Vitrification of mature oocytes is routinely adopted in human clinics and research laboratories, even if with suboptimal yields. Among the numerous variables that can still be optimized, the role of cumulus cells (CCs) during vitrification has not fully elucidated. Data in the literature are controversial, evidencing a protecting role of CCs against oocyte cryoinjuries in some species or a barrier effect against the proper penetration of cryoprotectants in others. The extent of cryoinjuries depends, in fact, on factors including size and shape of the cells, membrane permeability, and oocyte quality and sensitivity. Indeed, all these factors may be highly variable depending on the species, developmental stage and origin (for example, in vitro produced or in vivo derived) (Vajta & Kuwayama, Reference Vajta and Kuwayama2006).
In humans, mature oocytes showed significantly higher survival, cleavage and high-quality embryo rates when CCs were retained with the oocytes before freezing (Jin et al., Reference Jin, Song, Xin, Dai, Chen and Sun2012). In mice, mature oocytes vitrified with the surrounding cumulus cells (COCs, cumulus–oocyte complexes) were fertilized in vitro more efficiently than vitrified denuded oocytes (DOs) (Kohaya et al., Reference Kohaya, Fujiwara, Ito and Kashiwazaki2011). Fertilizability preservation was undoubtedly proved by the same group that recently obtained live rats from vitrified COCs fertilized in vitro with cryopreserved sperm (Fujiwara et al., Reference Fujiwara, Kamoshita, Kato, Ito and Kashiwazaki2017). Similarly, vitrified immature mouse oocytes had higher survival, maturation and embryo developmental rates, with respect to the denuded vitrified counterpart (Nikseresht et al., Reference Nikseresht, Toori, Rasti, Kashani and Mahmoudi2015). CCs also seemed to maintain the fertilization ability in vitrified bovine oocytes, as demonstrated by comparing blastocyst formation and cleavage rates from vitrified DOs and COCs. Both were significantly higher in vitrified multilayered COCs that, after IVF, successfully produced calves (Ishii et al., Reference Ishii, Tomita, Sakakibara and Ohkura2018). Conversely, others found lower survival rates in vitrified bovine COCs, with respect to fresh and denuded oocytes, even if cleavage, blastocyst and fertilization rates were higher in vitrified COCs than in DOs (Ortiz-Escribano et al., Reference Ortiz-Escribano, Smits, Piepers, Van den Abbeel, Woelders and Van Soom2016). These might be explained by considering results previously obtained in sheep, in which the complete removal of CCs before vitrification enhanced oocyte survival and meiotic competence but impaired the activity of proteins such as maturation-promoting factor (MPF) and mitogen-activated protein kinase (MAPK) that could affect the developmental competence of oocytes (Bogliolo et al., Reference Bogliolo, Ariu, Fois, Rosati, Zedda, Leoni, Succu, Pau and Ledda2007). Higher fertilization rates found in the presence of CCs can be accounted for by their action in the prevention of zona hardening and maintenance of cortical granule number, as observed in humans using electron microscopy (Tong et al., Reference Tong, Wu, Jin, Luo, Luan and Liu2012).
Transmission electron microscopy (TEM) is the gold standard in revealing the microtopography and the ultrastructural characteristics of the cells and, indeed, is a tool in assessing the presence and extent of alterations connected to vitrification. To our knowledge, clinical outcomes and biomolecular data available are not yet sustained by a morphological evaluation on CCs exposed or not to vitrification, and are necessary to better identify the cryosensistivity of these cells in order to corroborate or disprove the theory of a protecting role of CCs against cryoinjuries. Therefore, to better clarify if cryoprotectants might affect the integrity of CCs in humans, we report in this study our ultrastructural observations made by TEM on discarded patches of CCs isolated from high-quality COCs that were recovered by women undergoing intracytoplasmic sperm injection (ICSI) due to male infertility factor and vitrified/thawed by CryoTop. Fresh CCs were used as controls.
Materials and Methods
All materials were purchased from Sigma Chemical Co. (St. Louis, MO, USA), unless stated otherwise.
