Role of Major Endocannabinoid-Binding Receptors during Mouse Oocyte Maturation

Endocannabinoids are key-players of female fertility and potential biomarkers of reproductive dysfunctions. Here, we investigated localization and expression of cannabinoid receptor type-1 and -2 (CB1R and CB2R), G-protein coupled receptor 55 (GPR55), and transient receptor potential vanilloid type 1 channel (TRPV1) in mouse oocytes collected at different stages of in vivo meiotic maturation (germinal vesicle, GV; metaphase I, MI; metaphase II, MII) through qPCR, confocal imaging, and western blot. Despite the significant decrease in CB1R, CB2R, and GPR55 mRNAs occurring from GV to MII, CB2R and GPR55 protein contents increased during the same period. At GV, only CB1R was localized in oolemma, but it completely disappeared at MI. TRPV1 was always undetectable. When oocytes were in vitro matured with CB1R and CB2R but not GPR55 antagonists, a significant delay of GV breakdown occurred, sustained by elevated intraoocyte cAMP concentration. Although CBRs antagonists did not affect polar body I emission or chromosome alignment, GPR55 antagonist impaired in ~75% of oocytes the formation of normal-sized MI and MII spindles. These findings open a new avenue to interrogate oocyte pathophysiology and offer potentially new targets for the therapy of reproductive alterations.


Introduction
In mammals, the molecular processes leading to the production of female gametes are controlled by multiple interactions among different modulators, either hormones or paracrine factors [1,2]. Starting from the observation that exogenous plant-derived cannabinoids, as those present in cannabis extracts like hashish and marijuana, negatively impact fertility [3,4], several reports documented the key-role of the so called "endocannabinoid system" (ECS) on virtually all steps of female reproduction, from fertilization to oviductal transport, embryo implant and development, and pregnancy outcome [3,[5][6][7][8][9]. The ECS includes: I) lipid messengers termed "endocannabinoids" (eCBs), such as anandamide (N-arachidonoylethanolamine, AEA) and 2-arachidonoylglycerol (2-AG); II) their receptor targets type-1 (CB 1 R) and type-2 (CB 2 R) cannabinoid receptors, G-protein coupled receptor 55 (GPR55), transient receptor potential vanilloid type 1 channel (TRPV1); III) a number of metabolic enzymes among which the eCB-cleaving fatty acid amide hydrolase (FAAH). Moreover, additional "eCB-like" compounds like N-palmitoylethanolamine (PEA) and N-oleoylethanolamine (OEA), that do not bind to CB 1 R and CB 2 R but can activate GPR55 [10], contribute to modulate eCB signalling. Expression was normalized to Actb and values were reported as mean ± SEM of 4 independent replicates. * p < 0.05 vs GV oocyte of the same experimental group. (B) Representative western blot and quantification of cannabinoid receptor type-1 and -2 (CB1R and CB2R), G-protein coupled receptor 55 (GPR55), and transient receptor potential vanilloid type 1 channel (TRPV1) protein contents. Data are expressed as mean ± SEM of each receptor content after normalization with α/β-tubulin, used as loading control. Experiments were repeated 3 times. * p < 0.05 vs GV oocyte of the same experimental group; ** p < 0.05 vs MI of the same experimental group. GV = germinal vesicle; MI = metaphase I; MII = metaphase II.

Protein Levels of eCB-Binding Receptors in GV, MI, and MII Oocytes
Receptors of eCBs were immunodetected in lysates of in vivo matured oocytes. At GV stage, similar levels of CB1R, CB2R, and GPR55 proteins were found ( Figure 1B; p > 0.05). At MI, CB1R content was significantly reduced ( Figure 1B; GV vs MI and MII, p < 0.05), while CB2R and even more GPR55 expression levels showed a sharp increase ( Figure 1B; GV vs MI: p < 0.05). As reported in Figure 1B, in MII oocytes only GPR55 content raised dramatically ( Figure 1B; GPR55: GV vs MI vs MII, p < 0.05; CBR2: MI vs MII, p > 0.05). TRPV1 signal was always barely detectable at any meiotic stage analyzed ( Figure 1A,B; p > 0.05). Expression was normalized to Actb and values were reported as mean ± SEM of 4 independent replicates. * p < 0.05 vs. GV oocyte of the same experimental group. (B) Representative western blot and quantification of cannabinoid receptor type-1 and -2 (CB 1 R and CB 2 R), G-protein coupled receptor 55 (GPR55), and transient receptor potential vanilloid type 1 channel (TRPV1) protein contents. Data are expressed as mean ± SEM of each receptor content after normalization with α/β-tubulin, used as loading control. Experiments were repeated 3 times. * p < 0.05 vs. GV oocyte of the same experimental group; ** p < 0.05 vs. MI of the same experimental group. GV = germinal vesicle; MI = metaphase I; MII = metaphase II.

