In Vitro Maturation of Fully Grown Mouse Antral Follicles in the Presence of 1 nM 2-Hydroxyestradiol Improves Oocytes’ Developmental Competence

Cathecolestrogens are estradiol metabolites produced during folliculogenesis in the mammalian ovary. 2-Hydroxyestradiol (2-OHE2) is one of the most abundant although its role remains unknown. The aim of this study is to investigate whether the presence of 2-OHE2 during the germinal vesicle-to-metaphase II transition affects oocyte meiotic and preimplantation developmental competence. Mouse cumulus-oocyte complexes (COCs), isolated from fully grown antral follicles, were in vitro–matured (IVM) in the presence of 2-OHE2 (0.1, 1, 10 or 100 nM) for 6 or 15 h; then, their meiotic and developmental competence was evaluated using a number of cytological quality markers. With the exception of the highest dose (100 nM), the addition of 2-OHE2 to the IVM medium, did not alter, compared with untreated control, the frequency of oocytes that reached the MII stage. Instead, IVM in the presence of 1 nM 2-OHE2 highly increased the rate of preimplantation development and blastocyst quality. To understand whether this positive effect could be attributed to the events occurring during meiosis resumption, we analysed a number of specific cytological quality markers of the asymmetric division, such as PB-I volume and position, presence and extension of the cortical F-actin cap, meiotic spindle shape and area, and microtubule organisation centre localisation. The results highlighted how the presence of 1 nM 2-OHE2 significantly improved the overall cytological organisation required for a correct asymmetric division. Our results contribute a first step to acknowledge a potential role of this estradiol metabolite during the GV-to-MII transition, contributing to the acquisition of oocytes developmental competence. Electronic supplementary material The online version of this article (10.1007/s43032-020-00276-6) contains supplementary material, which is available to authorized users.


Introduction
The acquisition of the oocyte developmental competence occurs inside the ovarian follicle thanks to a continuous crosstalk between the germ cell and the surrounding cumulus and granulosa cells. The follicle somatic cell components contribute crucial factors required, through several signalling pathways and regulatory loops, to complete oocyte growth and meiotic maturation [1,2].
The importance to preserve the communication between oocyte and surrounding granulosa cells becomes evident when cumulus-oocyte complexes (COCs) are isolated from fully grown antral follicles and in vitro cultured [3,4]. In the mouse, the efficiency of COCs in vitro maturation (IVM) varies depending on both strain (i.e., outbred or inbred) [5] and culture conditions [6,7], although the meiotic and developmental competence of metaphase II (MII) oocytes obtained in vitro remains lower compared with that of ovulated MII oocytes [8,9].
Catecholestrogens have been shown to play a role in folliculogenesis both in Fish and Mammals. In teleosts (i.e., catfish and zebrafish), 2-OHE 2 has a crucial role as component of the gonadotrophin cascade of regulation of oocyte maturation and ovulation [22,23]. In vitro maturation of intact catfish follicles in the presence of 5 μM 2-OHE 2 increased significantly the frequency of oocytes attaining germinal vesicle breakdown (GVBD); instead, it did not change it when oocytes were cultured in the absence of surrounding follicle cells [24].
In the mammalian ovary, oestrogen hydroxylases are wellrepresented showing a 100-fold increase from small antral to fully grown follicles and corpora lutea [25]. The produced 2-OHE 2 and 4-OHE 2 metabolites are detected in the follicular fluid of antral follicles; they increase concentration during follicle maturation [26] and have an influence on the differentiation and proliferation of granulosa [27] and luteal [28] cells. Also, they have been shown to stimulate progesterone production in small follicles [29], to enhance cAMP production and, as a consequence, improve FSH activity [27]. When injected, 4-OHE2 induced ovulation in immature rats [30] and improved the overall implantation success rate [31].
Overall, the studies described above highlight several roles for catecholestrogens during follicle maturation, and suggest a link with the female gamete meiotic and developmental competence. The objective of this study was to investigate how the presence of 2-OHE 2 during the germinal vesicle-to-metaphase II transition affects oocytes meiotic and preimplantation developmental competence.
The main aim of the present study was to explore the effects of one of 2-OHE 2 at concentrations ranging from 0.1 to 100 nM, during IVM of mouse fully grown antral follicles from the GV to the MII stage. Oocytes that reached MII were inseminated with capacitated sperm and their preimplantation developmental competence recorded. Our results indicate a dose-dependent effect of 2-OHE 2 , that, when present in the IVM culture medium at 1 nM concentration, induces a significant beneficial effect on both meiotic resumption and preimplantation embryonic development.

