The Photoperiod-Driven Cyclical Secretion of Pineal Melatonin Regulates Seasonal Reproduction in Geese (Anser cygnoides)

Most seasonal changes in photoperiods and laying rates of the female Yangzhou geese are shown in Figure 1. These geese are among Chinese indigenous domestic goose breeds, mainly located in the north of Jiangsu Province in China (Figure 1A), and the natural photoperiods at the sampling site exhibited marked seasonal fluctuation (10.07–14.24 h). Their seasonal cycle of laying rates showed changes due to fluctuations in the annual photoperiod (Figure 1B). The laying rates gradually increased from January (8.66%) to February (29.37%); the highest levels were observed in March (38.70%), slowly decreasing until May (23.41%); they then stopped reproducing in June and started laying eggs again during the final 2 months of the year. Moreover, we measured the egg-laying time during the 24 h day at the SE and WS (Figure 1C,D). The egg-laying time of female Yangzhou geese was relatively concentrated; more than 50% of the eggs were laid from 4:00 a.m. to 8:00 a.m.

The morphology of ovarian development during the different breeding seasons further revealed dramatic seasonal changes in the reproductive performance (Figure 2). At the SE and WS, female geese had normal ovaries and many hierarchical follicles, while the gonads exhibited significant atrophy and degeneration from the SS to AE (Figure 2A). In addition, almost all follicles presented atresia at the AE with a sudden shortening of the photoperiod. The GSI levels significantly increased at the SE compared to WS, while, at the AE, a shortened photoperiod induced the lowest levels (F = 103.402, p < 0.001, Figure 2B). The ovarian area also exhibited a similar seasonal trend to that of the GSI, increasing from the WS to SE and shrinking from the SS to AE (F = 72.496, p < 0.001, Figure 2C).

The morphologic characteristics of the pineal gland also showed a relatively marked seasonal variation (Figure 3A). Significant increases in volume (F = 25.245, p < 0.001) and weight (F = 10.338, p < 0.001) were observed at the AE compared to the SS (Figure 3B,C). The anatomical observations found that the pineal gland was composed of the pineal organ (PO), pineal stalk (PS), and pineal choroid (PC) (Figure 3A). The connective tissue capsule was present in the form of stripes, extending in various degrees into the parenchyma, and most blood vessels were distributed across the lateral of the pineal organ parenchyma. The number of blood vessels was lower at the SS. The pineal parenchyma was distributed with a large number of follicular structures. There was a significant increase in the number of pineal follicle-like structures from the SS to the AE (F = 9.822, p < 0.001, Table 1). In addition, compared to other breeding seasons, the long diameter of pineal follicles reached its maximum at the WS (F = 16.871, p < 0.001), while the height of follicular cells decreased significantly at the SS (F = 10.218, p < 0.001).

The ultrastructure of the pineal gland also showed the typical seasonal features (Figure 3D). Pineal cells were visibly elliptic, circular, or irregular in shape and had many small surface protrusions. The nuclei of cells were comparatively large, and abundant heterochromatin was localized near the nuclear membrane. At the SS, the micrographs showed that the cell nuclei became smaller and heterochromatin density increased. In addition, pineal gland cells contained abundant organelles. The Golgi apparatus exhibited a small area and narrow width at the SS, while being composed of many flattened cisternae arranged in stacks with numerous small vesicles at the AE. During the period from the SS to the AE, the size and number of intracellular secretory vesicles increased. Lastly, the observation of ultrastructural characteristics demonstrated prominent glycogen granules in the cytoplasm.

