Follicle-stimulating hormone administration affects amino acid metabolism in mammalian oocytes^†

Abstract

Culture media used in assisted reproduction are commonly supplemented with gonadotropin hormones to support the nuclear and cytoplasmic maturation of in vitro matured oocytes. However, the effect of gonadotropins on protein synthesis in oocytes is yet to be fully understood. As published data have previously documented a positive in vitro effect of follicle-stimulating hormone (FSH) on cytoplasmic maturation, we exposed mouse denuded oocytes to FSH in order to evaluate the changes in global protein synthesis. We found that dose-dependent administration of FSH resulted in a decrease of methionine incorporation into de novo synthesized proteins in denuded mouse oocytes and oocytes cultured in cumulus-oocyte complexes. Similarly, FSH influenced methionine incorporation in additional mammalian species including human. Furthermore, we showed the expression of FSH-receptor protein in oocytes. We found that major translational regulators were not affected by FSH treatment; however, the amino acid uptake became impaired. We propose that the effect of FSH treatment on amino acid uptake is influenced by FSH receptor with the effect on oocyte metabolism and physiology.

Introduction

Conditions of in vitro maturation (IVM) influence the nuclear and cytoplasmic maturation of oocytes [1, 2]. In order to support the nuclear and cytoplasmic IVM of oocytes, culture media used in programs of assisted reproduction are commonly supplemented with gonadotropin hormones. Follicle-stimulating hormone (FSH), a pituitary gonadotropin glycoprotein hormone, regulates a number of transcriptional and metabolic events in the ovary that are essential for proliferation and differentiation during follicular growth and oocyte maturation [3, 4]. FSH acts through the FSH receptor (FSHR), a G protein coupled receptor. It is generally accepted that FSHR is expressed exclusively in the granulosa cells in ovarian follicles and in testicular Sertoli cells [5]. Activated FSHR stimulates many intracellular signaling pathways including events initiated by adenylyl cyclase activation, followed by the induction of cyclic adenosine monophosphate (cAMP), protein kinase A activation, and protein phosphorylation [6, 7]. The binding of FSH to its receptor is also implicated in intracellular calcium increase, mitogen-activated protein kinase (MAPK) activation, and inositol triphosphate stimulation [8].

During the growth phase of development, oocytes accumulate macromolecules in order to cease transcription at the completion of this stage. Gene expression in fully grown oocytes is then regulated based on the level of mRNA stabilization and translation [9]. Transcription remains suppressed during the meiotic progression of the oocyte, as well as during fertilization and early embryo development, until a species-specific time of embryonic genome activation [10, 11]. In particular, the process of protein translation in oocytes is controlled by the phosphorylation/dephosphorylation of eukaryotic initiation factors (eIFs) and their regulators [12–15]. Protein translation in germinal vesicle (GV)-stage oocytes is at low levels, while during germinal vesicle breakdown (GVBD), a three-fold increase of protein synthesis is preceded by the phosphorylation of eIF4E. In oocytes at the metaphase II (MII) stage, protein translation drops to low levels compared to the rate occurring at the GV-stage [15].

Protein synthesis in oocytes during meiotic maturation is crucial for the completion of meiosis [16] and pronuclear development in porcine fertilized oocytes [17, 18]. The results of [³⁵S]-methionine incorporation into ovine cumulus-oocyte complexes (COCs) suggested a more intense protein synthesis in oocytes exhibiting higher developmental competence [19]. It has been shown that protein translation of maternal mRNAs was enhanced in mouse oocytes and embryo development was improved when COCs were subjected to FSH in vitro (10 ng/ml) [20]. On the other hand, the negative effect of administration of gonadotropin in vivo on early embryo development has also been reported before [21–24] as well as in vitro treatment of oocytes with recombinant FSH (Gonal-F) inducing a lower developmental competence of early embryos in vitro [25, 26].

The expression of FSHR in the mammalian oocyte and in connection with the direct effect of FSH on oocyte physiology is controversial. Although IVM protocols use FSH, the effect of gonadotropins on oocytes is not well understood. We have found that FSHR is expressed in the mammalian oocytes and FSH shows effect on amino acid uptake in oocytes of various mammalian species including human.

Materials and methods

Ethics Statement

All animal work was conducted according to Act No. 246/1992 on the protection of animals against cruelty, issued by experimental project #215/2011, certificate #CZ02389. Bovine and porcine ovaries were obtained from local slaughterhouses where they are discarded without utilization (hence no ethics statement was required). Surplus human oocytes were provided for research only when written informed consent was obtained.

