Introduction

Infertility is a disease that affects about 17.5% of the population at reproductive age, with female, male and combined causes similarly distributed1. Controlled ovarian hyperstimulation (COS) is the key for successful Assisted reproductive technology (ART) treatments, aiming to achieve a sufficient number of competent oocytes in order to originate a heathy newborn (NB)2,3.

Although almost of the retrieved oocytes are mature at the metaphase-II stage (MII), around 10–30% are immature, either at the prophase-I stage, also known as germinal vesicle stage (GV), or at metaphase-I stage (MI)4,5. Several mechanisms were suggested to explain why some oocytes are unresponsive in-vivo to the exogenous trigger for triggering a full nuclear maturation, such as follicle growth desynchrony, premature oocyte retrieval, aspiration of smaller antral or degenerating follicles, and poor response to stimulation5. Apart special cases of recurrent predominant or total oocyte immaturity, which are generally due to structural or genetic situations6, most cycles present between 6–11% of GV oocytes and 9–12% of MI oocytes7.

Immature oocytes are generally discarded, except when the number of mature oocytes is insufficient for a successful embryo transfer program. In these situations, when a quick spontaneous maturation occurs (mainly in the case of MI oocytes), these oocytes are used in the normal intracytoplasmic sperm injection (ICSI) timing. In those cases where the rescue is delayed because in-vitro maturation (IVM) is needed, these oocytes are still used although in a latter ICSI timing7. Oocytes that have completed their maturation in-vitro can be fertilized by ICSI; however, they were associated with poor embryonic development potential8,9,10, and derived embryos showed a high frequency of multinuclear blastomeres as well as aneuploidies4,11,12.

Oocyte developmental competence is required for proper fertilization and subsequent embryo development. This is achieved during folliculogenesis, where the oocyte acquires nuclear and cytoplasmic competence in a synchronized maturation process13,14,15. During nuclear maturation, meiosis resumption occurs. Under the LH surge, the GV oocyte nuclear envelope ruptures (GV-breakdown stage), followed by chromosome condensation and alignment in the first meiotic spindle (MI stage), and then the first meiotic division occurs with extrusion of the first polar body (PB1)13,14,15. This process occurs simultaneous with cytoplasmic maturation, which involves remodeling of the cytoskeleton and organelles13concomitant with ooplasm accumulation of mRNA, proteins, nutrients and substrates16,17. Communication between the oocyte and the surrounding cumulus cells, as well as the energy metabolic pathways of the follicular environment, are also important for oocyte competence establishement17,18.

The morphology of the human oocyte was evaluated by transmission electron microscopy (TEM), enabling a qualitative description of organelle disposition along maturation. These involved the study of oocytes retrieved during infertility treatment cycles, donor oocytes, after vitrification and after IVM19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36.

Pires-Luís37and colleagues reported a pioneering three-dimensional quantitative ultrastructural analysis of GV oocytes using a stereological approach. The same approach was then applied to MI oocytes38. These studies first described the relative volumes of the organelles in each oocyte region and their specific spatial distribution.

As a final part of this effort to quantitatively evaluate the human oocyte during the three main maturation stages, we here present the quantitative evaluation of MII oocytes. For this it was used the same stereological method to describe the relative volumes occupied by organelles in order to uncover their spatial distribution and volume occupied in the mature oocyte. We believe that in addition to gathered knowledge, results will be useful to assist in improvements of IVM media and cryopreservation techniques applied to human oocytes.