Mtnrs showed a wide tissue distribution, being found in the pineal gland, hypothalamus, pituitary gland, and ovary (Figure 4). Mel-1a had an obvious seasonal fluctuation in the pineal gland and hypothalamus, and the expression level increase rapidly after the SS, reaching a higher expression at the AE (F = 39.577, p < 0.001; F = 49.654, p < 0.001, respectively) (Figure 4A,B). The expression levels of three Mtnrs all showed at least threefold changes at the AE compared to the values at the SS in the pituitary gland (F = 16.332, p < 0.001, Figure 4C). Among them, Mel-1b had a higher expression compared to other Mtnrs at the WS (p < 0.001). In the ovary, stable expressions of Mel-1a and Mel-1b were consistently at a lower level during the different seasons, while Mel-1c showed a 2.5-fold increase from the SS to AE (F = 85.114, p < 0.001, Figure 4D). Moreover, immunohistochemistry was used to assess the expression of Mtnr in the pineal gland. As shown in Figure 4E, Mtnrs were widely expressed in the pineal organ, with higher levels observed near the connective tissue capsule. Likewise, there were also seasonal variations in Mtnr expression levels (Figure 4F). The expression level of Mtnr decreased from the WS to SE, while it increased significantly at the AE with a shortened photoperiod compared to the SS (F = 28.199, p < 0.001). Lastly, there were consistent seasonal fluctuations between the expression levels of the Aanat gene and the contents of melatonin (MT). The expression of Aanat increased significantly with a shortening of the photoperiod from the SS to AE (F = 186.361, p < 0.001, Figure 4G). Serum melatonin levels were significantly elevated under a shorter photoperiod, reaching a maximum increase approaching threefold at the AE compared to the SS (F = 45.996, p < 0.001, Figure 4H).

The results for seasonal changes in reproduction-related genes are shown in Figure 5. GnRH showed a highest expression level at the SE, and then its expression decreased steadily from the SS to AE with a shortened photoperiod (F = 50.299, p < 0.001, Figure 5A). In the pituitary gland, the expression levels of FSHβ and LH gradually decreased with longer photoperiods, before increasing again at the WS (Figure 5B,C). Consistently, the results of FSHR and LHR also showed similar seasonal oscillations and were expressed at the highest levels at the SE (F = 103.388, p < 0.001; F = 99.268, p < 0.001, respectively) (Figure 5E,F). On the contrary, the expression of PRL in the pituitary gland reached a maximum peak at the SS and showed an almost fourfold increase compared to the AE and WS (F = 110.401, p < 0.001, Figure 5D). Furthermore, the expression of PRLR in the ovary demonstrated a similar trend, with the lowest level at the AE and the highest at the SS (F = 156.698, p < 0.001, Figure 5G).

The levels of reproductive hormone secretion through the different seasons are shown in Figure 6. Among them, the concentrations of GnRH, FSH, LH, E2, and P4 showed similar seasonal oscillation patterns in a short photoperiod and long photoperiod. At the SE, as the photoperiod lengthened, the concentration of these hormones in serum significantly increased compared to the WS, while the lowest levels were found at the AE (F = 75.921, p < 0.001; F = 240.707, p < 0.001; F = 49.034, p < 0.001; F = 81.526, p < 0.001; F = 59.137, p < 0.001, respectively). Although the result of PRL also showed obvious fluctuations over the different seasons, the highest values corresponded to the SS (F = 506.974, p < 0.001, Figure 6D).

This study observed a marked rhythm of seasonal reproduction in female Yangzhou geese, which we speculate is governed by seasonal changes in photoperiod. In the present research, the seasonal changes in photoperiod were crucial not only as a reliable environmental indicator of physiological changes but also as one of the most important environmental cues of seasonal reproduction control in vertebrate species [33,34]. During the feeding management of broilers, several researchers have successfully regulated the sexual maturation and egg production traits by changing the means of photoperiod [35,36]. In general, most studies using ambulatory recordings of seasonal photoperiodic changes have been conducted between latitudes 32° N and 51° N [11,14,37]. The site of this experimental research was in that same region, and it exhibited a significant photoperiod change of more than 4 h between the summer and winter. This meant that the location was highly suitable for exploring seasonal variations in photoperiod and the neuroendocrine control of reproduction. In seasonally breeding animals, timing of reproduction in females can influence offspring development and survival [11], and the laying performance is an important economic factor in goose production. Therefore, female Yangzhou geese were used as the subjects of this study. We found that they showed a significant increase in egg rates with prolongation of the photoperiod, after reaching the peak of laying at the SE. This is in agreement with earlier data by Yang et al. [38], where the laying rate of breeding geese was highest in February and then started decreasing. On the basis of this characteristic, the female Yangzhou geese exhibited a marked seasonal breeding pattern, similar to that found in other long-day breeders. This was also consistent with previous findings that the reproduction season of Sichuan white geese lasts for approximately 7 months [39].