Oocyte isolation and IVM

Mice (CD1 strain) were stimulated with pregnant mare serum gonadotropin (PMSG, Folligon, Merck Animal Health) 46 h prior to oocyte isolation; 5 IU per mouse. Oocytes were isolated by disrupting the ovaries into transfer media [27] supplemented with 100 μM 3-isobutyl-1-methylxanthine (IBMX, Sigma Aldrich) to prevent spontaneous meiotic resumption. Isolated mouse COCs or oocytes deprived of cumulus oocytes (denuded oocytes, DOs) were cultured in the presence of IBMX for 2 h in M16 medium supplemented with either 10 ng/ml (0.136 IU/ml) or 100 ng/ml (1.36 IU/ml) of FSH (Gonal-F; Serono Laboratories; Puregon, N.V. Organon) diluted in M16 medium (Millipore). Culture was performed in M16 medium pre-equilibrated at 37.5 °C and 5% CO₂. For IVM, isolated DOs were washed with IBMX and cultured for 12 h to MII-stage in M16 medium.

Follicles with 5–9 mm (bovine) and 3–5 mm (porcine) in diameter were dissected and punctured to isolate the oocytes. Bovine and porcine COCs were evaluated and selected according to the morphology of the cumuli. COCs with at least three layers of compact cumulus cells (CCs) were used for the experiments. Earlier, the culture COCs were deprived of CCs using hand micropipette (bovine) or by vortexing for ∼7 min (porcine). Bovine and porcine DOs (GV oocytes) were subsequently cultured in M-16 medium supplemented with 100 ng/ml FSH (Gonal-F) for 2 h. See also experimental schemes in Figure 1.

Figure 1

Experimental design schemes. (A) Labeling of GV oocytes with [³⁵S]-methionine in the presence (+) or absence (−) of FSH. (B) Labeling of MII oocytes with [³⁵S]-methionine. Isolated DOs were cultured with presence (+) or absence (−) of FSH and then labeled with [³⁵S]-methionine for 2 h. (C) Labeling of 2cell embryos with [³⁵S]-methionine. In vivo isolated zygotes were cultured for 22 h to 2cell stage. 2cell embryos were labeled with [³⁵S]-methionine for 2 h with presence (+) or absence (−) of FSH. (D) Scheme of culture of COCs and DOs. DOs were stripped from CCs prior culture. Cells from COCs were separated after culture. All cells types were frozen separately.

The collection of human oocytes was carried out in IVF center Reprofit International (Brno, Czech Republic). Ovarian stimulation and oocyte retrieval were performed as described earlier [28]. Donated immature oocytes were incubated overnight in continuous single culture (CSC) medium (#90165, Irvine Scientific, USA) at 37 °C in a humidified atmosphere of 5% O₂ and 6% CO₂. Next day, the developmental maturity of each oocyte was confirmed by the presence of a polar body and MII spindle (Octax polarAIDE, MTG, Germany). A total of 15 MII oocytes from 7 IVF patients and 6 egg donors (average age 30.15 years) were frozen using VT801 vitrification media and Cryotop—open system (Kitazato BioPharma, Japan) according to the manufacturer’s instructions and stored in liquid nitrogen (LN2) until thawing. Frozen oocytes were thawed using VT802 warming media (Kitazato BioPharma, Japan) according to the manufacturer’s instructions. After thawing, oocytes were incubated for 2 h in CSC medium at 37 °C in a humidified atmosphere of 5% CO₂. Next, oocytes were further processed for [³⁵S]-methionine labeling (Figure 1B). The use of spare human oocytes for research was approved by the Institutional Ethics Committee of Reprofit International (# 1/2015) and Faculty of Medicine, Masaryk University, Brno, Czech Republic (# 16/2016).

In vivo MII oocytes and 2cell embryos

To obtain embryos, mice were primed with 5 IU of human chorionic gonadotropin (hCG, Pregnyl, N.V. Organon) 46 h after PMSG administration and mated with males. Zygotes were collected from fallopian tubes 17 h after mating and cultured to the 2cell stage for 24 h in M16 medium at 37.5 °C under 5% CO₂. Subsequently, 2cell embryos were treated for 2 h with 100 ng/ml FSH (Serono Laboratories) diluted in M16 medium.

Natural stimulation of the mice

For natural stimulation, females that are previously not exposed to males were housed on bedding containing male urine and feces for three consecutive days prior to hCG administration. The bedding was changed daily in order to maintain the level of male pheromones. Mice were injected with 5 IU of hCG on the third day of natural stimulation.

Metaphase II oocytes from naturally stimulated mice were compared with in vivo maturated MIIs. Females were primed with 5 IU of PMSG and subsequently after 46 h with 5 IU of hCG. In both cases, MII oocytes were obtained by puncturing oviducts 17 h after hCG administration. Cumulus cells were removed by hyaluronidase (H-3506, Sigma Aldrich) treatment for 30 min and the immunocytochemistry (ICC) protocol was then applied to the DOs.