Source of CCs
In total, 50 patches of CCs were sampled from mature COCs, collected by ovum pick up from 30 women aged 25–40 years who were undergoing ICSI cycles due to male infertility factor at the Yazd Reproductive Sciences Institute. Among ARTs, the ICSI procedure was specifically selected as a unique source of discarded CCs, as they are routinely removed from high-quality COCs prior to subsequent oocyte injection (Mangoli et al., Reference Mangoli, Khalili, Talebi, Ghasemi-Esmailabad and Hosseini2018). Briefly, gonadotropin-releasing hormone (GnRH) antagonist stimulation protocols were used for all participants (Eftekhar et al., Reference Eftekhar, Aflatoonian, Mohammadian and Eftekhar2013). Transvaginal oocyte aspiration was performed 36 h after the trigger injections. COCs were removed from the collection fluid using a sterile pipette and washed in G-IVF (VitroLife, Sweden). Patches of CCs were then isolated with insulin needles from high-quality COCs and washed twice with G-Mops-V1 (VitroLife). Denuded oocytes were assessed for nuclear maturity under a stereomicroscope (Olympus Co., Japan) and used for patient treatment. Surplus CC patches were randomly divided into two groups: 1, fresh CCs (controls, n = 25); and 2, vitrified CCs (n = 25).
Vitrification and thawing
CCs were frozen utilizing a modified vitrification method (Al-Hasani et al., Reference Al-Hasani, Ozmen, Koutlaki, Schoepper, Diedrich and Schultze-Mosgau2007). The cells were equilibrated in a solution composed of 7.5% ethylene glycol (EG, Merck Co., Germany), 7.5% dimethyl sulphoxide (DMSO, Merck Co, Germany) in Ham's F10 medium supplemented with 20% human serum albumin (HSA, Plasbumin Co., USA) for 10 min at room temperature (RT). CCs were then transferred to a vitrification solution containing 15% EG, 15% DMSO and 0.5 M sucrose (Sigma Co., USA) in Ham's F10 medium supplemented with 20% HSA for 1 min at RT. The cells were then loaded on a CryoTop in a volume < 1 μl, and immediately immersed into liquid nitrogen (LN2) for storage. For thawing, the cryotops were directly submerged into pre-incubated Ham's F10, 20% HSA and 1 M sucrose solution for 1 min the next day. Finally, the thawed CCs were rehydrated in Ham's F10 and 20% HSA (v/v) containing 0.5 or 0.25 M sucrose at RT for 3 and 5 min, respectively. The samples were finally rinsed in Ham's F10 and 20% HSA three times.
Transmission electron microscopy
From each group, 15 CC samples were fixed and processed for TEM as previously reported (Bianchi et al., Reference Bianchi, Macchiarelli, Borini, Lappi, Cecconi, Miglietta, Familiari and Nottola2014; Familiari et al., Reference Familiari, Verlengia, Nottola, Tripodi, Hyttel, Macchiarelli and Motta1998; Macchiarelli et al., Reference Macchiarelli, Palmerini, Nottola, Cecconi, Tanemura and Sato2013; Nottola et al., Reference Nottola, Cecconi, Bianchi, Motta, Rossi, Continenza and Macchiarelli2006, Reference Nottola, Familiari, Micara, Aragona and Motta2016; Palmerini et al., Reference Palmerini, Antinori, Maione, Cerusico, Versaci, Nottola, Macchiarelli, Khalili and Antinori2014a, Reference Palmerini, Nottola, Leoni, Succu, Borshi, Berlinguer, Naitana, Bekmukhambetov and Macchiarelli2014b; Reference Palmerini, Zhurabekova, Balmagambetova, Nottola, Miglietta, Belli, Bianchi, Cecconi, Di Nisio, Familiari and Macchiarelli2017; Reference Palmerini, Belli, Nottola, Miglietta, Bianchi, Bernardi, Antonouli, Cecconi, Familiari and Macchiarelli2018). Briefly, CCs were fixed for 2–5 days in 2.5% glutaraldehyde (Agar Scientific, Cambridge Road Stansted Essex, UK)/PBS at 4°C, embedded in 1% agar, then post-fixed with 1% osmium tetroxide (SIC, Rome, Italy). The samples were dehydrated in increasing concentrations of ethanol and then immersed in propylene oxide for solvent substitution. CC patches were individually embedded in EMbed-812 (Electron Microscopy Sciences, 1560 Industry Road, Hatfield, PA, USA). In the next step, the CCs were sectioned for light and electron microscopy observations. The cells were sectioned using a Reichert-Jung Ultracut E ultramicrotome (Reichert Technologies, Munich, Germany) at a thickness of 0.5–1 μm for light microscopy and stained with toluidine blue. Ultrathin sections of 60–80 nm were cut with a diamond knife, mounted on copper grids, and contrasted with saturated uranyl acetate and lead citrate (SIC, Rome, Italy) before being examined and photographed using Zeiss EM10 and Zeiss EM900 electron microscopes (Germany) operating at 80 kV.