In addition, the growth and development of gonads were the critical seasonal indicators affecting the reproductive performance of poultry [40,41,42]. The gonads of seasonal breeding poultry development were accompanied by annual periods of quiescence and renaissance. During the breeding season, the HPG axis of birds was activated, resulting in a significant increase in the volume of the gonads, sometimes even more than 100-fold [42,43]. Interestingly, the HPG axis could automatically shut down after the end of the breeding season, and the gonads could also degenerate [11]. In this experiment, female geese had well-developed ovaries with distinct hierarchical follicles during the peak egg-laying period at the SE with an extended photoperiod, whereas the ovaries shrank at the SS and AE, before entering the ceased-laying period. Moreover, seasonal variation in GSI was also observed in female geese. Consistent with this observation, an increase in ovarian weight was associated with greater amounts of corpora lutea, thus indicating a higher reproductive performance [44]. Meanwhile, we found the GSI levels reached their peak at the SE with a long photoperiod. Notably, we found that the gonads degenerated gradually after breeding, even though the photoperiod still increased. This phenomenon is known as photorefractoriness, and the specific molecular mechanisms regulating this phenomenon are unclear at the moment [20]. Altogether, these findings might suggest a higher sophisticated photoperiodic mechanism driving the seasonal changes in HPG axis responsiveness in birds compared to other vertebrates.

The pineal gland of poultry is considered the most important light-sensing tissue, which exhibits a close association with the activity of the gonadal axis [45]. Studies have demonstrated that the pineal gland is an important neuroendocrine organ regulating reproduction by converting light-induced neural activity into endocrine hormones [46]. The pineal gland is thus defined as a photo-neuroendocrine converter and forms an essential part of the organism. It provides information about the photoperiod, thereby connecting the outside environments with the internal normal biochemical signaling and physiological needs of the body [47]. Typically, the size and activity of the pineal gland, which are related to the reproductive physiological state of animals, depend on environmental stimuli such as light stimulation. Seasonal breeding poultry mostly have a well-developed pineal gland [47]. Nonetheless, we observed a strange phenomenon in our study. We found that the weight and volume of the pineal gland in female geese showed seasonal changes contrary to gonadal activity, in accordance with similar results of Singh et al. (2007) in Perdicula asiatica [48]. Similar to most seasonal breeding mammalian species, there is an inverse relationship between the pineal gland and gonadal activity in some species of birds. Moreover, pineal functional modifications are also reflected in seasonal changes in structure. The histological structure of the goose pineal gland is similar to that observed by Wight et al. [45]. Pineal microvessels can offer vascular support for circannual periodic changes in the metabolic activity of the pineal tissue [49]. During the peak egg-laying period, along with the prolongation of the photoperiod came a decrease in the number of blood vessels around the connective tissue capsule of the pineal gland, resulting in a reduced secretory capacity. Therefore, the anatomical characteristics of the pineal gland under different photoperiods were closely related to the neuroendocrine regulation of seasonal reproduction in female geese.

The ultrastructure and functional changes in pineal cells also play an important role in regulating animal reproduction. In the experimental results of this study, the changes in the volume of the cell nucleus and the heterochromatin density might have been affected by the seasonal photoperiodic changes. As expected, on the basis of a previous study by Lee et al., there was a lower capacity of the pineal gland to secrete indoleamine hormones under a longer photoperiod, which resulted in promoting gonadal development [14]. Moreover, we observed through electron microscopy that the number of Golgi apparatus complexes increased slightly, gradually accumulating numerous vesicles at the AE. These characteristic morphological changes in the cell were suggestive of active secretory functions of the pineal gland, and the phenomena described here were consistent with the results reported by Frink et al. (1978) and McNulty et al. (1980) [50,51]. In addition, the current study confirmed that 5-HT is a fundamental component of melatonin biosynthesis [47]. Then, Aanat converts 5-HT into N-acetylserotonin, which is converted into melatonin by hydroxyindole-O-methyltransferase [52]. As such, we can only provide an educated guess here that the pineal cells might store a large amount of 5-hydroxytryptamine in order to synthesize melatonin under a short photoperiod of the AE. Meanwhile, glycogen was able to provide enough energy only for the physiological activities and biosynthesis of the cell [53]. More glycogen of pineal cells is needed in order to meet the consumption of melatonin synthesis. From this perspective, the ultrastructural changes in the pineal cells are an important indication of the seasonal physiological state of material synthesis and secretion.