[³⁵S]-methionine labeling

To analyze de novo protein synthesis, oocytes and early embryos were cultured either in the absence or presence of FSH as well as of the protein synthesis markers [³⁵S]-methionine or homopropargylglycine (HPG) for 2 h. To measure de novo protein synthesis, oocytes and embryos were exposed to 25 μCi/ml of [³⁵S]-methionine (Hartmann analytics) for 2 h. Samples were then washed in polyvinyl alcohol (PVA)/phosphate buffer saline (PBS) and stored in −80 °C prior to usage. Lysed samples were subjected to SDS-polyacrylamide gel electrophoresis (PAGE) and transferred to Immobilon-P membrane using a semidry blotting system for 25 min at 5 mA/cm² (the same method as immunoblotting). The labeled proteins were visualized by autoradiography on FujiFilm membrane (exposed for at least 7 days at −80 °C), scanned using BAS-2500 Photo Scanner (FujiFilm Life Science) and quantified by ImageJ software (http://rsbweb.nih.gov/ij/). Western blotting with glyceraldehyde 3-phosphate dehydrogenase (GAPDH) antibody was used as a loading control.

In situ translation

For nascent protein synthesis, oocytes were cultured in M16 with 50 μM L-HPG (C10186; Thermo Fisher Scientific) for 2 h in the presence or absence of FSH. For detection of HPG influx, cells were treated with 100 μM CHX for 2 h in M16. Oocytes were fixed in 4% paraformaldehyde (PFA) in PVA/PBS for 15 min. Homopropargylglycine was detected using Click-iT Cell Reaction Kit (Thermo Fisher Scientific). 4',6-diamidin-2-fenylindol (DAPI) was used for chromosome staining (H-1500; Vector Laboratories). Samples were visualized using an inverted confocal microscope in 16-bit depth (TCS SP5; Leica). Images were assembled in Photoshop CS3 and quantified by Image J software.

Immunocytochemistry

For protein visualization, oocytes and embryos were fixed for 15 min in 4% PFA (Sigma Aldrich) in PBS. Fixed oocytes were permeabilized in 0.1% Triton X-100 for 10 min, washed in PBS supplemented with PVA (Sigma Aldrich) and then incubated with primary antibodies overnight at 4 °C. Primary antibodies are listed in Supplementary Table 1. Oocytes were then washed 2× 15 min in PVA/PBS and the detection of primary antibodies was performed using relevant Alexa Fluor 488, 594 conjugates (Invitrogen) diluted 1:250, 1 h at room temperature. Washed oocytes (2× 15 min in PVA/PBS) were then mounted in Vectashield Mounting Medium with DAPI (Vector Laboratories).

Mouse ovaries and thigh muscles were mounted in tissue freezing medium (#14020108926, Leica), frozen in liquid nitrogen, and subsequently cut in cryotome (Leica CM1850). Slices of tissues were stored at −20 °C and fixed for 15 min in 4% PFA prior staining. A similar protocol for immuno-staining as mentioned above was followed, with washes in PBS. Rabbit polyclonal anti-FSH-R antibody with secondary antibody Alexa Fluor 594 conjugate were used. Phalloidin Alexa 488 conjugate (Thermo Fisher) was added for 10 min to visualize actin filaments. Inverted confocal microscope (Leica SP5) was used for sample visualization. Image quantification and assembly were performed using ImageJ and Adobe Photoshop CS3.

Western blotting

Oocytes, embryos or tissues (ovaries and muscle) were lysed with 6 μl of Millipore H₂O and 2.5 μl of 4× lithium dodecyl sulfate, sample buffer NP 0007, and 1 μl reduction buffer NP 0004 (Novex, Thermo Fisher Scientific) at 100 °C for 5 min. Lysates were separated using a 4–12% gradient SDS-PAGE (NP323BOX, Life Technologies) and transferred to an Immobilon-P membrane (PVDF; Millipore) using semidry blotting system (Biometra GmbH). Membranes were blocked for 1 h, in 1–5% skimmed milk dissolved in Tween-Tris-buffer saline (TTBS, pH 7.4) according to the antibody (list of primary antibodies and dilutions is below). After 3× 10 min of washing in TTBS, membranes were incubated at 4 °C overnight in 1–5% skimmed milk/TTBS with primary antibodies listed in Supplementary Table 1. After 3× 10 min washing in TTBS, the membranes were incubated for 1 h with secondary antibody Peroxidase Anti-Rabbit Donkey (711-035-152, Jackson Immunoresearch) 1:7500 in 1% milk/TTBS 1 h at room temperature. Immunodetected proteins were visualizd by ECL (Amersham, GE Healthcare life science), films were scanned using a GS-800 calibrated densitometer (Bio-Rad) and quantified using Image J software.