Protein Levels of eCB-Binding Receptors in GV, MI, and MII Oocytes
Receptors of eCBs were immunodetected in lysates of in vivo matured oocytes. At GV stage, similar levels of CB 1 R, CB 2 R, and GPR55 proteins were found ( Figure 1B; p > 0.05). At MI, CB 1 R content was significantly reduced ( Figure 1B; GV vs. MI and MII, p < 0.05), while CB 2 R and even more GPR55 expression levels showed a sharp increase ( Figure 1B; GV vs. MI: p < 0.05). As reported in Figure 1B, in MII oocytes only GPR55 content raised dramatically ( Figure 1B; GPR55: GV vs. MI vs. MII, p < 0.05; CBR2: MI vs. MII, p > 0.05). TRPV1 signal was always barely detectable at any meiotic stage analyzed ( Figure 1A,B; p > 0.05).

Localization of eCB-Binding Receptors in GV, MI, and MII Oocytes
In these experiments, the immunolocalization of CB 1 R, CB 2 R, GPR55, and TRPV1 was carried out in oocytes collected in vivo at GV, MI, and MII. At GV stage, all receptors showed a homogeneous distribution over the entire cytoplasm even if CB 1 R immunostaining was more intense than that of CB 2 R and GPR55 (Figure 2A,B). In MI oocytes, while CB 1 R signal intensity was restricted to some cytoplasmic dots, that of CB 2 R and GPR55 increased, being still homogeneously distributed across the germ cell (Figure 2A,B). At MII, fluorescence was very faint for CB 1 R, unchanged for CB 2 R but not for GPR55, as it was remarkably enhanced (Figure 2A,B). In keeping with molecular data, signal intensity of TRPV1 was barely detectable throughout meiotic maturation (Figure 2A,B).

Localization of eCB-Binding Receptors in GV, MI, and MII Oocytes
In these experiments, the immunolocalization of CB1R, CB2R, GPR55, and TRPV1 was carried out in oocytes collected in vivo at GV, MI, and MII. At GV stage, all receptors showed a homogeneous distribution over the entire cytoplasm even if CB1R immunostaining was more intense than that of CB2R and GPR55 (Figure 2A,B). In MI oocytes, while CB1R signal intensity was restricted to some cytoplasmic dots, that of CB2R and GPR55 increased, being still homogeneously distributed across the germ cell (Figure 2A,B). At MII, fluorescence was very faint for CB1R, unchanged for CB2R but not for GPR55, as it was remarkably enhanced (Figure 2A,B). In keeping with molecular data, signal intensity of TRPV1 was barely detectable throughout meiotic maturation (Figure 2A,B). Receptors were labelled with Cy-3 (red), DNA was counterstained by DAPI (cyan). In the upper right-hand corner of GV CB1R, the strong DAPI staining is due to undetached cumulus cell nuclei. Each image was taken at the equatorial plan of the oocyte. Magnification: ×630. Each inset represents a magnified part of ooplasm. GV = germinal vesicle; MI= metaphase I; MII= metaphase II; NC= negative control. (B) Mean fluorescence of CB1R, CB2R, GPR55, and TRPV1 receptors in mouse oocytes collected at different stages of meiotic maturation. Values are expressed as arbitrary units (AU) and are reported as mean ± SEM of 6 oocytes from 3 independent experiments. *p < 0.05 vs GV oocyte of the same experimental group; **p < 0.05 vs MI of the same experimental group.
Analysis of CB1R localization at oocyte plasma membrane (oolemma) revealed its presence at GV stage, while it failed to detect it at MI and MII stage ( Figure 3A). Unlike CB1R, the other 3 receptors were never found at oolemma at any meiotic stage analyzed (GV: Figure 3A; MI, MII: data not shown). To further characterize CB1R dynamics during the transition from GV to MI, oocytes were Analysis of CB 1 R localization at oocyte plasma membrane (oolemma) revealed its presence at GV stage, while it failed to detect it at MI and MII stage ( Figure 3A). Unlike CB 1 R, the other 3 receptors were never found at oolemma at any meiotic stage analyzed (GV: Figure 3A; MI, MII: data not shown). To further characterize CB 1 R dynamics during the transition from GV to MI, oocytes were collected at 0, 3, 5, and 8 h after human chorionic gonadotropin (hCG). It was found that the homogenous distribution of CB 1 R recorded at 0 h was lost 3 h later, when receptor localization was restricted to small microdomains, probably associated with lipid rafts ( Figure 3B). At 5 h, CB 1 R was almost completely compartmentalized in few microdomains, and its signal disappeared from oolemma when oocytes reached the MI stage ( Figure 3B). collected at 0, 3, 5, and 8 h after human chorionic gonadotropin (hCG). It was found that the homogenous distribution of CB1R recorded at 0 h was lost 3 h later, when receptor localization was restricted to small microdomains, probably associated with lipid rafts ( Figure 3B). At 5 h, CB1R was almost completely compartmentalized in few microdomains, and its signal disappeared from oolemma when oocytes reached the MI stage ( Figure 3B).

