The melatonin system is expressed in the ovine uterus: effect of the day of the oestrous cycle and undernutrition

Animals

Two experiments that involved multiparous adult Rasa Aragonesa ewes were performed at the experimental farm of the University of Zaragoza (Zaragoza, Spain; 41°41′N, 0°52′W) and followed a protocol approved by the Ethics Committee of the University of Zaragoza, which met the requirements of the European Union for Scientific Procedure Establishments.

RNA isolation and reverse transcription

A Quick-RNA Miniprep kit (Zymo Research, Orange, CA, USA) was used to extract total RNA. The concentration of RNA was measured based on its absorbance at 260 nm, purity was assessed based on the ratio of absorbance at 260/280 nm, and the integrity of the RNA was assessed by electrophoresis on a 1% agarose gel. All samples had A260/280 ratios between 1.8 and 2.1. A NZY M-MuLV First-Strand cDNA Synthesis Kit (NZYTech, Lisboa, Portugal) was used to reverse transcribe 1 μg total RNA to cDNA (one reaction per sample).

Quantitative real-time PCR (qPCR)

Quantitative real-time PCRs were performed using the NZYSpeedy qPCR Green master mix (NZYTech), with equimolar amounts of forward and reverse primers (400 nM, Table 1) and 2 μL cDNA. Samples were run in duplicate in two independent runs. Standard amplification conditions were 2 min at 95°C, and 40 cycles of 5 s at 95°C and 30 s at 60°C. At the end of each run, dissociation curves were analysed to confirm amplicon specificity and absence of contamination and or primer dimers. Ovis aries-specific intron-spanning primers were designed using the NCBI primer designing tool (https://www.ncbi.nlm.nih.gov/tools/primer-blast/), and checked by BLAST analysis to verify gene specificity (Table 1). qPCR data was analysed by the 2^−ΔΔCT method (Livak and Schmittgen 2001), where the changes in expression of the target genes relative to the housekeeping gene β-actin, were normalised with respect to a calibrator (pooled uterine cDNA from samples of each experiment).

Classic PCR

Classic PCR confirmed AANAT and ASMT mRNA in the ovine uterus. Reactions were carried out using NZYTaq II 2x Green Master Mix (NZYTech), forward and reverse primers at 250 nM, and 1.5 μL cDNA. Standard amplification conditions were 3 min at 95°C, followed by 35 cycles of 30 s at 94°C, 30 s at 58°C, and 15 s at 72°C, with a final extension of 5 min at 72°C. Primers were the same as those used for qPCR (Table 1). Amplified products were visualised by electrophoresis in 2% agarose gel.

Western blot

For protein extraction, samples of endometrium were mechanically disrupted in a radioimmunoprecipitation assay buffer (50 mM Tris pH 8, 150 mM NaCl, 0.1% SDS, 0.5% deoxycholate, 1% Triton X-100, and protease inhibitor cocktail, Sigma-Aldrich, St. Louis, MO, USA). After a 2 h incubation at 4°C, samples were centrifuged and supernatants recovered and stored at −20°C. Total protein concentration was determined based on a BCA (bicinchoninic acid) Protein Assay kit (Thermo Fisher Scientific, MA, USA).

For immunoblotting, 50 μg of protein lysate were separated in a 12% SDS-PAGE gel and transferred onto PVDF membranes by a Trans-Blot Turbo Transfer System (Bio-Rad, Hercules, CA). After blocking with 5% non-fat skim milk in 1% v/v Tween-20 PBS for 1 h, membranes were probed with polyclonal rabbit anti-AANAT (1:1000, Abcam, Cambridge, UK, Cat# ab3505) or anti-ASMT antibodies (1:200, Bioss Antibodies, Woburn, MA, USA, Cat# BS-6961R) overnight at 4°C. Following a 1 h incubation with a peroxidase-conjugated secondary goat anti-rabbit antibody (Santa Cruz Biotechnology, CA, USA), immunoblots were exposed to an enhanced chemiluminescence substrate (Clarity Western ECL substrate, Bio-Rad) and scanned in a VersaDoc imaging system (Bio-Rad). Each blot was performed twice. Densitometric quantification was performed by the Quantity One software (Bio-Rad).