Figure 2

Effect of FSH on [³⁵S]-methionine incorporation to mouse oocytes and 2cell embryos. (A) A representative autoradiography image of [³⁵S]-methionine incorporation into de novo synthetized proteins during 2 h labeling of denuded GV and MII oocytes and 2cell embryos exposed to FSH (100 ng/ml). GAPDH protein levels were used as a loading control. (B) Incorporation of [³⁵S]-methionine to oocytes cultured with absence (DO) or presence (COC) of CCs treatedwith absence or presence of FSH. GAPDH was used as a loading control. (C) Incorporation of [³⁵S]-methionine in FSH-treated NIH3T3 cells. GAPDH expression was used as a loading control. (D) Quantification of GAPDH and autoradiography of FSH-treated/non-treated oocytes and embryos shown in A. Non-treated controls were set as 100% (means ± SEM). Data from three independent experiments are presented with equal number of cells per experimental group. Statistical differences were tested using Student’s t-test; NS—not significant. (E) Quantification of GAPDH and autoradiography shown in B. The value of GAPDH expression and [³⁵S]-methionine incorporation after FSH-treatment is compared to the values of non-treated COCs and DOs (means ± SEM, n ≥ 3). ³⁵S-Met labeled COCs were stripped of CCs after culture. Student’s t-test; NS—not significant. (F) Quantification of results shown in C. The value of GAPDH expression and [³⁵S]-methionine incorporation after FSH-treatment is compared to the values of non-treated NIH3T3 cells (means ± SEM); Student’s t-test; NS—not significant. (G) Representative confocal images of methionine analog HPG incorporation into DOs treated with FSH (100 ng/ml) compared to non-treated ones. (H) Quantification of results shown in Figure 2G. The value of HPG incorporation was compared to the values of non-treated DOs (means ± SEM); data from three independent experiments. Student’s t-test; NS—not significant. See also Figure 1 and Supplementary Figure 1.

Figure 3

Expression of FSHR in mouse oocytes and 2cell embryos. (A) Quantification of Fshr mRNA expression of by real time PCR. The values of Fshr expression in MII DOs and 2cell embryos were compared to the values of GV-stage DOs. ANOVA, data from three independent experiments, n = 30 oocytes per group, NS—not significant. See also Supplementary Figure 2. (B) Representative WB image of FSHR protein expression in GV and MII DOs and 2cell embryos. See also Supplementary Figure 3. (C) Quantification of WB results shown in B. Follicle-stimulating hormone receptor and GAPDH expression levels are relative to the values in GV DOs (means ± SEM); ANOVA; NS—not significant. Data from at least three independent experiments. (D) Representative WB image of FSHR and GAPDH protein expression in FSH-treated and non-treated NIH3T3 cells. 2cell embryos were used as positive control. (E) Quantification of WB results shown in D. Data from three independent experiments; means ± SEM; Student’s t-test.

Figure 4

Effect of FSH on translational regulators. (A) A representative WB images of protein expression of selected markers involved in translation regulation in GV and MII oocytes and 2cell embryos treated and non-treated with FSH. (B) Quantification of WB results shown in A. The values are normalized to the non-treated controls; (means ± SEM); data from at least three independent experiments; Student’s t-test; NS—not significant. (C) Representative confocal images of amino acid uptake (HPG) in GV oocytes with ceased translation. CHX was used for suppression of translational and methionine analog HPG staining serves as amino acid uptake marker. (D) Quantification of confocal images shown in C. Data from at least three independent experiments; n ≥ 10; Student’s t-test.

Figure 5

Expression of FSHR in bovine, porcine, and human DOs and effect of FSH treatment on the rate of methionine incorporation into bovine, porcine, and human DOs. (A) A representative image of FSHR protein detection by WB in mouse, bovine, porcine, and human oocytes. Tubulin and GAPDH were used as a loading control. (B) Quantification of WB results shown in A relative to the levels of FSHR and Tubulin in mouse, bovine, porcine, and human oocytes (means ± SEM, data from three independent experiments; ANOVA). (C) Incorporation of [³⁵S]-methionine into proteins during 2 h labeling of bovine, porcine, and human DOs. GAPDH was used as a loading control. (D) Quantification of autoradiography and GAPDH WB data shown in C relative to the levels in control bovine, porcine, and human DOs (means ± SEM, data from at least two independent experiments; Student’s t-test).

Silver staining

The membranes were incubated in staining solution (1 g of sodium citrate, 0.4 g FeSO4, 0.1 g AgNO3, 50 ml Milli-Q water) for 10 min. The staining reaction was terminated by rinsing the membranes in Milli-Q water (5 times for 2 min) and membranes were subsequently dried.