Results

Oocyte morphological description

Nuclear maturity and absence of extracytoplasmic and intracytoplasmic dimorphisms of the studied oocytes were confirmed in live-cells, and in semithin and ultrathin sections. All present observations were common to the five MII oocytes. Under the inverted microscope (Fig. 1A), live MII oocytes presented a spherical shape, a homogeneous fine granular ooplasm, a thick translucent zona pellucida (ZP) and the PB1. In semithin sections (Fig. 1B, C), nuclear maturity was confirmed by the presence of the PB1 and metaphase plate-II, with absence of oocyte dimorphisms. In ultrathin sections, nuclear maturity was further confirmed by the presence of the PB1 and metaphase plate-II (Fig. 2A, B). The PB1 was identified by the presence of chromosomes and MII ooplasm cortical organelles, such as cortical vesicles, mitochondria and SER vesicles (Fig. 2A). The metaphase plate-II was located in the ooplasm near to the PB1 region, crossing the three oocyte regions (Fig. 2A, B). Dense cortical vesicles formed rows under the oolemma, with a few being observed isolated in the subcortex region (Fig. 2A, C). Tiny coated-vesicles were also observed under the oolemma (Fig. 2D). Mitochondria, smooth endoplasmic reticulum (SER) small vesicles (SER-SV), SER medium vesicles (SER-MV), SER large vesicles (SER-LV), SER isolated tubules (SER-IT) and SER tubule aggregates (aSERT) were found in all oocyte regions (Fig. 3A-E). Lysosomes were only observed in the inner ooplasm (Fig. 3F). Dictyosomes, vesicles containing zona pellucida-like materials, multivesicular bodies, annulate lamellae, rough endoplasmic reticulum, lipid droplets and polyribosomes were not observed.

Fig. 1
figure 1

Metaphase-II (MII) oocyte. (A) Live oocyte observed in the inverted microscope. (B, C) Oocyte observed in semithin sections. The oocyte (O) has a spherical shape and was separated from the zona pellucida (ZP) by the perivitelline space (pvs), where the first polar body (PB1) was located. In the PB1, it was possible to detect chromosomes (chr) and cortical vesicles (cv). The oocyte metaphase II plate (mp) extended from the cortex to the inner region. The ooplasm presented a homogeneous appearance. In the oocyte it could be detected surface microvilli (mv), cortical vesicles (cv) under the oolemma, mitochondria (mi), smooth endoplasmic reticulum (SER) small tubular aggregates (aSERT), SER large vesicles (LV), and lysosomes. Bars: (A) 20 μm; (B, C) 10 μm.

Full size image
Fig. 2
figure 2

Ultrastructural images of metaphase-II oocytes. (A) First polar body (PB1), showing chromosomes (chr), chromosomal microtubules (mt), mitochondria (mi), smooth endoplasmic reticulum (SER) small vesicles (SV) and cortical vesicles (cv). The PB1 is inserted in the perivitelline space (pvs), between the zona pellucida (ZP) and the oolemma. The oocyte surface contained microvilli (mv), outside the PB1 apposition region. In the ooplasm it can be observed cortical vesicles (cv), mitochondria (mi), smooth endoplasmic reticulum (SER) small (SV), medium (MV) and large (LV) vesicles, SER isolated tubules (IT), and SER small tubular aggregates (*). (A, B) Oocyte chromosomes (chr) at the metaphase-II plate. (A, C, D) Cortical vesicles (cv) were observed to form layers under the oolemma, with some presenting decondensed contents. (D) Tiny coated-vesicles (arrows) were observed at the oocyte surface. Bars: (A, B, C) 1 μm; (D) 0.5 μm.

Full size image
Fig. 3
figure 3

Ultrastructural images of metaphase-II oocytes. (A) Delimitation of the three oocyte regions, cortex (C), subcortex (SC) and inner cytoplasm (IC). (A-C) Oocyte cortex. (A, C) Oocyte subcortex. (D-F) Oocyte inner cytoplasm. Cortical vesicles (cv) accumulate under the oolemma but are also observed isolated in the cortex and subcortex. The ooplasm is rich in mitochondria (mi), smooth endoplasmic reticulum (SER) small (SV), medium (MV) and large (LV) vesicles, SER isolated tubules (IT) and SER small tubular aggregates (*). Lysosomes (Ly), rich in lipid droplets (L), were observed only in the inner cytoplasm. ZP: zona pellucida; pvs: perivitelline space; mv: microvilli. Bars: (A, B, C, D, E, F) 1 μm.