Animals and Reagents
Four 6-week-old female and 6-month-old male CD1 mice were purchased from Charles River (Como, Italy). Animals were maintained under controlled room conditions (22°C, with 60% air moisture and 12/12 light/dark photoperiod), and investigations were conducted in accordance with the guiding principles of European (n. 2010/63/UE) and Italian (n. 26/2014) laws protecting animals used for scientific research. All chemicals used were purchased from Sigma-Aldrich (St. Louis, MO, USA), unless otherwise stated. Ultrapure water used for preparing media and solutions was obtained with a MilliQ® IQ 7000 system (Merck) equipped with a Biopak® Ultrafiltration filter (Merck).

COC Isolation and In Vitro Maturation
Females were injected with 5 I.U. Folligon (Intervet Srl, Italy), and 48 h later, COCs were isolated from the ovary. Briefly, ovaries were placed in 1 ml HEPES-buffered isolation medium: α-MEM plus GlutaMAX (Life Technology; cat.no. 32561-029) supplemented with 6 mg/ml HEPES, 5% fetal bovine serum (FBS), 0.23 mM sodium pyruvate, 1 mg/ml Fetuin, 100 I.U./ml penicillin and 75 μg/ml streptomycin in a 30-mm Petri dish (Corning, Euroclone, Italy) under a stereomicroscope. Fully grown antral follicles were punctured from the ovarian surface with a thin and sharp-pulled sterile Pasteur pipette. Only intact COCs with layers of cumulus cells completely surrounding the oocyte were used in further experiments. After washing in fresh HEPES-buffered isolation medium, COCs were cultured in 100 μl drops of α-MEM supplemented with 5% FBS; 0.23 mM sodium pyruvate; 1 mg/ml fetuin; 100 I.U./ml penicillin and 75 μg/ml streptomycin; 50 mI.U./ml FSH; and 10 ng/ml epidermal growth factor (EGF) in the absence (IVM) or presence (IVM-2-OHE 2 ) of 0.1 nM, 1 nM, 10 nM or 100 nM 2-OHE 2 (prepared in dimethyl-sulfoxide, DMSO, at a maximum concentration of 0.002%. As reported earlier [32] and confirmed by our results (Tables 1S and 2S), the presence of DMSO in IVM medium, at concentrations up to 0.1%, did not have adverse effects on oocyte maturation and on further preimplantation development. Thus, DMSO was always present in our IVM medium, independently on the presence or absence of 2-OHE 2 .
IVM drops were layered with mineral oil and placed in an incubator at 37°C, 100% humidity in 5% CO 2 , 5% O 2 and 90% N 2 for 6 h (MI stage) or 15 h (MII stage). At the end of the culture period, MI or MII (confirmed by the presence of PB-I) oocytes were used for further experiments.
In order to collect in vivo-matured oocytes, females were injected with 5 I.U. Folligon followed, 48 h later, by 5 I.U. Corulon (Intervet) and, after 6 h (MI) or 15 h (MII), gametes were isolated from ovarian surface or from the oviducts, respectively.
MII oocytes, with the exception of those used for in vitro fertilisation (IVF), were denuded of their cumulus cells by a brief treatment with 500 I.U. hyaluronidase.
Double labelling of αand γ-tubulin, raised in the same species, was done through sequential detection of both epitopes, i.e. primary and then secondary antibodies for the first epitope followed by primary and then secondary antibodies for the second epitope.
Samples were examined using a Leica TCS SP8 confocal microscope equipped with lasers for AlexaFluor488, AlexaFluor555, AlexaFluor633 and DAPI fluorochromes. Stacks were obtained with axial distances of 0.3 μm. Immunolocalisation of OCT4 and CDX2 on blastocysts was analysed with an Olympus BX60 fluorescence microscope (Olympus, Ibaraki, Japan) equipped with single-bandpass filters for DAPI, AlexaFluor488 and AlexaFluor555; images were captured with a DP72 camera (Olympus) controlled by the CellSens Dimension 1.4.1 software (Olympus) and processed with ImageJ (http://imagej.nih.gov/ij/).