Reverse transcriptase PCR and quantitative RT-PCR

The total RNA was from 20 mouse oocytes, 20 2cell embryos, 20 cumulus layers from 20 COCs, as well as pieces of the ovary and thigh muscle which were isolated using a RNeasy Plus micro kit (Qiagen) according to the manufacturer’s instructions. Isolated RNA was stored at −80°C. Complementary DNA was synthetized by qPCRBIO cDNA synthesis kit (PCR Biosystems) using oligo (dT) (Thermo Scientific) and random hexamer primers (Thermo Scientific). The reaction was performed for 30 min at 42  °C (PTC200, Bio-Rad). For PCR, PPP Mix kit (Top-Bio) was used according to the manufacturer’s instructions: 94  °C for 1 min followed by 40 cycles of 94 °C for 15  s, 58 °C for 15 s, 72 °C for 20 s, and 72 °C for 7 min. Products were verified by 1.5% agarose gel electrophoresis with ethidium bromide staining. Quantitative PCR was performed by CFX96 Realtime system (Biorad) using appropriate primers (primer names and sequences are listed in Supplementary Table 2) by TaqMan Gene Expression Master Mix XS (Applied Biosystems) according to manufacturer’s instructions: 50  °C for 2  min and heated at 95  °C for 10 min followed by 40 cycles of 95 °C for15 s, 50 °C for 20  s, and 58  °C for 60  s. The data are from at least three biological replicates. Products were verified by melting analysis. The relative concentrations of templates in different samples were determined using method 2^(-ddCT). The results were normalized according to the relative internal standard glyceraldehyde 3-phosphate dehydrogenase (Gapdh).

Statistics

All experiments were repeated at least three times. Western blot (WB) and radiography images were analyzed using ImageJ software (http://rsbweb.nih.gov/ij/), ICC and immunohistochemical (IHC) images were processed with LAS X (Leica) and ImageJ. Mean and standard deviation values were calculated using MS Excel, the statistical significance of the differences between groups were tested using Student’s t-test or ANOVA and p < 0.05 was considered as statistically significant. P values were distinguished: p < 0.05; p < 0.01; and p < 0.001. Statistical results that were not significant are designated with “NS”.

Results

FSH suppresses de novo methionine incorporation into oocytes and 2cell embryos

To elucidate the influence of FSH on methionine incorporation, we analyzed two different maturation stages of oocytes (GV and MII), 2cell embryos, and CCs; two different groups of oocytes were tested: either cultivated in the presence (COCs) or absence of CCs (DOs) (Figure 1). In FSH-treated cells, the incorporation of [³⁵S]-methionine protein synthesis marker was significantly (p < 0.01) reduced to 17% (± 3%) in cumulus-free GV-stage DOs, 15% (± 5%) in MII oocytes, and 19% (± 2%) in 2cell embryos, respectively (Figure 2A). The expression of the loading control GAPDH was the same in oocytes and 2cell embryos regardless of FSH treatment (Figure 2A). The effect of FSH on [³⁵S]-methionine incorporation into oocytes was dose dependent. The decrease of [³⁵S]-methionine incorporation was evident in oocytes treated with 10 ng/ml FSH (Supplementary Figure 1A and B). To avoid the possibility that the observed effect on [³⁵S]-methionine incorporation was specific to a particular commercial FSH, we compared the effect of Gonal with a different recombinant FSH, Puregon. Treatment of DOs with either FSH-Gonal or FSH-Puregon resulted in both cases in decreased [³⁵S]-methionine incorporation (Supplementary Figure 1C and D).

Next, we examined the difference in [³⁵S]-methionine incorporation into GV-stage DOs and COCs. COCs treated with FSH (100 ng/ml) for 2 h and stripped of cumulus (Figure 1D) after the treatment showed a similar decrease in [³⁵S]-methionine incorporation (22% ± 3%; p < 0.05) as was the case in DOs, without any significant difference (p > 0.05) (Figure 2B and E). In CCs originating from FSH-treated COCs (Figure 1D), a 48% reduction in [³⁵S]-methionine incorporation was induced (Supplementary Figure 1E and F). Importantly, treatment of fibroblasts (NIH3T3) with FSH (100 ng/ml) did not show any significant (p > 0.05) change in methionine incorporation (Figure 2C). Using an additional protein synthesis marker, the methionine analog HPG, showed significant (p < 0.001) reduction (54% ± 20%) of HPG incorporation into FSH-treated (100 ng/ml) GV-stage DOs occurred (Figure 2G and H).

We found that FSH had a biological effect on de novo protein synthesis measured as methionine incorporation into the oocytes and 2cell embryos. The FSH-treated GV and MII oocytes and 2cell embryos exhibited decreased [³⁵S]-methionine incorporation into proteins suggesting a direct effect of FSH on the amino acid metabolism in DOs and embryos.

FSH receptor expressed in mouse oocytes and 2cell embryos

As our results revealed the inhibitory effect of FSH on [³⁵S]-methionine incorporation into oocytes and 2cell embryos (Figure 2A), we further investigated the putative expression of FSHR in oocytes and embryos. Applying qRT-PCR, follicle stimulating hormone receptor (Fshr) mRNA was detected in GV and MII oocytes and in 2cell embryos (Figure 3A). We found that Fshr mRNA was present in oocytes during meiotic maturation and was significantly decreased in 2cell embryos (Figure 3A, Supplementary Figure 2A). To exclude possible contamination by mRNAs from transzonal projections of CCs, we analyzed the presence of Fshr transcripts in zona pellucida enclosed (ZP+) and zona pellucida free (ZP−) oocytes. ZP− oocytes exhibited a similar amount of Fshr mRNA as ZP+ samples (Supplementary Figure 2B). Sequence analysis of RT-PCR product confirmed that PCR product is Fshr specific.