Although the pineal gland is very important for seasonal breeding animals, melatonin needs to act on the hypothalamus–pituitary–gonadal axis by binding to the corresponding receptors [8]. Previous studies have confirmed that melatonin is intimately involved in regulation of the circadian rhythm, light signal transmission, and seasonality of the annual reproductive cycle [14,54]. In the present study, several notable findings were observed. For female geese, distribution analysis showed that the expression of three receptor genes had different seasonal changes in the pineal gland, hypothalamus, pituitary gland, and ovary. This is in contrast to the situation in mammals and fish, whereby Mel-1c was not detected in Atlantic salmon and mice, indicating adaptive evolution of organisms under varying environments. In addition, studies in seasonal breeding animals have found periodic changes in the expression levels of melatonin receptor genes concomitant with the photoperiod, and most of these have been reported in gonads (alongside some studies in brain tissue) [55]. Melatonin has been shown to act directly on the gonadal tissues [56,57,58]. In addition, the hypothalamus releases gonadotropin in response to melatonin, which indirectly inhibits its secretion [59]. In our study, the wide and specific tissue distribution of Mtnrs implied a series of processes regulated by melatonin. Alternatively, the three subtypes of Mtnrs (Mel-1a, Mel-1b, and Mel-1c) might have displayed a divergent distribution due to functionalization. However, unlike other nonseasonal breeding animals, changes in the expression levels of Mtnrs were present during the periodic changes, revealing the seasonal characteristics of melatonin in regulating reproduction.

Some studies on the control of sexual maturation in birds have explored the relationships between the seasonal photoperiod and the reproductive endocrine system [20]. Lee et al. observed an oscillation of melatonin levels with the natural pattern of seasonal photoperiod conditions [14]. In seasonal breeding animals, melatonin secretion fluctuates with photoperiod changes to regulate reproductive endocrine function. In addition, the ability of melatonin to regulate the reproduction of seasonal breeding animals has been proven several times in previous studies [60,61]. According to some studies, melatonin suppressed GnRH gene expression in GT1–7 cells and suppressed GnRH secretion by about 45%, indicating that melatonin could regulate GnRH neurons [62,63]. In the present study, we systematically assessed whether the rhythm of melatonin secretion could be correlated with seasonal changes in the neuroendocrine system at all levels of the HPG axis. Numerous studies have substantiated the crucial involvement of Gths, E2, and P4 in the developmental processes of gonads [64]. Among them, E2 assumes a pivotal role in various physiological functions encompassing growth, development, and reproduction [65]. Notably, it is plausible that the release of gonadal reserves of E2 could be triggered by GnRH [66]. In the present study, a lower level of melatonin regulated secretions of GnRH to stimulate secretions of LH, PRL, E2, and P4, along with gonadal development and reproductive activities, under a shorter photoperiod at the SE. Interestingly, the secretion of melatonin was inhibited by exposure to longer photoperiods, while the lowest amount of serum GnRH content was observed at the SS. There are results from the literature that also attest to this phenomenon. For instance, the findings by Yang et al. showed a close association of the reproductive hormones and their corresponding transcript expressions with the reproductive behaviors observed in geese [38]. Furthermore, some studies have observed that a high concentration of PRL could inhibit the secretion of pituitary gonadal hormone in the late laying period, leading to the interruption of reproductive behavior [11,67]. Afterward, the increase in melatonin content would further inhibit the activity of gonads with the increased daylength at the AE, bringing on the ceased-laying period for female geese. This suggests that the ocular melatonin signal is directly or indirectly related to reproduction. Further studies are needed to confirm the regulation methods underlying the reproductive state, but these findings suggest that, in female geese, melatonin can regulate seasonal reproduction upon receiving seasonal photoperiodic information.