First Polar Body Morphology and Size
CCs-free MII oocytes were photographed using an inverted Olympus IX71 microscope equipped with a JVC KY-F58 3-CCD camera and a micromanipulator. For each oocyte, images (under × 2x objective) were captured with a frontal and lateral PB-I orientation obtained moving the oocytes using the micromanipulator micropipette. Only intact PB-Is were considered for further analyses. Pictures were analysed using the CellSens Dimension 1.4.1 software and the PB-I volumes calculated drawing three lines corresponding to the three radii of each PB-I.

Amplitude Angle Between PB-I and Oocyte Chromosomes
CCs-free MII oocytes were stained with 0.05 μg/ml bisBenzimide H-33342 trihydrochloride (Hoechst 33342; cat. N. D2261) in M2 medium for 15 min. Then, by using an inverted Olympus IX71 fluorescence microscope equipped with a micromanipulation system, single oocytes were rotated until metaphase chromosomes of both PB-I and oocyte appeared on the same focus. Pictures were taken, under × 20 objective, with a JVC KY-F58 3-CCD camera and analysed using the CellSens Dimension 1.4.1 software. Two lines were drawn from the centroid of the oocyte towards either the oocyte's or PB-I's chromosomes and the resulting angle was measured.

F-Actin Cap Intensity Profile and Amplitude Angle
The F-actin cap intensity was calculated, on the phalloidin red channel image, using the ImageJ software. A line was drawn (see Fig. 3) that, from the cortical region (p), intersected the MI or MII chromosomes, passed through the centroid (c) and reached the opposite pole (d). The fluorescence intensity was measured as p/d ratio. MI or MII oocytes of the three experimental conditions were divided into two groups: group I, oocytes with intensity (I) comprised between 1.8 < I ≤ 21.8; and group II, 1.1 < I ≤ 1.8.
The amplitude angle of the F-actin cap was determined drawing two lines starting from the centroid of the oocyte and the ends of the F-actin cap defined by phalloidin staining (see Fig. 4a).

Spindle Area and Shape
Using the CellSens Dimension software, the spindle area, identified by the α-tubulin signal, was calculated considering the spindle as a barrel made of two trapezoids with a common side at the equator width, the shorter sides at the two spindle poles and the height as half the length between the two poles (Fig. 5).
Based on the ratio between the equator width and the mean of the widths at the two poles, a barrel and a rectangular configuration were observed.

MTOC Distribution
The microtubule organisation centre (MTOC) distribution was established based on the γ-tubulin signal. Two different patterns were identified: (i) a single focus, in which all the γtubulin signals were distributed as a single cluster, localised at each spindle pole; and (ii) multiple foci, in which the γtubulin signals were distributed as two or more clusters along the width of the two poles.

Statistical Analysis
Statistical analyses were carried out using the Sigma Stat 3.5 software. Data, obtained from at least five independent experiments, were analysed by the Student's t test for comparing two conditions or by the one-way ANOVA for comparing more than two groups. In the presence of significant differences, one-way ANOVA test was followed by the Fisher LSD Method (post hoc test). When data were not normally distributed, statistical analysis was performed either by the Mann-Whitney test (when the comparison involved two samples) or by the Kruskal-Wallis test together with the Dunn's multiple range test (for the comparison of more than two samples). Parametric and nonparametric data were expressed as the error standard ± mean (SEM). The analysis of the absolute frequencies was performed using a Fisher's exact test. Differences were considered significant for p values ≤ 0.05.

Results
To test the effects of 2-OHE 2 on oocyte developmental competence, COCs were in vitro-matured for 15 h in the presence of 0.1, 1, 10 or 100 nM 2-OHE 2 .