Immunofluorescence

Tissue localisation of AANAT and ASMT was assessed by indirect immunofluorescence on paraffin 5-μm uterine transverse sections following the protocol of Zaqout et al. (2020), with slight modifications. Briefly, deparaffinised sections were exposed to a heat-mediated antigen retrieval with 0.01 M sodium citrate (pH 6.0), and blocked with 5% bovine serum albumin in 1X PBS/0.25% v/v Triton X-100 for 1 h at room temperature (RT). Rabbit anti-AANAT (1:100, Abcam, Cambridge, UK) or rabbit anti-ASMT (1:100, Bioss Antibodies, Woburn, MA, USA) were then applied to samples and incubated overnight in a humid chamber at 4°C. Negative controls were not incubated with primary antibodies. Sections were then rinsed with 1X PBS, incubated with Alexa Fluor 488 goat anti-rabbit IgG (1:400, Invitrogen, Carlsbad, CA, USA) for 1 h at RT, rinsed again, and treated with a solution containing 10 mM CuSO₄ and 50 mM NH₄Cl (pH 5) to reduce autofluorescence. Finally, the sections were washed with dd-H₂O and mounted with ProLong Gold antifade reagent with DAPI (4′,6-diamidino-2-phenylindole) (Invitrogen).

Statistical analysis

Normality was tested by Shapiro–Wilk tests. The differences between days of the oestrous cycle were assessed based on a one-way ANOVA and a Tukey’s multiple comparison test. Pairwise comparisons between nutritional treatments were assessed by a t-test. All statistical analyses were performed in GraphPad Prism 9.0 (GraphPad Software Inc., San Diego, CA, USA). Results are presented as mean ± s.e.m. and were considered statistically different if P ≤ 0.05, and as tendency to differ if 0.05 < P ≤ 0.10.

AANAT and ASMT are expressed in the ovine uterus

The expression of AANAT and ASMT mRNA in the sheep endometrium was demonstrated based on specific primers for quantitative (Figs 1a and 2a) and classic PCR (Figs 1b and 2b). Sheep pineal gland served as a positive control, and it had the highest expression levels of the two genes. In qPCR, the threshold cycle (Ct) values obtained for AANAT and HIOMT pineal expression were approximately 16, whereas in the endometrium they ranged between 28 and 32. The PCR products from the uterine samples were unambiguously at the expected size for each primer set and of the same size as that of pineal gland. Uterine protein expression for AANAT and ASMT was confirmed by Western blot and immunofluorescence after binding to specific antibodies. AANAT displayed a clear band of about 23 kDa (compatible with the expected size of the protein) in the pineal and uterine samples and, in most cases, a less intense band at about 18 kDa (Fig. 1c). ASMT produced a single band of about 45 kDa, compatible with the expected molecular weight of ASMT, and a less intense band in the pineal gland and in the uterus of remarkable expression (Fig. 2c). The expression of the two proteins occurred at all stages of the oestrous cycle (Experiment 1, Figs 1 and 2) and under nutritional restriction (Experiment 2).

Fig. 1.

Arylalkylamine N-acetyltransferase (AANAT) expression in Rasa Aragonesa sheep endometrium on days 0 (oestrus), 5, 10 and 14 of the oestrous cycle. Expression in pineal gland (P) served as a positive control. (a) AANAT mRNA relative expression as determined by qPCR (n = 3); (b) mRNA expression by classic PCR (M = MW ladder; W = water); (c) Western blot and densitogram analysis of AANAT protein expression (n = 2). *P < 0.05.

Fig. 2.

N-acetylserotonin-O-methyltransferase (ASMT) expression in Rasa Aragonesa sheep endometrium on days 0 (oestrus), 5, 10 and 14 of the oestrous cycle. Expression in pineal gland (P) served as a positive control. (a) ASMT mRNA relative expression as determined by qPCR (n = 3); (b) mRNA expression by classic PCR (M = MW ladder; W = water); (c) Western blot and densitogram analysis of ASMT protein expression (n = 2). *P < 0.05.

As revealed by immunofluorescence, both AANAT and ASMT were expressed cytoplasmically mainly in the luminal epithelium and stratum compactum (i.e. glandular epithelia and stroma next to the luminal epithelium, Fig. 3), but were almost absent in the glands and stroma near myometrium. We noted that AANAT was strongly detected in the apical membrane of epithelial cells, mostly in the superficial glands, and to a lesser extent, in the luminal epithelium. In contrast, ASMT was more evenly distributed in the epithelial cells.

Fig. 3.