Effects of Receptor Antagonists on Intraoocyte cAMP Concentration
In these experiments, the effects of SR1 and SR2, antagonists of CB1R and CB2R respectively, on the kinetics of meiotic resumption and cAMP concentration were tested. Antagonists were used alone or in combination at the final concentration of 0.5 µM. As shown in Figure 4A, 30 min after starting culture, almost all oocytes were still arrested at GV stage (Ctr vs SR1, SR2, SR1+SR2; p > 0.05). At 60 min, germinal vesicle breakdown (GVBD) occurred in about 40% of Ctr oocytes, and in about 20% of oocytes treated with SR1, SR2 or both antagonists ( Figure 4A; Ctr vs SR1, SR2, SR1+SR2; p < 0.05).

Effects of Receptor Antagonists on Intraoocyte cAMP Concentration
In these experiments, the effects of SR1 and SR2, antagonists of CB 1 R and CB 2 R respectively, on the kinetics of meiotic resumption and cAMP concentration were tested. Antagonists were used alone or in combination at the final concentration of 0.5 µM. As shown in Figure 4A, 30 min after starting culture, almost all oocytes were still arrested at GV stage (Ctr vs. SR1, SR2, SR1+SR2; p > 0.05). At 60 min, germinal vesicle breakdown (GVBD) occurred in about 40% of Ctr oocytes, and in about 20% of oocytes treated with SR1, SR2 or both antagonists ( Figure 4A; Ctr vs. SR1, SR2, SR1+SR2; p < 0.05).

Effects of Receptor Antagonists on Polar Body I Emission and Spindle Morphology
In these experiments, it was ascertained whether antagonists of CB1R, CB2R, and GPR55 could affect polar body I (PBI) emission and/or the morphology of MI-MII spindles. The presence of SR1, SR2 or ML193 (0.5 µM) did not perturb in vitro maturation (IVM), as the percentage of oocytes reaching MI (>90%; vs Ctr, p > 0.05) and MII stage (~80% PBI; vs Ctr, p > 0.05) were comparable with On the basis of the kinetics of meiotic resumption, in the next set of experiments we tested the hypothesis that CB 1 R and CB 2 R, both able to activate G αi proteins [26], could be involved in meiotic resumption by modulating cAMP intraoocyte concentration [27,28]. To this end, cAMP concentration was determined in oocytes cultured in vitro up to 120 min either in the absence (Ctr) or in the presence of SR1, SR2 or SR1+SR2. At the beginning of culture, cAMP content was 0.30 ± 0.01 fmol/oocyte, and 30 min later it showed a slight yet not significant decrease under all experimental conditions ( Figure 4B; p > 0.05). A sharp decrease in cAMP concentration occurred at 60 min in SR1-, SR2-, and SR1+SR2-treated cells (~0.21 ± 0.012 fmol/oocyte) and even more in Ctr (0.16 ± 0.007 fmol/oocyte) ( Figure 4B, vs. Ctr; p < 0.05). At 90 min, cAMP concentration was undetectable in Ctr oocytes, while it was~0.11 ± 0.015 fmol/oocyte in SR1-, SR2-, and SR1+SR2-treated cells ( Figure 4B; Ctr vs. SR1, SR2, SR1+SR2, p < 0.05). After 120 min, cAMP was no longer detectable in all groups ( Figure 4B; Ctr vs. SR1, SR2, SR1+SR2, p > 0.05). Similar results were obtained in the presence of both antagonists ( Figure 4B).

