuchi et al. 2015).Results also displayed that GnRH exhibited gradually increased during the light phase and gradually decreased during the dark phase, which is contrary to the circadian secretion pattern of melatonin.Furthermore, under LL conditions, there was a decrease in the peak concentration (ZT2) of plasma melatonin, while the GnRH peak concentration (ZT14) increased.It's reasonable to deduced that decreased melatonin promote GnRH secretion, consistent with research on melatonin inhibitory effect on GnRH secretion in Chrysiptera cyanea (Imamura et al. 2022).Several studies have highlighted NPY role in regulating circadian rhythms and food intake (Huhman et al. 1996;Pedrazzini et al. 2003).Data from this study demonstrated that in the LD condition, NPY levels in the plasma of L. crocea displayed significant circadian rhythm.When the photoperiod shifted to LL and DD, the median and amplitude of NPY secretion levels were higher than those in the LD group, but their rhythms remained unaffected, indicating its rhythmic role in promoting feeding in L. crocea.These results suggest that GnRH and NPY have the rhythmic secretion to exert their phycological roles such gonad development and feeding by response to the photoperiodic factors in L. crocea.Meanwhile will offer the theoretical insights for optimizing the breeding environment of L. crocea.

Conclusion
Our findings indicate the presence of five clock genes (LcClocka, LcClockb, LcBmal, LcPer1 and LcPer2) in L. crocea, suggesting the existence of a potentially functional circadian clock in this teleost.These clock genes exhibit broad tissue expression, with notably high levels in the pineal and pituitary, underscoring the presence of a complex peripheral biological clock in this teleost.The prominence of these two tissues implies their crucial role in executing circadian functions.Under different light regimes (LD, LL and DD), the expression patterns of L. crocea clock genes varied significantly and showed tissue-specific characteristics.This indicates that the endogenous rhythm (circadian) is influenced by exogenous (zeitgeber-driven) signals.Conversely, the circadian rhythm of melatonin secretion in L. crocea displays selfsustaining characteristics, providing an adaptive advantage to this teleost in coping with changes in natural light conditions.Furthermore, hormones GnRH and NPY in L. crocea exhibit rhythmic secretion in response to photoperiod, emphasizing their ability to modulate physiological functions through the perception and response to light cues.In summary, these findings serve as a theoretical foundation for exploring the detailed mechanisms of circadian rhythm regulation in teleost.However, a comprehensive investigation is warranted to understand the effects of different photoperiods on the growth and reproductive phenotypes of this teleost.
LcClocka has an open reading frame (ORF) spanning 2643 bp, encoding a polypeptide consisting of 880 amino acids.Similarly, LcClockb ORF spans 2712 bp, encoding a 903-amino acid protein.Two genes share conserved functional domains, including a typical helix-loop-helix (HLH) domain, two PAS

Figure 1 .
Figure 1.(a) Phylogenetic tree comparing the deduced cDNA sequences of Clocka, Clockb, Bmal, Per1, Per2 from Larimichthys crocea (red) and representative organisms mainly including mammals, amphibian, and reptiles.The tree was constructed using the maximum-likelihood method based on the Tamura -Nei model with MEGA X.The numbers represent the frequencies with which the tree topology presented was replicated after 1000 iterations.(b) Conserved domains within the five clock genes of L. crocea.

Figure 6 .
Figure 6.Daily melatonin secretion patterns in plasma (a and b) and eye (c and d) of Larimichthys crocea exposed to different photoperiod.The L. crocea were reared under 12 h light:12 h dark cycle (LD, orange line), continuous light (LL, green line), and continuous darkness (DD, blue line).Asterisks indicate that the concentration level at this point is significantly different compared to LD (*p < 0.05, **p < 0.01, ***p < 0.0001).The values (means ± SD, n = 3) sharing a common letter within the same color were not significantly different (p ≥ 0.05).The shaded part represents the dark phase in the LD environment.

Figure 7 .
Figure 7.The secretion patterns of GnRH (a and b) and NPY (c and d) secretion patterns in plasma of Larimichthys crocea examined under varying photoperiod conditions.The L. crocea were reared under 12 h light: 12 h dark cycle (LD, orange line), continuous light (LL, green line), and continuous darkness (DD, blue line).Plasma was collected every 4 h during the 24 h light -dark cycle.Asterisks indicate that the concentration level at this point is significantly different compared to LD (*p < 0.05, **p < 0.01, ***p < 0.0001).The values (means ± SD, n = 3) sharing a common letter within the same color were not significantly different (p ≥ 0.05).The shaded part represents the dark phase in the LD environment.

Table 1 .
Primers used for RT-qPCR of Larimichthys crocea clock genes and reference gene.
h, respectively; Table4 and Figure

Table 3 .
Rhythm parameters (mean ± SD) of melatonin levels in plasma, eye of Larimichthys crocea reared under different light regimes.

Table 4 .
Rhythm parameters (mean ± SD) of GnRH and NPY levels in plasma in Larimichthys crocea reared under different light regimes.