The following parameters were evaluated by TEM and used in the qualitative assessment of CCs: general features; cell membrane integrity; characteristics of the nucleus, chromatin, and nuclear envelope; type and quality of organelles and inclusions; and presence and extent of cytoplasmic vacuolization (Nottola et al., Reference Nottola, Cecconi, Bianchi, Motta, Rossi, Continenza and Macchiarelli2006; Reference Nottola, Familiari, Micara, Aragona and Motta2016).
ImageJ software (http://rsbweb.nih.gov/ij/) was used to measure cell dimension on low magnification TEM micrographs of control and vitrified CCs. For each experimental group, at least 15 CCs were selected for morphometric analysis.
All data were expressed as means ± standard deviation (SD). Statistical comparisons were performed using an unpaired t-test (GraphPad InStat). Differences in values were considered significant if P-values were <0.05.
Group 1 (fresh CCs)
Fresh CCs [14.43 µm ± 0.874 µm, medium diameter ± standard deviation (SD)] showed a normal ultrastructure. Most of the CCs were round or oval and delimited by a continuous cytoplasmic membrane, occasionally protruding in short microvilli. Clustered cells showed small linear gap junctions (Fig. 1a). Large oval nuclei, with one or more nucleoli, were delimited by a continuous and electron-dense nuclear membrane (Fig. 1a). Spots of heterochromatin were clustered beneath the nuclear membrane (Fig. 1a). The organelles were abundant and uniformly scattered in the cytoplasm. Major organelles were those typical of steroidogenic cells as mitochondria, smooth endoplasmic reticulum (SER), Golgi apparatus and lipid droplets (Fig. 1a). Mitochondria appeared mainly ovoid or elongated, with numerous tightly packed tubulo-vesicular cristae (Fig. 1b). Tubular and vesicular SER elements were abundant and uniformly distributed throughout the cytoplasm (Fig. 1c). Cisternae and associated vesicles belonging to the Golgi apparatus were visible in the cytoplasm of fresh CCs. Numerous, spherical dark lipid droplets occupied the cytoplasm of fresh CCs (Fig. 1c). Free ribosomes, occasional lysosomes and electron-lucent vacuoles, sometimes containing debris, were also present (Fig. 1c).
Group 2 (vitrified CCs)
By TEM, vitrified CCs appeared mainly spherical (12.29 µm ± 0.701 µm, medium diameter ± SD; Group 2 versus Group 1: P > 0.05), isolated and delimited by a continuous cytoplasmic membrane (Fig. 2a). We observed two populations of CCs: light CCs had a smooth appearance, with a few short microvilli; in dark CCs the plasma membrane folded into numerous and long microvilli (Fig. 2a–c). The ratio nucleus/cytoplasm was occasionally higher respect what observed in fresh CCs (Fig. 2a, c). The nucleoplasm contained dispersed chromatin with one or more nucleoli, except for peripheral patches of heterochromatin, found attached to the inner leaflet of the nuclear membrane (Fig. 2a). The cytoplasm appeared populated with different types of organelles, with some differences with respect to group 1 (Fig. 2b). Organelles as mitochondria, tubular and vesicular elements of the SER and Golgi apparatus and ribosomes did not show significant variations with respect to Group 1. Lipid droplets were more numerous and larger than in fresh CCs in the 33.3% of vitrified patches and were packed around the peripheral cytoplasm or located in proximity of the nucleus, often close to elements of the endoplasmic reticulum (Fig. 2a–c).
Compared with group 1, in the 20% of vitrified patches the cytoplasm presented numerous large and clear vacuoles (Fig. 2a, c). In 40% of the observed 251 patches a concomitant increase in vacuoles and lipid droplets was found, while 6.7% of samples were mostly similar to the fresh controls.