Effects of Receptor Antagonists on Polar Body I Emission and Spindle Morphology
In these experiments, it was ascertained whether antagonists of CB 1 R, CB 2 R, and GPR55 could affect polar body I (PBI) emission and/or the morphology of MI-MII spindles. The presence of SR1, SR2 or ML193 (0.5 µM) did not perturb in vitro maturation (IVM), as the percentage of oocytes reaching MI (>90%; vs. Ctr, p > 0.05) and MII stage (~80% PBI; vs. Ctr, p > 0.05) were comparable with control. Similar results were obtained when oocytes underwent IVM in the presence of the 3 antagonists (MI: >90%; vs. Ctr, p > 0.05; MII:~82% PBI; vs. Ctr, p > 0.05) ( Table 1). Note: For complete and detailed information on the different IVM procedures see section 4.9 Materials and Methods. All oocytes were analyzed by confocal microscopy for the evaluation of spindle morphology.

Discussion
Our results demonstrate that (i) the four major eCB-binding receptors are expressed in mouse oocytes, but CB 1 R, CB 2 R, and GPR55 expression changes throughout in vivo meiotic maturation. Conversely, TRPV1 expression is always low/undetectable; (ii) CB 1 R and CB 2 R can play a role in meiotic resumption, while GPR55 could be involved in spindle organization.
We found that Cnr1, Cnr2, and Gpr55 mRNAs are expressed in oocytes at GV, while their amounts decrease drastically at MI and MII in keeping with the well-known repression of transcription that follows GVBD [29]. Part of these data are different from those recently reported on CB 2 R mRNA, that was found to be expressed throughout in vivo maturation [25]. This could be due to the different methodology used for MI and MII oocyte recruitment utilized in the two studies. Indeed, we obtained MI by puncturing ovarian follicles 8 h after hCG and MII oocytes from the oviducts 12 h after hCG, while López-Cardona and colleagues retrieved both MI and MII from the oviducts 14 h after hCG [25]. As for humans, only CB 1 R transcripts were reported in mouse oocytes [21], though cells of different meiotic stages were pooled together, thus preventing any conclusion on their modulation during meiosis.
In our experiments, localization of CB 1 R at GV oolemma is modified during GVBD, as CB 1 R is progressively clustered in large microdomains probably associated with lipid-rafts, a preferential localization documented also in neuronal cells [30]. We hypothesized that CB 1 R disappearance from oolemma could be relevant for fertilization and/or early embryo development. It has been demonstrated that in the ampulla, where fertilization occurs, AEA concentration is low, but that spermatozoa here present have undergone CB 1 R-dependent acrosome reaction (AR) in the isthmus, where AEA concentration is high [12,31]. As a consequence, we cannot exclude that the permanence of CB 1 R at the oocyte plasma membrane could trigger inappropriate signalling during the meiotic maturation and fertilization process. Moreover, it is worth to notice that both CB 1 R and CB 2 R are expressed in the early embryo (CB 2 R in the zygote and CB 1 R from 2 cell-embryo stage onward [32]), but with different and still unexplained roles. In fact, while CB 2 R is unresponsive to agonist stimulation and its role has not been yet identified [25,33], CB 1 R plays a key role in embryonic development [33,34]. To date, nothing is yet known about the other receptors.
Our results show that CB 2 R, GPR55, and TRPV1 receptors have different dynamics during meiotic maturation. While TRPV1 is weakly expressed at any meiotic stage, CB 2 R content increases from GV to MI and that of GPR55 throughout the whole meiotic maturation. Our results on CB 1 R and CB 2 R are different from those reported by López-Cardona and colleagues [25], who showed that localization of CB 1 R is differentially influenced by in vivo (periphery of the oocyte from GV to MII) or in vitro maturation (firstly in the cytoplasm at GV, and then peripherally at MII). Such a different distribution is unexpected, since oocytes collected at GV display identical properties regardless of conditions adopted for subsequent (in vivo or in vitro) maturation. Also, for CB 2 R, López-Cardona and colleagues [25] found a homogeneous immunofluorescence in the cytoplasm at all meiotic stages, while we recorded a sharp increase of CB 2 R signal from GV to MI. Although the different procedures used to collect oocytes might explain this discrepancy, the quantification of protein expression performed here strongly supports confocal images.