Immunofluorescence staining of AANAT and ASMT proteins in the ovine endometrium. Representative microphotographs of day 10 of the oestrous cycle. AANAT and ASMT (green) were localised cytoplasmically mainly in the superficial glandular epithelium (GE), superficial stroma (S) and luminal epithelium (LE). Nuclei were counterstained with DAPI (blue). No specific staining (green colour) was observed in the negative control (no primary antibody). Scale bar = 50 μm for all microphotographs.

Effects of the oestrous cycle

Gene expressions of AANAT and ASMT were affected by the days of the oestrous cycle (P < 0.05), with higher concentrations at day 10 compared to day 14 (P < 0.05, Figs 1a and 2a). The protein expression of AANAT also seemed to be higher at day 10 than at day 14 (Fig. 1c), but the low sample number per day (n = 2) precluded reliable statistical analyses. In turn, ASMT protein level did not seem to vary (Fig. 2c).

For the melatonin receptors, the days of the oestrous cycle did not have a significant effect on the gene expression of MT1, but it affected the abundance of MT2 mRNA (P < 0.001), being higher at day 10 than at days 0, 5 and 14 (P < 0.01, Fig. 4).

Fig. 4.

Gene expression as assessed by qPCR of melatonin receptors and metabolising enzymes in the endometrium of Rasa Aragonesa sheep on days 0 (oestrus), 5, 10 and 14 of the oestrous cycle (n = 3). MT1 and 2, melatonin receptors 1 and 2; IDO1 and 2, indoleamine 2,3-dioxygenase isoforms 1 and 2; MPO, myeloperoxidase. *P < 0.05; **P < 0.01; ***P < 0.001.

The oestrous cycle affected some of the melatonin-catabolising enzymes; specifically, IDO1 (P < 0.01) and MPO (P < 0.05, Fig. 4). Gene expression of IDO1 was upregulated from oestrus (day 0) to day 5 (P < 0.01) and to day 10 (P < 0.05), and decreased on day 14 (P < 0.05); however, the oestrous cycle did not regulate the isoform IDO2. For MPO, mRNA expression was lowest on day 0 and was significantly (P < 0.05) higher at day 5.

Effects of undernutrition

Undernutrition significantly (P < 0.01) upregulated AANAT transcript expression (Fig. 5a); however, protein expression of AANAT was significantly (P < 0.05) lower in undernourished ewes (Fig. 5c). ASMT was unaffected by the nutritional treatment at either the gene (Fig. 5b) or the protein level (Fig. 5d). Undernutrition did not influence the gene expression of MT1, but it tended (P = 0.06) to increase MT2 mRNA expression (Fig. 6). Similarly, undernutrition increased transcript expression of IDO2 (P < 0.05), but did not have a significant effect on IDO1 or MPO expression (Fig. 6).

Fig. 5.

Arylalkylamine N-acetyltransferase (AANAT) and N-acetylserotonin-O-methyltransferase (ASMT) uterine expression in Control (1.5 times the daily requirements for maintenance) and Low (undernourished, 0.5 times the daily requirements for maintenance) Rasa Aragonesa ewes on day 14 of the oestrous cycle. Expression in pineal gland (P) was used as a positive control. (a) AANAT and (b) ASMT mRNA expression as determined by qPCR (relative to β-actin) (n = 4). (c) AANAT and (d) ASMT Western blot and densitometry analysis of protein expression (n = 4). *P < 0.05; **P < 0.01.

Fig. 6.

Uterine mRNA expression of melatonin receptors and metabolising enzymes in Control (1.5 times the daily requirements for maintenance) and Low (undernourished, 0.5 times the daily requirements for maintenance) Rasa Aragonesa ewes on day 14 of the oestrous cycle (n = 4). MT1 and 2, melatonin receptors 1 and 2; IDO1 and 2, indoleamine 2,3-dioxygenase isoforms 1 and 2; MPO, myeloperoxidase. *P < 0.05; ^#P = 0.06.