Oocytes In Vitro Maturation
Based on the presence of the PB-I, more than 90% of oocytes in all five experimental conditions reached the MII phase. Compared with the control in vitro-matured COCs (IVM) ( Table 1), oocytes whose COCs were cultured in the presence of 2-OHE 2 (IVM-2-OHE 2 ), spontaneously resumed meiosis and reached MII without significant differences (p > 0.05), with the exception of those treated with the highest 2-OHE 2 concentration (100 nM) in which a decreased MII rate was observed (90.4 ± 1.5%; p = 0.015). This latest experimental group also showed significantly higher frequency of fragmented oocytes (3.1 ± 1.4%; p = 0.002).
Overall, these results indicate a highly significant improvement of the preimplantation developmental rate to blastocyst of those antral oocytes whose COCs were matured in the presence of 1 or 10 nM 2-OHE 2 . Comparison between ovulated MII oocytes (OV MII ) and IVM or IVM-2-OHE 2 showed that the former performed always significantly better (Table 1).

Blastocyst Cell Number
The quality of blastocysts obtained with the 1 or 10 nM 2-OHE 2 experimental conditions was evaluated based on their total number of cells and on the number of trophectoderm (TE) or inner cell mass (ICM) blastomeres, compared with control IVM blastocysts or to those derived from OV MII . Embryos were processed for the immunofluorescence localisation of OCT4 and CDX2 proteins, markers of ICM and TE cells, respectively. As shown earlier [9], the total cell number of blastocysts derived from IVM-COCs (32.3 ± 1.6) was significantly lower than that of control OV MII oocytes (52.5 ± 1.0; p < 0.001) ( Table 2). Compared with IVM blastocysts, the presence of 1 nM 2-OHE 2 during COC maturation significantly (p = 0.011) increased the number of cells (43.2 ± 4.6), instead that of 10 nM IVM-2-OHE 2 (34.4 ± 1.7) was not significantly different (p = 0.561). This higher number of cells could be attributed to an increased number of ICM blastomeres (9.9 ± 0.7 vs. 7.4 ± 0.6; p = 0.003).
In summary, the results described highlight a significant improvement of the preimplantation developmental rate when COCs were matured in the presence of 1 nM or 10 nM 2-OHE 2 , and better blastocyst quality with 1 nM concentration.
Next, we addressed the question of whether this observed improvement in the presence of 1 nM 2-OHE 2 could be attributed to the latest events of meiotic resumption, when the first asymmetric division occurs culminating with PB-I extrusion. To this end, we employed a number of known cytological quality markers of the meiotic division to compare MI or MII oocytes matured in vitro, either in the absence or in the presence of 1 nM 2-OHE 2 , and those obtained by ovulation. Specifically, we analysed (i) the PB-I volume [6,[34][35][36][37][38][39][40][41][42][43] and its position in respect to the oocyte's spindle [44]; (ii) the presence and extension of the cortical F-actin cap [45][46][47]; (iii) the meiotic spindle shape and area [48][49][50][51] and (iv) the microtubule organisation centre (MTOC) localisation [8,48,52].

PB-I Volume and Position
Based on the PB-I volume, we identified the presence of three main MII oocytes groups: group I, with a volume comprised  ) were classified in group II (p = 0.601) or III (p = 0.003), respectively. Group III oocytes appeared similar to a 2-cell embryo, indicating the occurrence of an almost symmetrical division.
The PB-I position was calculated as the amplitude of the angle measured at the oocyte's centroid between oocyte's and PB-I's chromosomes (Fig. 2a). IVM-2-OHE 2 MII oocytes showed significantly smaller angle compared with IVM MII (p = 0.044) oocytes, but similar (p = 0.399) to that measured in OV MII oocytes (Fig. 2b).