This study was performed on 760-day-old female Yangzhou geese raised according to common farming practices by Yangzhou Tiange Goose Industry Development Co., Ltd., Yangzhou, Jiangsu Province, China (32°64′ N, 119°34′ E). The geese lived in natural environmental conditions regarding the photoperiod. All animal experiments were approved by the Ethics Committee on Animal Experiments of Yangzhou University (Approval Code: 121-2018., Government of Jiangsu Province, China). These experiments did not involve the utilization of any rare or endangered animals. The protocols and methods were in accordance with the Regulations for the Administration of Affairs Concerning Experimental Animals, approved by the State Council of the People’s Republic of China.

The female Yangzhou geese selected for the experiments were hatched on 19 February 2018, 23 May 2018, 24 August 2018, and 22 November 2018. All Yangzhou geese (42 females and 7 males) were fed the same diet throughout the duration of the study (Table S1), which combined coarse and concentrated material. Feed and water were provided ad libitum. The animals were tested at the spring equinox (SE), summer solstice (SS), autumn equinox (AE), and winter solstice (WS) in 2020, representing different seasonal patterns. The egg-laying rates of 42 female Yangzhou geese were recorded each day, and the photoperiod changes were also continuously monitored throughout the year. In each sampling season, six female geese were selected, and differences in the individual weight of the experimental animals in different seasons were within 3% of the mean weight (N = 6 per point). After blood sampling, serum was acquired via the centrifugation of blood samples at 3000× g for 15 min. Subsequently, the geese were sacrificed by anesthetizing them with sodium pentobarbital, before dissecting them to collect the pineal gland, hypothalamus, pituitary gland, and ovary, which were frozen in liquid nitrogen and stored at −80 °C until RNA extraction. Lastly, the pineal gland and ovary from the remaining six geese were collected and further subjected to histological observation.

The daily photoperiod is described as the change in day length, recorded throughout the year. In addition, the egg-laying rates of the geese were calculated on a daily basis as follows: egg-laying rate (%) = (number of eggs/total geese) × 100%. After MS-222 anesthesia, the sexually mature female geese (N = 6) were dissected to obtain the intact pineal gland and gonadal tissues in the different seasons, and the weight and gonadosomatic index (GSI) were measured. The latter was determined using the following formula: GSI (%) = [gonad weight (g)/total body weight (g)] × 100%. Next, samples from the pineal gland were fixed with 4% paraformaldehyde for 24 h at room temperature, dehydrated with graded ethanol, and embedded in paraffin. Sections of paraffin-embedded tissue (4 μm thick) were prepared for hematoxylin and eosin staining. The photographs were scanned using a Nanozoomer scanner (Hamamatsu, Sydney, Australia), and an image analysis system (Image-Pro Plus, Media Cybernetics, Rockville, MD, USA) was used to calculate the diameter of pineal gland follicles and the height of the pineal gland follicular cells for each sample.

For immunohistochemistry, the slides were dewaxed three times in xylene for 15 min each, and then heated with a microwave oven for antigen retrieval in Tris-ethylenediaminetetraacetic acid (EDTA) buffer (pH 9.0) at 95 °C for 20 min. The endogenous peroxidase activity was quenched with 3% H₂O₂ for 10 min. Then, the slides were incubated with the primary antibody anti-Mtnr (Mtnr, 1:1000, ab87639; Abcam, Cambridge, UK) at 4 °C overnight. The WB experiments were performed to verify the specificity of the Mtnr antibody (Figure S1). Slides were washed three times in phosphate-buffered saline (PBS) for 5 min each and incubated with the secondary antibody rabbit anti-mouse IgG (ab6728; Abcam, Cambridge, UK) for 30 min, followed by three additional washes in PBS and staining with 3,3′-diaminobenzidine (DAB, 1:20 dilution) for 10 min at room temperature. Lastly, after counterstaining with hematoxylin, samples were dehydrated in a graded ethyl alcohol series (70%, 90% and 100%), and cover slips were placed on. Bright-field and fluorescence images (GFP filter) of the stained sections were captured using an EVOS FLc imaging system (Thermo Fisher Scientific, Waltham, MA, USA). Image-Pro Plus (Image-Pro Plus, v6.0) was used to evaluate the mean integrated optical density (IOD) of the immunohistochemical results.