Effects of the oestrous cycle

Uterine physiology is under the cyclic regulation of ovary steroid hormones; therefore, we investigated whether the melatonin system was regulated as well. Gene expression of AANAT varied throughout the oestrous cycle, and seemed to be paralleled by the protein expression detected by Western blots. Although ASMT gene expression also changed during the oestrous cycle, protein expression did not seem to vary, which is consistent with a rather constitutive expression of this enzyme that occurs in the pineal gland in which ASMT exhibits little change in expression in response to dark and light (Ganguly et al. 2002). Expression of the AANAT and ASMT genes, and presumably the AANAT protein, was greatest in the late luteal phase (day 10), and lowest at oestrous and at the initiation of luteolysis (day 14), which suggests that oestrogen and or progesterone regulate these proteins. In the uterus of gilts, AANAT mRNA was upregulated at the end of the luteal phase (Bae et al. 2020). An inhibitory effect of oestrogens on melatonin synthesis has been demonstrated in rat pineal gland (Cardinali et al. 1974; Hernández-Díaz et al. 2001). Although, to our knowledge, that has not been demonstrated in extrapineal tissues, the results of our study indicated the lowest melatonin synthesis around oestrous, when oestrogen levels are highest. On the other hand, the upregulation of melatonin synthetic enzymes in the progesterone prevailing phase of the oestrous cycle is consistent with the roles of melatonin in early embryo development and implantation, when progesterone levels remain high in order to sustain pregnancy. Indeed, Bae et al. (2020) observed an increase in the gene expression of AANAT and ASMT in early pregnancy in the endometrium of gilts. Furthermore, crosstalk between melatonin and steroid hormones might occur (Cipolla-Neto et al. 2022), although its importance in the uterus has not been investigated thoroughly. In the ovine endometrium, for instance, exogenous melatonin seems to regulate progesterone receptor expression (Vázquez et al. 2013), although the mechanisms underlying this effect remain to be assessed.

In our study, the uterine expression of MT2 mRNA was influenced by the day of the oestrous cycle since there was a pronounced increase on day 10. The variation in melatonin receptor expression during the oestrous cycle was specific, as no differences were observed in MT1. In general, the results were similar to those reported in the uterus of gilts (Bae et al. 2020) because expression of MT1 remained constant, while expression of MT2 was highest in the late luteal phase. Oestradiol reduces the expression of the two melatonin receptors in the sheep oviduct (Hu et al. 2020). In line with our results, treatment with progesterone increased melatonin uterine binding in ovariectomised rats (Zhao et al. 2002). However, oestradiol showed a similar effect, which is difficult to reconcile with the low expression of MT2 mRNA at oestrus in the sheep uterus, although differences in species and experimental protocols probably account for the discrepancies. Indeed, to our knowledge, this is the first study to investigate MTs expression in the ovine uterus throughout the oestrous cycle.

Whether the strength of gene expression reflects protein abundance at the receptors is unknown but, if these findings corroborate at the protein level, the results of our study suggest an increase in uterine synthesis and sensitivity to melatonin in the luteal phase. In mice, melatonin acting via MT2 regulates the endometrial structure and glandular density, which improves uterus receptivity and suggests that melatonin is involved in the endometrial preparation for early gestation (He et al. 2015), and similar phenomena might occur in the ovine uterus. Nevertheless, interpretation of the results of our study is not straightforward because the binding of melatonin to each receptor subtype might activate different signalling pathways, and melatonin receptors can act individually or by forming homo- or heterodimers, and several events can regulate signal transduction (Dubocovich et al. 2010). Furthermore, another factor is the genetic variation of the melatonin receptors (Jockers et al. 2016). For instance, the polymorphisms of the MTNR1A gene sequence can influence reproductive performance in ewes and rams (Carcangiu et al. 2009; Abecia et al. 2020b; Starič et al. 2020).

Unlike melatonin synthesis, melatonin catabolism is less well understood and much more complex because it includes enzymatic, pseudoenzymatic, and direct transformations that occur after interactions with oxygen- and nitrogen-reactive species (Tan et al. 2015). Primarily, two isoforms of the indoleamine 2,3-dioxygenase (IDO1 and IDO2) and myeloperoxidase (MPO) are involved in the enzymatic cleavage of melatonin (Tan et al. 2007; Fatokun et al. 2013); however, neither IDO or MPO exhibit specificity for melatonin and have high affinity for a wide range of substrates including tryptophan (Ferry et al. 2005). In our study, the day of the oestrous cycle affected the gene expression of IDO1 and MPO in the sheep endometrium, with upregulation from oestrus to day 5, which was maintained up to day 10 in the case of IDO1. These results suggest a high rate of removal of melatonin from the cells, but also an increase in the bioactive melatonin metabolites. Enzymatic melatonin catabolism produces the metabolite N1-acetyl-N2-formyl-5-methoxykynuramine and its derivatives, which act as antioxidants and might mediate and amplify some of the functions of melatonin (Tan et al. 2007). In mammals, IDO1 acts as an immunomodulatory molecule that defends against infections and tumour invasion (Fatokun et al. 2013). In the uterus, that role might be important in the preparation of the endometrium for the arrival of an embryo. Indeed, IDO1 might contribute to the establishment of immune tolerance in early pregnancy so that the embryo is not rejected (Mellor et al. 2002; Ott 2019). The role of those enzymes in uterine physiology has not been identified and, probably, several individual and combined actions occur.