CCs are known to be nourishing cells providing the supply of nutrients and signalling molecules through gap junctions and paracrine factors, necessary to sustain oocyte maturation. CCs also produce adhesive proteins useful to favour oocyte pick up and oviductal transport (Familiari et al., Reference Familiari, Verlengia, Nottola, Renda, Micara, Aragona, Zardi and Motta1996). In addition, they actively participate in the spermatozoa selection for a successful fertilization (Nottola et al., Reference Nottola, Albani, Coticchio, Palmerini, Lorenzo, Scaravelli, Borini, Levi-Setti and Macchiarelli1998; Huang & Wells, Reference Huang and Wells2010; Coticchio et al., Reference Coticchio, Dal Canto, Mignini Renzini, Guglielmo, Brambillasca, Turchi, Novara and Fadini2015; Zhou et al., Reference Zhou, Wu, Shen, Wang, Kong, Lu, Li, Zhou, Zhao and Liang2016; references therein). CCs may even have a role in positively affecting the microenvironment in which fertilization and early embryo development occur, both in vivo and in vitro, by producing small amounts of estrogens and progesterone, as revealed by their ultrastructure (Nottola et al., Reference Nottola, Heyn, Camboni, Correr and Macchiarelli1991; Motta et al., Reference Motta, Nottola, Pereda, Croxatto and Familiari1995).
In clinical practice, CCs are usually mechanically or enzymatically removed from COCs, to ascertain the meiotic stage of the oocyte before subsequent ARTs. When CCs are retained with the oocyte to ensure full developmental competence, there is the possibility that cryopreservation of COCs may compromise the functional coupling between oocyte and CCs, as seen in immature feline COCs (Luciano et al., Reference Luciano, Chigioni, Lodde, Franciosi, Luvoni and Modina2009) and in immature and mature porcine COCs (Wu et al., Reference Wu, Rui, Dai, Zhang, Ju, Xie, Lu and Zheng2006). Noteworthily, in equine immature COCs, vitrification induced the death of CCs invariably at the periphery of the cumulus layers, without compromising gap junction function or the ability of the oocytes to complete nuclear maturation (Tharasanit et al., Reference Tharasanit, Colleoni, Galli, Colenbrander and Stout2009). In humans, data present in the literature are controversial, with a group reporting that the denudation determined a significantly higher incidence of post-thaw meiotic spindle visualization and survival rates in DOs, when compared with COCs (Minasi et al., Reference Minasi, Fabozzi, Casciani, Ferrero, Litwicka and Greco2012); another showing higher survival, cleavage and high-quality embryo rates when CCs were retained with the oocytes before freezing (Jin et al., Reference Jin, Song, Xin, Dai, Chen and Sun2012) and a third that did not evidence any ultrastructural damage or difference in the clinical outcomes between fresh controls, vitrified COCs and vitrified DOs, except for a higher fertilizing capability in vitrified COCs subjected to conventional IVF (Tong et al., Reference Tong, Wu, Jin, Luo, Luan and Liu2012). Overall, as we previously reported (Khalili et al., Reference Khalili, Shahedi, Ashourzadeh, Nottola, Macchiarelli and Palmerini2017), the most cryo-sensitive structures of the human oocytes are microvilli, mitochondria, vacuoles and cortical granules with obvious specific differences between immature and mature oocytes, due to the presence of the nucleus and to the absence of the meiotic spindle in the formers.
To further clarify the role of CCs during vitrification in human, we aimed to analyze the ultrastructure of human CCs discarded from high-quality COCs undergoing ICSI, due to male infertility factor.
After vitrification, some CC cells showing a slightly dark cytoplasm and an irregular profile were found, with the plasma membrane folded into numerous and long microvilli. These cells resemble those belonging to a peculiar CC cell sub-population that, in vivo, appears specifically involved in sperm selection through phagocytosis (Pereda & Coppo, Reference Pereda and Coppo1984; Motta et al., Reference Motta, Nottola, Pereda, Croxatto and Familiari1995; Nottola et al., Reference Nottola, Albani, Coticchio, Palmerini, Lorenzo, Scaravelli, Borini, Levi-Setti and Macchiarelli1998). Probably, the vitrification process induces a sort of ‘hyperactivation’ in some CC cells, with the expression of ultrastructural markers such as membrane irregularities and evaginations. Nuclear and cytoplasmic characteristics of these CC cells, however, are similar to those shown by the other CC cells and, at least in part, superimposable to those of fresh CCs, as reported below.