Presence and Extension of the Cortical F-Actin Cap
The eccentric position of the metaphase spindle and the consequent asymmetric meiotic division lead to the formation of two daughter cells with very different volumes. A crucial event during this cortical polarisation is the formation of a thick F-actin cap at the site of PB, just prior to its extrusion. Oocytes from OV, IVM or IVM-2-OHE 2 groups were labelled with phalloidin after 6 h (MI) or 15 h (MII) of maturation. Based on the fluorescence intensity of the F-actin cap, we classified two main groups of oocytes (Fig. 3a): group I with an intensity comprised between 1.8 and 21.8; and a much fainter, group II, comprised between 1.1 and 1.8. All OV MI belonged to group I, compared with 53.8% (p < 0.001) of IVM MI oocytes, whilst the remaining 46.2% IVM MI oocytes showed a fainter, group II, F-actin cap (Fig. 3b). The presence of 2-OHE 2 in the culture medium significantly increased (p = 0.016) to 86.7%; the number of MI oocytes with a clearly visible F-actin cap was not significantly (p = 0.112) different compared with that of OV MI oocytes (Fig. 3b). Similarly, the frequency of IVM-2-OHE 2 MII oocytes with marked F-actin cap (83.3%) resulted not significantly different (p = 0.059) compared with that of OV MII oocytes (97.5%) (Fig. 3c).
Then, we measured the amplitude of the angle formed between the oocyte centroid and the cortical F-actin cap ends in both in vitro-and in vivo-matured oocytes (Fig. 4a). Whilst all OV MI showed an angle comprised between 80°and 130°, 63.3% (p < 0.001) of IVM MI oocytes displayed an angle > 130° (Fig. 4b). When cultured in the presence of 2-OHE 2 , the majority (88.5%) of IVM-2-OHE 2 MI oocytes showed an angle similar to that of OV MI (p = 0.105; Fig. 4b), whereas only 11.5% oocytes had a wider (> 130°) angle. The same Fig. 1 PB-I size. MII oocytes from the in vitro-maturated (IVM), 1 nM IVM-2-OHE 2 or ovulated (OV) groups were analysed to determine the PB-I volume (V) and divided (absolute percentage) into three groups (a): group I, with a volume comprised between 1 × 10 3 < V ≤ 15 × 10 3 μm 3 ; group II, 15 × 10 3 < V ≤ 30 × 10 3 μm 3 and group III, V > 30 × 10 3 μm 3 . (b) Following treatment with 2-OHE 2 , we observed a significant increase of MII oocytes with a correct group I PB-I size, although it remained lower compared with that of OV MII . In the same group, different superscript letters indicate the presence of a significant difference. Bar, 40 μm analysis performed on MII oocytes gave similar results (Fig.  4c).

Meiotic Spindle Shape and Area
Oocytes from the three experimental groups were fixed after 6 h (MI) or 15 h (MII) of maturation, immunolabelled with anti-alpha tubulin to visualise their spindle and counterstained with DAPI.
The shape of the metaphase spindle was calculated as the ratio between the central width (Fig. 5a, y) and the mean between the width at the two poles (Fig. 5a, x and x′). Two main spindle shapes were observed: a barrel-like spindle (Fig.  5b) with a ratio value ≥ 1.8 and a rectangular-like spindle (Fig.  5c) with a ratio value < 1.8.
All the OV MI oocytes displayed a barrel-like shape with an area of 173.6 ± 8.6 μm 2 (Table 3). When compared with IVM MI , the latter showed a significant (p = 0.004) decreased frequency (75%) of oocytes with a barrel-like spindle. The maturation of COCs in the presence of 2-OHE 2 decreased the frequency of oocytes with a rectangular-like spindle to values not significantly different (13.3%; p = 0.077) compared with that of OV MI .
Similarly, all OV MII oocytes showed a barrel-like spindle, whereas the two IVM conditions produced both barrel-and rectangular-like spindles. Again, IVM in the presence of 2-OHE 2 gave a significantly (p = 0.011) higher frequency (81%) of oocytes with barrel-like spindle compared with those matured in the absence of the hormone (53.8%).

MTOC Localisation
Gamma-tubulin, a component of the microtubule organisation centres (MTOCs), plays a key role in spindle formation during meiosis [53]. Here, we analysed the MTOCs localisation in MI and MII oocytes obtained from the three experimental conditions. Double immunostaining with αand γ-tubulin showed two main patterns of MTOCs localisation: one with all MTOCs clustered at each of the two spindle poles (single signal) (Fig. 6a); the other, with the MTOCs dispersed at the two spindle poles (multiple signals) (Fig. 6b).
The great majority of control OV MI (97.2%) or OV MII (93.3%) oocytes displayed only a single signal pattern of MTOC localisation associated to a barrel-like spindle (Table 4). Instead, only 33.3% of IVM MI or 14.3% IVM MII oocytes showed the combination of a barrel-like spindle and a single MTOC signal. These frequencies increased significantly when COCs were matured in the presence of 2-OHE2 with 57.9% (p = 0.039) of IVM-2-OHE 2 MI and 57.1% (p < 0.001)