Western blot analysis revealed that the FSHR protein was expressed in GV and MII mouse oocytes and in 2cell embryos (Figure 3B) with non-significant differences between the groups (Figure 3C). As expected, FSHR was expressed in the mouse ovary and it was not present in the negative control, muscle tissue (Figure 3B). We also analyzed the expression of FSHR in the separated CCs, where the presence of FSHR protein had previously been confirmed [29, 30]. Similarly as in the oocyte sample, we detected the high expression of FSHR in the CCs (Supplementary Figure 3). The presence of FSHR was not detected in mouse fibroblasts (NIH3T3 cells) (Figure 3D and E) and muscle tissue (Figure 3B and C; Supplementary Figure 3A and B). Additionally, ICC revealed the presence of FSHR protein in the oocytes, embryos, CCs, and ovary, but not in the muscle (Supplementary Figure 4). The signal was distributed evenly in the oocytes (Supplementary Figure 4A), however, without Triton X-100 permeabilization of cytoplasmic membrane FSHR showed abundant membrane localization (Supplementary Figure 4B).

Our results reveal that the FSHR protein is expressed in GV and MII oocytes and persists until at least the 2cell embryo stage.

FSH treatment does not affect the translational pathway in oocytes and 2cell embryos, despite the negative FSH effect on methionine incorporation

Our results demonstrate that FSH negatively affects the incorporation of the global protein translation marker [³⁵S]-methionine into oocytes and 2cell embryos, in which FSHR is expressed. We assumed that the key translational regulators are possibly affected by FSH. Active mTOR (Ser2448) and ERK1/2(Thr202/Thr204) kinases, translational repressor 4E-BP1, and elongation factor eEF2(Thr56) were analyzed, as well as initiation factor eIF2α and translational stress repressor marker. However, the expression and phosphorylation of tested key players of translational regulation were not changed in FSH-treated oocytes and embryos (Supplementary Figure 4A and B) and, moreover, phosphorylation of eIF2α (Ser51) was not affected (Supplementary Figure 4A and B). Furthermore, silver staining of the WB membrane did not reveal any FSH effect on the alteration of the global protein quantity (Supplementary Figure 5A and B).

These results suggest that FSH has no influence on the activity of translational activators (mTOR and ERK), repressors (4E-BP1, eIF2a), or elongation factor (eEF2), and except for the suppression of methionine incorporation FSH does not impose any stress on global protein translation in oocytes and 2cell embryos.

Similar to methionine, in situ translational marker HPG is negatively influenced by FSH (Figure 2G and H). To detect amino acid transport to the oocyte, we used HPG in combination with translational repressor CHX. In the presence of FSH detection of intracytoplasmic HPG, the fluorescence signal was significantly reduced to 48% (±3%; p < 0.001) (Figure 4C and D).

Our data show that despite the translational machinery not being influenced by the presence of FSH, the influx of exogenous amino acid is significantly reduced.

FSHR protein is expressed in oocytes from various mammalian species and treatment of FSH negatively influences methionine incorporation

We further investigated if FSHR was expressed in other mammalian species. Expression of similar levels of FSHR protein was observed in bovine, porcine, and human DOs (Figure 5A and B). Exposure of bovine, porcine, and human DOs to FSH exhibited a similar suppression of [³⁵S]-methionine incorporation as in mouse oocytes (Figure 5C and D).

Our data clearly show that FSHR is expressed in the oocytes of at least three mammalian species. Treatment with FSH significantly suppressed methionine incorporation into newly synthesized proteins in mouse, bovine, porcine, and human oocytes.

Exogenous gonadotropins influence MII-spindle morphology

We studied whether the oocyte morphology was also influenced by FSH treatment. Although the oocytes treated with FSH accomplished first meiotic division with polar body extrusion, the morphology of MII spindle was altered. We measured the MII spindle morphology at two axes (Figure 6A) and we found that the spindle length and width was significantly larger in in vitro FSH-treated oocytes (Figure 6B). Moreover, MII oocytes obtained from naturally stimulated (Whitten effect; [31, 32]; see methods) and PMSG-primed mouse females showed similarly altered spindle morphology (Figure 6C).

Figure 6

The effect of exogenous FSH on MII-spindle morphology. (A) Depiction of spindle axis measurements in control and FSH oocytes. Tubulin depicted in green, chromosomes in blue. Dashed lines represents measured spindle axes. Scale bars 20 μm. (B) Quantification of MII spindle axes in in vitro oocytes cultured with presence or absence of FSH. Data from threeindependent experiments, means ± SEM; n ≥ 15; Student’s t-test. (C) Quantification of MII spindle axes in in vivo MII oocytes form naturally and PMSG-primed females. Data from three independent experiments; means ± SEM; n ≥ 15; Student’s t-test.