An additional point of interest is that López-Cardona and colleagues [25] concluded that CB 1 R is more important than CB 2 R in the control of oocyte maturation. Instead, here a role for both receptors in the control of meiotic resumption is supported by experiments with selective receptor antagonists SR1 and SR2. Indeed, the significant delay of GVBD and higher cAMP concentration allow us to conclude that both receptors can modulate oocyte adenylyl cyclase activity, possibly through G αi proteins coupled to them, as shown in other cell systems [26]. It is well known that meiotic arrest in GV oocytes is maintained by a high intra-oocyte cAMP concentration [27,28]. The autonomous production of this cyclic nucleotide is ensured by the presence of active GPR3 receptors coupled to G αs proteins, which maintains the high levels of cAMP necessary for GV arrest via activation of adenylyl cyclase type 3 [28,35]. Following gonadotropin stimulation, cAMP concentration in the oocyte falls down sharply around the time of GVBD, due to gap junction closure and PDE3A activation [35][36][37]. Lowther et al. [38] found that also GPR3 endocytosis through a beta-arrestin/GRK-independent mechanism participates in meiotic resumption. Since we observed that CB 1 R becomes entrapped in large clusters soon after GVBD and disappears from the oolemma at MI, we hypothesize that the concomitant internalization of CB 1 R and GPR3 could be part of the mechanism(s) involved in the control of meiotic resumption. As a matter of fact, receptor endocytosis modulates in the control of GVBD in vertebrate oocytes [39].
The co-existence in GV oocytes of serpentine receptors coupled to G αi proteins, localized either in plasma membranes (CB 1 R) [26,30] or intracellularly (CB 2 R) [40], could be instrumental to properly regulate intra-oocyte cAMP concentration and GVBD [41]. This hypothesis is further corroborated by the presence of functional microdomains for cAMP production in plasma membrane and cytoplasm of many cell types [42], as in fish oocytes [43] and in rat oocyte nucleus [44].
From our experiments it is also evident that the use of SR1 and SR2 during IVM does not induce any variation in meiotic spindle structure and in the percentage of MII oocytes extruding normal PBI. This result is in agreement with the observation by López-Cardona and colleagues [25] on Cnr1 or Cnr2 knockout mice, although the presence of both receptors during oocyte growth and maturation is essential for successful fertilization and embryogenesis.
Conversely, a novel role of GPR55 in the formation of MI and MII spindles has been revealed by experiments with its antagonist. In fact, despite nearly all the oocytes cultured with ML193 extrude normal PBI, a high percentage of them exhibits variation in spindle size at both MI (75%) or MII (75%) stage compared with Ctr (19% for MI, 5% for MII), without compromising chromosome alignment at both metaphase plates. To the best of our knowledge, this is the first time that a link between GPR55 and spindle organization has been found in mammalian cells. In mammalian oocytes, spindle length is controlled by a complex interplay of many proteins [45][46][47], and its normal length is considered a marker of oocyte quality [48]. Recently, Wang and collaborators proposed that spindle size and timing of meiotic progression are efficiently controlled more by cytoplasmic than by nuclear components [49]. In keeping with this hypothesis, dynein, dynactin and NuMA are able to control microtubule length, while γ-tubulin regulates the polymerization of α-tubulin [50]. Although we do not know how GPR55 could affect spindle morphology yet, here we show that GPR55 protein is entirely localized in the cytoplasm and its expression raises in a stepwise manner from GV to MI and, more greatly, to MII. This finding supports a possible role for this receptor not only in meiotic maturation but, at later times, also in fertilization and embryo development. It is of interest that GPR55 can mediate Ca 2+ mobilization from IP3-sensitive intracellular stores via G q , G α12 , RhoA, PLC, and actin [51], and that an essential role for GPR55 activation in the Ca 2+ -dependent regulation of human sperm motility and capacitation has been proposed [52,53]. How GPR55 can participate in the oocyte spindle formation is currently under study in our laboratories. Altogether, these observations suggest that we are still far from understanding the importance of the whole ECS in female reproduction.