Effects of undernutrition

Undernutrition can impair reproduction in sheep (Abecia et al. 2006; Meikle et al. 2018) by delaying embryo development, increasing embryo mortality, and altering uterine gene expression in the oestrous cycle and early pregnancy (Abecia et al. 2006; Sosa et al. 2009; de Brun et al. 2021). Exogenous melatonin improves reproductive outcomes (Abecia et al. 2008, 2019), and modulates oviductal gene expression and uterine sensitivity to progesterone (Vázquez et al. 2013, 2019). Therefore, we sought to determine whether the uterine melatonin system is involved in these effects. Undernutrition to half maintenance energy requirements affected the expression of AANAT on day 14 of the oestrous cycle, but gene and protein expressions were uncoupled. Undernutrition increased AANAT gene expression, but the protein expression was reduced. Others have reported that changes in AANAT mRNA had no effect on AANAT protein expression or activity in cells from bovine pineal gland (Schomerus et al. 2000). Rather, AANAT activity seems to be influenced by changes in protein expression and other activating mechanisms such as cyclic AMP (Ganguly et al. 2002). The reduced AANAT protein expression in undernourished ewes in the late luteal phase might reduce endometrial melatonin production. Given the important roles of melatonin in uterine physiology, in the establishment of a receptive endometrium and or in embryo development, implantation, and survival (Carlomagno et al. 2018; Olcese 2020), this finding might contribute to the detrimental effects of undernutrition on reproduction in ewes. In our study, undernutrition did not have a similar effect on ASMT expression at the gene or at the protein level. Similarly, ASMT mRNA was higher at day 10 of the oestrous cycle, but its protein expression remained unchanged, showing stable integral expression.

In our study, uterine expression of MT2 mRNA was higher in undernourished ewes than it was in well-fed ewes, but MT1 gene expression did not differ significantly which suggests, perhaps, increased sensitivity to melatonin in the uterus of undernourished ewes by day 14 of the oestrous cycle. Possibly, this is a compensatory mechanism in response to reduced uterine melatonin (as reflected in the reduced expression of AANAT in these sheep), given melatonin’s roles in uterine remodellation, which is vital at the time of maternal recognition of pregnancy (i.e. day 14).

In our study, undernutrition increased the expression of IDO2 mRNA in the uterus of sheep by day 14 of the oestrous cycle, which suggests an increase in melatonin catabolism that in addition to the impairment of melatonin synthesis (i.e. reduced AANAT expression), might result in a reduction in the melatonin available in the uterus of undernourished sheep. Furthermore, these results might explain why the treatment of undernourished ewes with exogenous melatonin in early pregnancy improves embryo survival (Vázquez et al. 2013; Abecia et al. 2019). Of the two indoleamine 2,3-dioxygenase isoforms, IDO2 was discovered most recently and less is known about its regulation and functions. IDO2 is thought to have less enzymatic activity than IDO1, and it might have non-enzymatic functions (Fatokun et al. 2013; Merlo, Mandik-Nayak 2016). Both isoforms serve immunomodulatory functions but, unlike IDO1, IDO2 mediates inflammatory responses (Merlo et al. 2022). In mice uterine stromal cells, IDO2 suppresses proliferation and promotes apoptosis (Li et al. 2015). Possibly, an increase in IDO2 in the uterus of undernourished sheep contributes to an unsuitable environment for embryo implantation, although the underlying mechanisms are unknown.

In conclusion, in this study, we have provided evidence of the expression of the melatonin system (i.e. receptors and synthetic and catabolic enzymes) in the endometrium of sheep throughout the oestrous cycle. The phase of the oestrous cycle affected the expression of most of the molecules studied, which suggests that they are under the influence of sexual steroid hormones, and links the melatonin system to the uterine physiology. Undernutrition altered the normal uterine pattern of expression of genes involved in sensitivity to melatonin and in the melatonin synthetic and catabolic pathways, which might help to explain the adverse effects of undernutrition on reproduction in sheep and the success of exogenous melatonin treatments for improving reproductive outcomes.