Full size image

Stereological analysis

The stereological analysis was performed in a total of 556 photos that included 197 representations of the cortex, 199 of the subcortex, and 310 of the inner cytoplasm. Cortical vesicles, mitochondria, SER elements (SER-SV, SER-MV, SER-LV, SER-IT and aSERT) and lysosomes were quantified.

Globally (Table 1.), total SER and mitochondria were the most prevalent organelles, presenting a homogeneous distribution, followed by aSERT (predominance in the peripheral oocyte region). The total SER included SER-SV (homogeneous distribution), SER-IT (predominance in the peripheral regions), SER-MV (predominance in the cortical and inner regions) and SER-LV (predominance in the cortical and inner regions). Cortical vesicles (predominance in the cortical region), and lysosomes (only observed in the inner region) were the least predominant organelles. Per oocyte region (Table 1.), the most predominant organelles in the cortex region were total SER, mitochondria and aSERT, followed by cortical vesicles. In the subcortex region, the most predominant organelles were also total SER, mitochondria and aSERT. In the inner cytoplasm region, the most predominant organelles were total SER and mitochondria. No lysosomes ware observed in the cortex or subcortex, nor cortical vesicles in the inner cytoplasm.

Table 1. Relative volume (Vv) of organelle per total MII oocyte and per MII oocyte regions.
Full size table

The Kruskal–Wallis test was used to compare the three oocyte regions (Table 2). No significant differences were detected in the Vv regarding the distribution of mitochondria, SER-SV, SER-MV, SER-IT and aSERT. Significant differences were found for lysosomes, cortical vesicles and SER-LV. The Mann–Whitney U-test, Bonferroni corrected, was used for strict pairwise comparisons for the three regions (Table 2), which showed significant differences in the Vv of lysosomes between cortex vs inner cytoplasm and between the subcortex vs inner cytoplasm; significant differences in the Vv of cortical vesicles between all regions of the oocyte; and significant differences in the Vv of SER-LV between the cortex vs inner cytoplasm and subcortex vs inner cytoplasm.

Table 2 Comparison of the means of organelle relative volume (Vv) between MII oocyte regions.
Full size table

Discussion

The knowledge of human oocyte morphology is a prerequisite to uncover the physiological mechanisms that regulate oocyte development, on which are based development of COS protocols, IVM media and cryopreservation techniques.

This knowledge has been sustained by ultrastructural qualitative descriptions of the different oocyte maturation stages. These included analysis of ovarian tissue39,40, fetal tissue41, different stages after IVM of antral follicles retrieved from the ovary42, mature oocytes retrieved after COS41, immature oocytes retrieved after COS22,23,24,25,27,43,44,45, different stages after IVM of immature oocytes retrieved after COS20,36,46,47, cryopreserved mature oocytes after IVM of immature oocytes retrieved after COS31,34,48, and of cryopreserved mature oocytes retrieved after COS29,30,32,33,49,50.

In those qualitative observations, cortical vesicleswere observed to originate from dictyosomes at the GV stage21,40, increasing in number in the MI stage19,24,25,27, with progressive migration toward the surface where they formed a row under the oolemma in MII oocytes19,24,25. Upon fertilization, cortical vesicles undergo exocytosis in a calcium-dependent process, which is responsible for inhibiting polyspermy through ZP modifications51,52. Dictyosomeswere observed very active at the GV stage21, then dispersed and migrated towards the oocyte cortex40, being rarely observed at the MII stage21,24,27. Mitochondriawere described as the most abundant organelle23, concentrated in the central area of GV, and then becoming homogenously dispersed during maturation26, with an apparent increase in number throughout maturation27. The SERwas also referred as the most predominant organelle, being observed as uniformly distributed25,27,42, with an apparent increase in number throughout maturation22,27. aSERTfirst appeared at the MI oocyte stage22,23,24,25,26,27,42, becoming concentrated in the cortex and subcortex regions of MII oocytes; they behave as calcium stores that participate in the induction of calcium oscillations after gamete fusion, then disitegratting25,53. At the MII stage, in the oocyte cortex and subcortex, SER-SV, SER-MV, SER-LV and aSERT became surrounded by mitochondria. These SER-mitochondria complexes were shown to act as calcium stores25,53. Lysosomeswere observed homogeneously distributed in all stages, with their number diminishing with maturation24,27. There are several other organelles and structures barely reported, mainly in GV oocytes, such as multivesicular bodies22,24, annulate lamellae27,40, rough endoplasmic reticulum, polyribosomes and lipid droplets24.