Animals
Mus musculus Swiss CD1 female mice (23-25 day old; Charles River Laboratories, Lecco, Italy) were housed in an animal facility under controlled temperature (21 ± 1 • C) and light (12 h light/day) conditions, with free access to food and water.
All mice were injected with PMSG (5 IU, i.p.). Forty-two to forty-four hours later, mice were sacrificed to obtain preovulatory germinal vesicle (GV)-stage oocytes or were injected with hCG (5 IU, i.p.) and sacrificed 3-12 h after hCG injection (depending on experimental protocols). MI and MII oocytes were retrieved 8 and 12 h after hCG injection, respectively.

Collection of In Vivo Matured Oocytes
Fully-grown, GV-stage oocytes surrounded by cumulus cells (oocyte-cumulus cell complexes, OCC) were collected in MEM-HEPES supplemented with 0.23 mM pyruvic acid, 2 mM l-glutamine and 0.3% BSA (here referred as MEM), and immediately devoid of cumulus cells by gentle pipetting. MI and MII oocytes were recovered from ovaries 8 h after hCG and from fallopian tubes 12 h after hCG, respectively. When needed, cumulus cells were removed by a brief hyaluronidase treatment [54]. Oocytes were immediately used for morphological or molecular analysis.

Quantitative Real-Time PCR Analysis
Total RNA was extracted from GV, MI and MII oocytes (20 oocytes/sample) using the RNeasy extraction kit (Qiagen, Crawley, UK) as suggested by the manufacturer. Starting with 100 ng of RNA, complementary DNA (cDNA) was prepared using M-MLV reverse transcriptase kit (ThermoFisher Scientific, Waltham, MA, USA). Quantitative PCR analysis was performed using SYBR Green I Master and the LightCycler 480 System (Roche, Basel, Switzerland) on a DNA Engine Opticon 2 Continuous Fluorescence Detection System (BioRad, Hercules, CA, USA). The reaction was performed using the following qRT-PCR program: 95 • C for 10 min, followed by 40 amplification cycles of 95 • C for 10 s, 57 • C for 30 s, and 72 • C for 30 s. The primer used for the amplification of Cnr1, Cnr2, Gpr55, and Trpv1 [55,56] were listed in Table 2 and all the data were normalized to the endogenous reference gene β-Actin. Relative quantitation of mRNAs was performed by the comparative ∆∆Ct method [57].  4.6. Western Blotting Analysis GV, MI, and MII oocytes (150 oocytes/sample) were lysed in sample buffer containing protease inhibitors (2 mM phenylmethyl sulphonyl fluoride, 10 µg/mL aprotinin, 0.1 mM sodium pyrophosphate, 10 mM sodium fluoride, and 1 mM sodium orthovanadate). Lysates were separated by electrophoresis and transferred to nitrocellulose membranes (Hybond C Extra, Amersham, UK). Membranes were incubated with antibodies against CB 1 R (1:200), CB 2 R (1:200), GPR55 (1:200), and TRPV1 (1:200) overnight at 4 • C. HRP-conjugated goat anti-rabbit IgG (1:5000) was used as secondary antibody (1 h, room temperature); peroxidase activity was detected using a SuperSignal West Pico Chemiluminescent substrate. Membranes were examined by Alliance LD2-77WL imaging system (Uvitec, Cambridge, UK). Densitometric quantification was performed with the public-domain software NIH Image 167 V.1.62 and standardized using tubulin as loading control. Negative controls were prepared using specific blocking peptide (for CB 1 R and CB 2 R and GPR55). For the anti-TRPV1 antibody used in this study, there are no blocking peptides commercially available.

Immunofluorescence
To detect presence and distribution of receptors at GV, MI and MII stage, oocytes (15/sample) were fixed in 4% paraformaldehyde for 10 min at r.t., permeabilized with 0.1% Triton X-100 for 30 min at 37 • C [54,58]. Afterwards, oocytes were incubated with the following primary antibodies diluted in a PBS blocking solution (containing 2% BSA, 2% powder milk, 2% normal goat serum, 0. For each set of experiments, negative controls (NC) were prepared using specific blocking peptide (for CB 1 R and CB 2 R and GPR55) and omitting the primary antibody for TRPV1 before addition of the secondary antibody. For the anti-TRPV1 antibody used in this study there are no blocking peptides commercially available. All the oocytes were observed by confocal microscopy (Leica System TCS SP5 confocal microscope, Wetzlar, Germany). Images were taken at the equatorial plan using the LAS AF software (Leica Microsystems).
For image analysis, data from high-resolution images of 6 oocytes from 3 independent experiments were acquired for each sample. Quantification of the intracellular mean fluorescence of CBRs was carried out using the public-domain software NIH Image 167 V.1.62 after the subtraction of the background intensity calculated from the images of NC.