Discussion
This study demonstrates a dose-dependent response to the presence of 2-OHE 2 during the GV-to-MII transition on the oocyte meiotic and developmental competence (Fig. 7). Although the overall quality of oocytes cultured in the presence of 2-OHE 2 remained lower compared with that of control OV oocytes, the presence of this catecholestrogen clearly produced a highly significant improvement when compared with COCs that were matured in the absence of 2-OHE 2 . Our results show that whilst the rate of meiotic resumption remained unchanged at doses comprised between 0.1and 10 nM, it decreased significantly at the highest dose employed. It was the combined observation of the developmental rate and blastocyst quality that highlighted the highly significant improvement obtained following GV-to-MII maturation in the presence of 1 nM dose, with a sharp 21% increase of embryos that completed preimplantation development. Furthermore, these blastocysts displayed higher number Fig. 3 Relative fluorescence intensity of the F-actin cap. The F-actin cap intensity was calculated, on the phalloidin red channel image, drawing a line that, from the cortical region (p), intersects the MI or MII chromosomes, passes through the centroid (c) and reaches the opposite pole (d). The fluorescence intensity was measured as p/d ratio using the ImageJ software. MI or MII oocytes of the three experimental conditions were divided into two groups (a): group I, oocytes with an intensity (I) comprised between 1.8 < I ≤ 21.8; and group II, 1.1 < I ≤ 1.8. Following 2-OHE 2 treatment, the matured MI (b) and MII (c) oocytes (absolute percentage) displayed an Factin fluorescent intensity comparable with that observed in OV MI or OV MII oocytes. Different superscript letters indicate the presence of a significant difference. Bar, 20 μm  of blastomeres, a sign of an intensified cell cycle rate [9,54,55].
This improved developmental competence contradicts earlier studies with bovine COCs that, when matured in the presence of 3 μM of this catecholestrogen, display a 5-fold decrease of development to blastocyst [56]. Also, COC maturation in the presence of a 2-OHE 2 metabolite (2-methoxyestradiol, 2-ME 2 ) caused spindle aberrations, chromosome congression failure and non-disjunction at the time of mouse oocytes meiosis resumption [57] and had negative effects on bovine oocyte developmental competence [56]. The improved results we obtained might find an explanation in the doses used in our experiments (0.1-100 nM), which are closer to the 15 nM physiological condition described in pig preovulatory follicles [26].
To understand the reason of the recorded oocyte developmental competence improvement, we used a number of known cytological quality markers of meiotic resumption. All these markers showed that the presence of 2-OHE 2 smoothed the GV-to-MII transition, favouring a more physiological asymmetric division (Fig. 7). Further studies will be needed to understand the mechanism of action of this catecholestrogen, although three of its described features may give us clues to explain our results. One of these refers to the fact that 2-OHE 2 contains a phenolic functional group that has been shown, in smooth muscles, to work as a potent antioxidant eliminating free radicals [58,59] and protecting membrane phospholipids against peroxidase [60]. Indeed, COCs cultured in the presence of antioxidants (i.e. sodium citrate, α-lipoic acid and acetyl-L-carnitine) complete the GV-to-MII transition with significant reduced number of oocytes with abnormal spindles [61]. In alternative, 2-OHE 2 could work as an E 2 antagonist. Earlier studies showed that culturing catfish post-vitellogenic follicles in the presence of 1-20 μM 2-OHE 2 induced a 6-17-fold increase in the number of oocytes that resumed meiosis [22][23][24]62]. This improvement correlated with increased steroid production [63,64], inhibition of aromatase activity [65] and decreased follicular cAMP, which would result in the removal of the inhibitory effects of E 2 on oocyte maturation and meiotic resumption. Another possible explanation on how 2-OHE 2 improves the female gamete quality is related with its role in the production of nitric oxide, a key factor that regulates germinal vesicle breakdown and PB-I extrusion in mouse oocytes [66]. Interestingly, a recent study showed that 10 nM 2-OHE 2 induced an increase of nitric oxide production by ovine uterine artery endothelial cells, whereas 100 nM was unable to produce this effect [67].
In conclusion, we demonstrated that, when present at 1 nM concentration during mouse COC in vitro maturation, 2-OHE 2 has a beneficial effect on the meiotic and