Hence our data suggest that exogenous gonadotropins affect the spindle morphology of MII oocytes progressing through meiosis in both in vitro and in vivo conditions.

Discussion

In fully grown mammalian oocytes, gene expression is regulated mainly at the level of protein synthesis, since the transcription is ceased during meiotic maturation [33]. Gene expression in oocytes is regulated almost exclusively at the level of mRNA translation and posttranslational modifications of proteins. It has been documented earlier that increased concentration of gonadotropins in culture medium results in an increase of the percentage of oocytes reaching MII, a normal configuration of the spindle and correct chromosomal alignment, cortical granule migration, and mitochondrial aggregation. Accumulation of oocyte proteins associated with improved oocyte quality is increased when COCs are incubated with FSH [20]. However, in our experiments, exposure to FSH results in a considerable decrease of [³⁵S]-methionine incorporation into the newly synthesized proteins in mouse COCs, DOs, 2cell embryos, as well as in CCs. Moreover, we detected the suppression of [³⁵S]-methionine incorporation in bovine, porcine, and human DOs. Furthermore, this study and Wetzels et al. [22] show a significant decrease of [³⁵S]-methionine uptake in mouse oocytes, 2cell embryos, and blastocysts after in vitro or in vivo gonadotropins administration. It has also been shown in starfish oocytes that when exposed to the maturation-inducing hormone 1-methyladenine [34], amino acid uptake is reduced. The authors conclude that the nearly immediate decrease in permeability for amino acids indicates that the site of action of 1-methyladenine is on the surface of oocytes.

The effect of FSH on oocytes and early embryos has been reported in the number of studies, however, the results of these studies are rather different. An in vitro stimulatory effect of FSH on protein synthesis in porcine CCs and bovine granulosa cells, as well as on the number of pig oocytes MII stage has been documented [35, 36]. On the other hand, a negative effect of FSH and PMSG treatment on oocyte and embryo development has been also shown. When bovine oocytes are subjected to IVM in the presence of purified pituitary FSH (pFSH), lower cleavage and decreased blastocyst rates were observed after fertilization, and when pFSH was replaced by recombinant FSH, the reduction of embryo development was more pronounced [25]. In bovine oocytes treated with high gonadotropin concentrations of genes implicated in spindle formation, cell cycle control and methylation was downregulated [21]. In vitro embryo development and in vivo blastocyst formation in super-ovulated mice was delayed [23] and a lower cell number with decreased mitosis index in in vitro gonadotropin-stimulated mouse embryos was observed [24]. In FSH-treated sheep oocytes, the passage of labeled choline was suppressed [37] and the decrease of uridine and choline uptake was also reported in porcine oocytes exposed to FSH [38]. The results of our experiments show suppression of both [³⁵S]-methionine and methionine analog HPG incorporation into oocytes and early embryos treated with recombinant FSH (Gonal-F). In Gonal-F-treated NIH3T3 cells, no changes in [³⁵S]-methionine incorporation rates have been observed. It has been previously reported that the FSHR is expressed in human endometrial cells [39] and that the proliferation of these cells treated with recombinant FSH (Gonal-F and Puregon) has been significantly inhibited [40]. These published data and our presented results suggest that in cells expressing FSHR, recombinant FSH negatively affects the proliferation (endometrial cells) and amino acid incorporation in oocytes and 2cell embryos.

FSH influences granulosa and Sertoli cells, in which a receptor for FSH (FSHR) is expressed naturally. In vitro FSH treatment caused a decrease of nuclear proteins synthesis in porcine Sertoli cells [41] and elicited morphological changes in rat Sertoli cells [42]. In FSH-treated human or rat granulosa cells, a reduction in the synthesis of the adherent junction proteins and a considerable suppression of the [³⁵S]-methionine incorporation to vinculin, α-actinin, and actin was detected, suggesting that FSH did not affect protein turnover, but rather induced changes in protein synthesis [43–45].

We and others [46–52] clearly show the presence of Fshr mRNA in oocytes and early embryos of various mammalian species including human. We have confirmed FSHR expression on the protein level in mouse, bovine, porcine, and human GV oocytes, in mouse MII oocytes and 2cell embryos employing WB and ICC. While our WB data reveal a single FSHR band of 75–77 kDa in mouse, bovine, and porcine DOs and 2cell mouse embryos, in control mouse ovarian extracts, additional FSHR bands are also apparent. It is possible to deduce that the additional FSHR bands are specific to FSHR expression in adult rodent ovary, similar to those reported earlier in the hamster [53]. The molecular weight of the FSHR protein and the specificity of our findings are confirmed by the previously published studies on human ovary and rat granulosa cells [54, 55]. Moreover, Fshr mRNA is present at the actively translating polyribosomal fractions [56] of mouse oocyte and zygote. Although no other WB data of FSHR detection in oocytes has been published, our findings of FSHR expression in oocytes and 2cell embryos are well supported by the results of IHC FSHR localization.