These qualitative observations, although of extreme importance, are not able to provide the real three-dimensional spatial relative distribution of the organelles inside the oocyte, as well as their relative occupied volume. To solve this question, our group initiated a stereological analysis of human oocytes during the three main developmental stages. In this context, we previously described the quantitative analysis for GV37and MI38oocytes. In the present study, we employed the same stereological approach to analyze the organelle spatial relative volume distribution in mature MII oocytes, thus enabling a strict comparison to the previous immature oocyte stages. A previous study developed an alternative method of quantification, although only regarding the distribution of mitochondria54; in that work, authors used spontaneous in-vitro maturated GV and MI oocytes; after mitotracker labelling, live immature oocytes were exposed to confocal microscopy during 42 h; results were then confirmed by TEM. Despite this long-term exposure to a laser beam, and the use of IVM, authors evidenced similar results to the present observations regarding mitochondria.

Comparisons between the present results on MII oocytes to our previous observations on immature oocytes were possible due to the use of the same quantitative approach; this enabled the evaluation of the changes in organelle distribution along the three oocyte developmental stages, having been found relevant significant differences (Tables S1-S4, Fig. 4).

Fig. 4
figure 4

Graphical representation of the variations in organelle relative volume (Vv) observed in (A) the total oocyte, and per oocyte region, (B) cortex, (C) subcortex and (D) inner cytoplasm. Mi: Mitochondria; Di: dictyosomes; Ly: lysosomes; CV: cortical vesicles; VZP: vesicles containing zona pellucida-like material; VZ: vesicles containing granular material; MVB; multivesicular bodies; AL: annulate lamellae; SER: smooth endoplasmic reticulum; VLV: SER very large vesicles; LV: SER large vesicles; MV: SER medium vesicles; SV: SER small vesicles; IT: SER isolated tubules; aSERT: SER tubular aggregates.

Full size image

In GV oocytes there was an exclusive presence of certain organelles, such as very large vesicles, vesicles containing ZP-like materials, multivesicular bodies and annulate lamellae; in MI oocytes there was an exclusive presence of small vesicles containing ZP-like materials; and MII oocytes did not present dictyosomes. The rough endoplasmic reticulum, polyribosomes and lipid droplets were not found at any oocyte stage.

Dictyosomes were observed in the cortex of GV and MI oocytes, in the subcortex of GV oocytes, and in the inner ooplasm of GV and MI oocytes. From GV oocytes to the MI oocyte, dictyosomes significantly decreased, entirely disappearing in MII oocytes. Lysosomes were observed in the cortex and subcortex of GV and MI (NS predominance) oocytes, and in the inner ooplasm of GV, MI and MII oocytes. They were more prevalent in MI oocytes, and in MII oocytes they were only observed in the inner cytoplasm.

Cortical vesicles were observed in the cortex of GV (isolated), MI (isolated) and MII (NS predominance, forming rows under the oolemma) oocytes, in the subcortex of GV, MI and MII oocytes (isolated), and in the inner ooplasm of GV and MI oocytes (isolated). They were predominantly found in the cortical region throughout all oocyte stages, being absent from the inner cytoplasm in MII oocytes, indicating that progressively migrated to the oolemma.