The pineal glands were placed in electron microscopy liquid fixative (0.1% paraformaldehyde in 0.1 M sodium cacodylate) for TEM section preparation. The tissues were removed after a minimum of 12 h of fixation. Slides of the pineal gland were dehydrated in an ascending ethanol series, and then transferred to propylene oxide and embedded in Epon (Sigma Aldrich, St. Louis, MO, USA). Semithin sections, stained with 1% toluidine blue (Reanal, Budapest, Hungary), were used for orientation; then, ultrathin sections were cut with a glass knife, mounted on grids, and allowed to dry. Uranyl acetate and lead citrate (Ted Pella Inc., Redding, CA, USA) were used as post-embedding staining. The slide samples were studied using a JEOL 1010 transmission electron microscope (JEOL, Peabody, MA, USA) operating at 80 kV.

The following experiments were conducted to measure gene expression: “experiment 1”—quantification of seasonal Aanat expression across different tissues; “experiment 2”—tissue distribution of Mtnrs expression in the four seasons; “experiment 3”—seasonal expression of genes relevant to the hypothalamus–pituitary–gonadal (HPG) axis.

Total RNA was extracted using Trizol (Invitrogen, San Diego, CA, USA) according to the instructions of the manufacturer. The concentration and purity of RNA samples were determined using a NanoDrop™ ND-1000 spectrophotometer (NanoDrop Technologies, Wilmington, DE, USA), and the RNA integrity was verified by electrophoresis on 1% agarose gel. One microgram of total RNA was reverse-transcribed with a FastQuant RT Kit (with gDNase) (Takara Biotechnology Co., Ltd., Dalian, China) following the supplier’s protocol. The RT reaction was performed as follows: the reaction mixture was incubated for 15 min at 42 °C to synthesize cDNA, followed by 3 min at 95 °C. The resulting cDNA samples were diluted fivefold and stored at −20 °C until further analysis. The primers for qRT-PCR (Table 2) were designed using Primer software (Version 5.0, Primer, Kingston, ON, Canada) and were synthesized by TSINGKE Biological Technology (Nanjing, China). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as the reference control gene to normalize target gene expression. Then, qRT-PCR was carried out using SYBR Green Master Mix (ABclonal, Wuhan, China), and data were analyzed in Quant Studio 5 (Applied Biosystems, Thermo Fisher Scientific, Waltham, MA, USA). The following qRT-PCR thermal cycling program was employed: 95 °C for 5 min, and then 41 cycles of 95 °C for 20 s and 60 °C for 30 s, which included data acquisition. Relative expression levels were determined using the 2^−ΔΔCt method, where the ΔCt value was derived as a function of the difference between the Ct value of each tested gene and that of the reference gene. Each reaction was performed in triplicate, and the data were the average of three independent experiments.

For the estimation of hormones, blood was centrifuged, and serum was collected for the analysis of melatonin (MT), gonadotropin-releasing hormone (GnRH), follicle-stimulating hormone (FSH), luteinizing hormone (LH), prolactin (PRL), estradiol (E2), and progesterone (P4) levels using commercially available enzyme-linked immunosorbent assay (ELISA) kits (Wuhan EIAab Science Co. Ltd., Wuhan, Hubei, China). Operations were carried out according to the manufacturer’s protocols. The sensitivities of the ELISAs were 10.0 pg/mL, 10.0 pg/mL, 0.56 mIU/mL, 0.19 mIU/mL, 0.195 pg/mL, 0.056 ng/mL, and 0.15 ng/mL, respectively. There was no cross-reaction with other structural analogues, and all the intra-assay and inter-assay coefficients of variation (CVs) for each hormonal assay were less than 10% and 15%, respectively. Lastly, the absorbance of 450 nm was read for each well using a microplate reader, Model 680 (BioRad, Hercules, CA, USA). The ELISA standard curves were analyzed using a four-parameter logistic equation:

Standard curves with R² ≥ 0.98 were accepted.

Results were expressed as the mean ± SEM (standard error of the mean). All data were checked for normality using the Kolmogorov–Smirnov test of normality before being analyzed. Depending on the design of the experiment, data were analyzed using a one-way (experiments 1 and 3, as well as hormone and histological experiments) or two-way ANOVA (experiment 2), followed by Tukey’s post hoc test. Significance was set at the 0.05 level (p < 0.05). Statistical analyses were performed using SPSS 20.0 software package (IBM Corp, Armonk, NY, USA).

The data presented in this study are available upon request from the corresponding author.