In sections of ovaries from mice, mare, porcine, and human, FSHR-staining has been observed [48, 57, 58]. Our data of ICC analysis reveal a diffuse distribution of FSHR and this finding corresponds to different non-membrane bound FSHR variants and the receptor precursors that accumulate in the cells [54, 59]. Additionally, we show uniform membrane localization of FSHR when membrane lipid structures are preserved which suggest that FSHR might be present at the cell membrane with detergent soluble structures. Although there are no other published data on FSHR detection by ICC in oocytes, a distinct localization of the [¹²⁵I]-labeled FSH suggests the presence of translated FSHR in human and porcine DOs [48]. We and others show that gonadotropins can affect the size of MII spindles. Pregnant mare serum gonadotropin used in mammalian superovulation treatment protocols exhibits FSH and luteinizing hormone activity [60]. The MII spindles in in vitro cultured oocytes from PMSG-primed mice are significantly larger than those from un-primed mice [61, 62]. Consistent with the published data, our results reveal different MII-spindle proportions in oocytes originating either from PMSG-stimulated or naturally stimulated mice. Follicle-stimulating hormone treatment can also affect meiotic spindle organization both in in vitro and in vivo conditions, as, for example, in matured mouse oocytes [63, 64] where spindle misalignments occur in oocytes stimulated in vivo with FSH [65]. Moreover, mouse oocytes treated with high doses of FSH (2 μg/ml) showed chromosome displacement of the meiotic spindle that has been observed [66, 67]. Moreover, it has also been shown that expression of genes that function in spindle formation, cell cycle control, and methylation was downregulated in bovine oocytes treated with FSH [21].

Here we show the occurrence of FSHR transcript and translated protein in oocytes of various mammalian species and we propose that FSH exerts its effect on oocytes and embryos through this receptor similar as in granulosa and CCs [4, 68]. Moreover, we have not detected any effect of FSH on [³⁵S]-methionine incorporation in mouse embryonic fibroblasts, which do not contain a FSHR. As the incorporation of [³⁵S]-methionine is a marker of translational levels, we have examined the state of key factors controlling protein translation in oocytes and embryos treated with FSH. Proteins with a known role in translational regulation were tested: mTOR [13, 69], 4E-BP1 [13, 14, 70, 71], eIF2α [72], eEF2 [73], and ERK1/2 [74, 75]. Interestingly, in FSH-treated oocytes and 2cell embryos, we did not detect any changes in the phosphorylation of the mentioned components of the protein translation pathway. This suggests that the decrease of methionine incorporation into newly translated proteins in FSH-treated cells was not caused by decreased activity of these proteins. Amino acid transport through the cytoplasmic membrane of the oocyte is affected, as we also observed lower incorporation of methionine analog HPG in FSH-treated oocytes in conditions when the translation was blocked by cycloheximide. Here, we showed that total protein level is not affected and assumed that oocyte/embryo is able to recycle internal proteins through proteasome system [76] and used its internal amino acids to maintain its physiological requirements. The mild phenotypic effect on the oocyte maturation might be due to sufficient usage of internal amino acids in this cell type which may explain why we do not detect any influence on amino acid-sensing pathway which we studied through analysis of mTOR and stress effector eIF2α.

Our data confirm that recombinant FSH of different sources (Gonal-F and Puregon) has a negative effect on methionine uptake in mouse oocytes, supported by results on human endometrial cells expressing FSHR, where Gonal-F and Puregon significantly inhibited cell proliferation [39, 40]. However, the possible different effect of other types of FSH cannot be excluded, since the accumulation of oocyte proteins associated with improved oocyte quality has been reported in mouse COCs incubated with ovine FSH [20].

In conclusion, we show that FSH treatment negatively affects methionine incorporation into mammalian oocytes and 2cell embryos, possibly through FSHR expressed in cumulus-free mouse, bovine, porcine, and human oocytes on mRNA and protein levels. As FSH is frequently used for the in vitro culture of oocytes and, moreover, PMSG treatment is applied for superovulation induction in females, these results may provide vital new insights into the physiology of female germ cells and should be taken into consideration by specialists in the field of assisted reproduction and animal breeding.

Acknowledgement

We thank Jaroslava Supolikova and Marketa Hancova for excellent technical assistance with experiments.

Conflict of Interest

The authors have declared that no conflict of interest exists.

Author notes

^†

Grant Support: This research was funded by MSMT (EXCELLENCECZ.02.1.01/0.0/0.0/15_003/0000460 OP RDE), GACR (18-19395S; 19-13491S) and Institutional Research Concept RVO67985904.

Conference Presentation: Presented in part at the Visegrad Group Society for Developmental Biology Inaugural meeting, 2018, Brno, Czech Republic.