SER-SV were observed in all regions of GV (predominance to MI), MI and MII (predominance to MI) oocytes. They thus revealed a significant decrease when moving from the GV to MI stage and a significant increase from MI to MII stage. In the cortex of MII oocytes most were associated with one mitochondrion. SER-MV were observed in all oocyte regions of GV, MI (predominance to GV) and MII (predominance to GV) oocytes. In contrast to SER-SV, they exhibit a significant increase from GV to MI stage, followed by a significant decrease from MI to MII stage. In the cortex of MII oocytes they were associated with mitochondria. SER-LV were observed in the cortex and inner ooplasm of GV, MI and MII (NS predominance to GV and MI) oocytes. In the cortex of MII oocytes they were associated with mitochondria. They did not change between GV and MI, although there was a non-significant increase from MI to the MII stage. SER-IT were observed in all regions of GV, MI (predominance to GV and MII) and MII oocytes. There was thus a significant increase from GV oocytes to MI oocytes, followed by a significant decrease from MI oocytes to MII oocytes, probably due to the formation of aSERT in MII oocytes. Regarding total SER, elements were observed in the cortex of GV, MI (predominance to GV and MII) and MII (predominance to GV) oocytes, and in the subcortex and inner ooplasm of GV, MI (predominance to GV and MII) and MII oocytes. There was thus a significant increase from GV oocytes to MI oocytes, followed by a significant decrease from MI oocytes to MII oocytes.

The aSERT were observed in all regions of MI and MII (predominance to MI) oocytes, with a progressive decrease from the cortex to the inner ooplasm. They are a mark of MII oocytes, and were associated with mitochondria in the cortex and subcortex.

Finally, mitochondria were observed in the cortex of GV, MI (predominance to GV) and MII (predominance to GV) oocytes, being equally distributed in the subcortex and inner ooplasm of GV, MI and MII oocytes. Mitochondria were found predominantly in the central area of GV oocytes, and thenceforth scattered throughout the cytoplasm in MI and MII oocytes, forming SER-M complexes in MII oocytes in the cortex and subcortex.

In conclusion, we here present the first stereological analysis of MII oocytes, and additionally discuss the quantitative spatial distribution of organelles along the major developmental stages of oocyte maturation. We expect that this data will be critical to support more precise developments in ovarian stimulation protocols, in-vitro maturation media and cryopreservation methods applied to human oocytes.

Methods

Ethical approval

Ethical guidelines were followed conducting the research, with written informed consent obtained before experiments. This work did not involve experimentation on humans or animals, and thus the approval of the Ethics Committee and the Helsinki Declaration, revised in Tokyo 2004, on human experimentation does not apply to this work. Surplus donor oocytes donated for research came from a fresh oocyte donation program of a private IVF clinic. According to the determinations of the National Law of Medically Assisted Procreation (Law of 2017) and guidelines of the National Council for Medically Assisted Procreation (CNPMA-2021), the use of clinical databases and patients’ biological material for diagnosis and research may be used without additional ethical approval, as long as the biological material obtained for research is used under strict individual anonymity and after informed and written consent from the patient. The use of human oocytes in laboratory experimentation (processing for microscopy) was authorized by the Ethics Committee ICBAS/CHUP with project number: 2019/CE/P017 (266/CETI/ ICBAS).

Patients

This research was performed on five surplus donor MII oocytes taken from a fresh donor oocyte program. Donor oocytes were recovered from five different oocyte donors (OD), and one surplus MII oocyte per oocyte donor was used for research. OD1 had 33 years old, and MII oocytes were donated for two different recipients using ICSI for fertilization. After transfer of a fresh blastocyst, one recipient had a clinical abortion. The other recipient received a froze-thaw blastocyst and achieved a term pregnancy. OD2 had 25 years old, and oocytes were donated for one recipient using ICSI. After transfer of a fresh blastocyst, the recipient achieved a term pregnancy. OD3 had 31 years old, and only two mature oocytes were retrieved. In this case, the donation cycle was cancelled. OD4 had 29 years old and oocytes were donated for one recipient using ICSI. After transfer of a fresh blastocyst, the recipient achieved a term pregnancy. OD5had 22 years old and oocytes were donated for one recipient using IVF. After transfer of a fresh blastocyst, the recipient achieved a term pregnancy (Table S5). Although these outcomes do not prove that the oocytes here analyzed would enable a LBD, they nevertheless reassure the oocyte competence of the retrieved pool. This is very important as, in a previous study, we could demonstrate that donor oocytes, with proven nuclear maturity, may exhibit cytoplasmic immatury55.