Effects of CB 1 R and CB 2 R Antagonists on Intraoocyte cAMP Content
OCCs were collected in MEM supplemented with cilostamide (1 µM) to maintain meiotic arrest [60]. After washing, OCCs were (i) in part devoid of somatic cells to obtain GV stage oocytes (t = 0) that were immediately stored at −80 • C; (ii) in part cultured at 37 • C in 5% CO 2 for 30, 60, 90, and 120 min in the absence (Ctr) or presence of CB 1 R antagonist SR141716 (SR1) and CB 2 R antagonist SR244528 (SR2), alone or in combination. The antagonists were all used at 0.5 µM because it was the lowest effective dose after preliminary experiments, and for CB 1 R and CB 2 R was in line with previous study [57]. Culture medium was alpha MEM supplemented with 0.23 mM pyruvate, 2 mM l-glutamine and 0.05% DMSO (hereafter referred as αMEM-DMSO). At each time point, OCCs were deprived of cumulus cells to record the percentage of GVs, according to the presence or absence of the GV in the ooplasm.
The amount of cAMP was determined in groups of 120 oocytes incubated stored after GV assessment, by using a Cyclic AMP EIA Kit (581001, Cayman Chemical Company, Anne Arbore, MI, USA) according to manufacturer's instructions. Absorbance at 420 nm was measured in a Model 550 microplate reader (BioRad).

Effects of Receptor Antagonists on Polar Body I Emission and Spindle Formation
To evaluate the effects of antagonists on the morphology of MI spindle, OCCs were cultured for 8 h in 300 µL αMEM-DMSO in the absence (Ctr, n = 30 oocytes) or presence of 0.5 µM SR1 (n = 45 oocytes), SR2 (n = 45 oocytes), ML193 (n = 80 oocytes), or a combination of the three antagonists (SR1 + SR2 + ML193; n = 50 oocytes). Following cumulus cells removal, only oocytes undergoing GVBD were fixed as described in the following procedure.
The analysis of antagonists' effect on PBI and MII spindles were performed by retrieving OCCs 8h after hCG, i.e., at MI in vivo, and then by culturing them in the absence (Ctr, n = 30 oocytes) or presence of 0.5 µM SR1 (n = 30 oocytes), SR2 (n = 50 oocytes) or ML193 (n = 80 oocytes), or a combination of the three antagonists (SR1+SR2+ML193; n = 50 oocytes) for 5 h. This experimental design was chosen in order to reduce the times of oocyte in vitro culture. By the end of culture period, the percentage of normal PBI was recorded, and oocytes were then fixed as described above. Afterwards, oocytes were incubated for 1 h at 37 • C with anti-tubulin primary antibody (1:100) and then with anti-mouse secondary antibody conjugated with Alexa Fluor 488 (1:1000). Chromosomes were labelled with DAPI (1:1000) [58,59]. Spindle sizes (length [49] and area [61]) were measured by the software ZEN 2009 Light Edition (Carl Zeiss MicroImaging GmbH), as previously described [49].

Statistical Analyses
All experiments were performed at least 3 times, and data obtained were expressed as the mean ± S.E.M. Statistical analysis was performed using ANOVA followed by the Tukey-Kramer post-test for comparison of multiple groups, by Bonferroni post-test for comparison among treatments and control groups and by the chi-square test for comparison of percentages. Values of p < 0.05 were considered significantly different.

Conclusions
Our results demonstrate that in mouse oocytes the major eCB-binding receptors are differentially expressed and modulated during meiotic maturation. Present data support a prominent role for CB 1 R and CB 2 R in the control of meiosis resumption, and the engagement of GPR55 in MI and MII spindle organization. These findings open a new avenue to interrogate oocyte pathophysiology and offer potentially novel biomarkers for fertility problems.

Acknowledgments:
The authors thank Mariangela Pucci (University of Teramo), Valeria Gasperi (Tor Vergata, University of Rome), Annalisa Castellucci and Alessio Ferrari (University of L'Aquila) for their kind help in the preliminary steps of this project.

Conflicts of Interest:
The authors declare no conflict of interest.