Transmission electron microscopy

Donor MII oocytes were fixed with Karnovsky (2.5% glutaraldehyde, 4% paraformaldehyde, 0.15 M sodium cacodylate buffer) (Sigma-Aldrich, St. Louis, USA; Merck, Darmstadt, Germany) at room temperature for 30 min, followed by 2 h at 4ºC. After washing in 0.15 M sodium cacodylate buffer, pH 7.3 (Merck) for 2 h at 4ºC, oocytes were post-fixed with 2% osmium tetroxide (Merck) in buffer containing 0.8% potassium ferricyanide (Merck) for 2 h at 4 °C, washed in buffer for 10 min, serially dehydrated in ethanol (Panreac, Barcelona, Spain), equilibrated with propylene oxide (Merck) and embedded in Epon (Sigma). Semithin and ultrathin sections were performed with diamond knifes (Diatome, Hatfield, Switzerland) in a LKB ultramicrotome (Leika Microsystems, Weltzlar, Germany). The first cut was determined using a Random Number Table. Then, oocytes were completely serially sectioned and sampled every 10 µm. Ultrathin Sects. (500–700 nm) were collected on 100 mesh formvar carbon-coated copper grids (Taab, Berks, UK) and stained with 3% aqueous uranyl acetate for 20 min (BDH, Poole, UK) and Reynolds lead citrate for 10 min (Merck) at room temperature in a light-protected environment. Ultrathin sections were observed on a JEOL 100CXII transmission electron microscope (JEOL, Tokyo, Japan) at 60kV43.

Stereological and statistical analysis

A systematic sampling was carried out on each microscope grid (Fig. S1), with images taken at alternating TEM field areas when the MII oocyte cytoplasm occupied more than 50% of the field. Images were taken at × 5300 magnification and printed at 20.2 cm × 20.2 cm. A classical manual stereological technique based on point-counting with a suitable stereological grid was used. The grid was positioned over printed images, and the number of grid points located over each organelle was registered. The relative volume (Vv) of each organelle was obtained by applying the formula Vv (organelle, oocyte) = [number of dots (organelle)/number of dots (oocyte)] × 100 (%)56.

Organelles included in the present evaluation were cortical vesicles, mitochondria, lysosomes, smooth endoplasmic reticulum (SER) small vesicles, SER medium vesicles, SER large vesicles, SER isolated tubules, small aggregates of SER tubules and total SER (Fig. S2).

Each MII oocyte was divided into three regions, from the oolema (oocyte membrane) up to the cell center: cortex (5 μm), subcortex (5–10 μm) and inner cytoplasm (> 10 μm) (Fig. S3). A stereological procedure was adopted, applying the formula Vv (organelle, cortex/ subcortex/inner cytoplasm) = [number of points (organelle)/number of points (cortex/subcortex/inner cytoplasm)] × 100 (%)37.

Statistical analysis was performed using the Microsoft Excel 2022 [Microsoft Corporation, Redmond, WA, USA (https://www.microsoft.com/en-us/microsoft-365/excel)] and SPSS version 27.0 software [IBM Corp, Foster City, California, CA, USA (https://www.ibm.com/products/spss-statistics)]. The results are presented as mean, standard error of the mean (SEM = standard deviation/n1/2) and coefficient of variation (CVar = standard deviation/mean), determined using Microsoft Excel 2022. Normal distribution was tested with the Kolmogorov–Smirnov test using SPSS version 27.0 software. Considering the samples did not exhibit a normal distribution, non-parametric tests were applied. The means of Vv (organelle, oocyte), Vv (organelle, cortex), Vv (organelle, subcortex) and Vv (organelle, inner cytoplasm) were compared using the Kruskal–Wallis test and the Mann–Whitney U-test with Bonferroni correction in SPSS version 27.0 software. The level of statistical significance was set at P < 0.05.