Intercellular connections were identified only in occasional clusters of fresh CCs, in which small linear gap junctions were observed. The absence of evident cell-to-cell interactions in vitrified CCs seems in contrast with the features shown by fresh or vitrified ovarian tissues, in which neighbouring cells maintained the established contacts after cryopreservation (Keros et al., Reference Keros, Xella, Hultenby, Pettersson, Sheikhi, Volpe, Hreinsson and Hovatta2009; Nottola et al., Reference Nottola, Macchiarelli, Familiari, Stallone, Sathananthan and Motta2011; Sheikhi et al., Reference Sheikhi, Hultenby, Niklasson, Lundqvist and Hovatta2011). This discrepancy can be surely due to the use of expanded CCs that, after vitrification, thawing and the subsequent preparative for TEM observations are clearly more subjected to isolation.
After vitrification, even if the ratio of nucleus/ cytoplasm appeared higher with respect to what was observed in fresh samples, CCs showed well preserved nuclei, as demonstrated by the presence of a continuous nuclear membrane, with patches of condensed chromatin attached on its inner leaflet. In our study, these features are comparable with those shown by the fresh counterpart. Similar characteristics are also observable in fresh COCs and vitrified human ovarian tissues containing metabolically active cells (Nottola et al., Reference Nottola, Albani, Coticchio, Palmerini, Lorenzo, Scaravelli, Borini, Levi-Setti and Macchiarelli1998; Mazoochi et al., Reference Mazoochi, Salehnia, Valojerdi and Mowla2008; Keros et al., Reference Keros, Xella, Hultenby, Pettersson, Sheikhi, Volpe, Hreinsson and Hovatta2009; Sheikhi et al., Reference Sheikhi, Hultenby, Niklasson, Lundqvist and Hovatta2011).
Morphology of organelles in vitrified–thawed samples as mitochondria, SER elements, Golgi apparatus and ribosomes did not differ with respect to controls, in agreement with previous data from cryopreserved ovarian tissue (Salehnia et al., Reference Salehnia, Abbasian Moghadam and Rezazadeh Velojerdi2002; Sheikhi et al., Reference Sheikhi, Hultenby, Niklasson, Lundqvist and Hovatta2011 and Keros et al., Reference Keros, Xella, Hultenby, Pettersson, Sheikhi, Volpe, Hreinsson and Hovatta2009). Pleomorphic mitochondria, of different size and shape and containing tubular or vesicular cristae, were a typical feature of luteinizing CCs, concomitantly with a rich SER network and a well developed Golgi apparatus (Motta et al., Reference Motta, Nottola, Familiari, Makabe, Stallone and Macchiarelli2003).
Major ultrastructural differences between fresh and vitrified CCs involved lipid droplets and vacuoles. In our study, the 33.3% of vitrified patches of CCs showed an increased amount of electron-dense lipid droplets in the cytoplasm, centrally or eccentrically packed, in respect to those found in fresh CC. Accumulation of lipids, with the help of specific cytoskeletal activity allowing their compartmentalization, is one of the early signs of luteinization (Motta et al., Reference Motta, Nottola, Pereda, Croxatto and Familiari1995). However, physical change of lipids at freezing temperature is one of the major causes of cellular cryodamage (Zhou & Li, Reference Zhou and Li2013) and, therefore, our data confirmed the cryosensitivity of CCs against freezing, probably due to the used cryoprotectants (Fu et al., Reference Fu, Shi, Zhang, Zhao, Yan, Hou, Zhou, Fan, Suo, Wusiman, Wang and Zhu2009; Tayefi Nasrabadi et al., Reference Tayefi Nasrabadi, Gavami, Akbarzadeh, Beheshti, Mohammadnejad and Abedelahi2015).
The sensitivity of CCs to vitrification was further demonstrated by an increased vacuolization found in the 20% of the vitrified patches and in the 40% of patches with the concomitant presence of increased lipid droplets. This finding is in agreement with other studies on human and mouse tissues (Salehnia et al., Reference Salehnia, Abbasian Moghadam and Rezazadeh Velojerdi2002; Keros et al., Reference Keros, Xella, Hultenby, Pettersson, Sheikhi, Volpe, Hreinsson and Hovatta2009) but in contrast with one study on vitrified human ovarian tissue, in which granulosa cells showed uniform contact with no obvious increase in the level of vacuolization (Sheikhi et al., Reference Sheikhi, Hultenby, Niklasson, Lundqvist and Hovatta2011). These data could be a probable consequence of the step-wise vitrification method used in this former study. It should be noted that the appearance of vacuoles is considered a non-specific response to cryodamage or osmotic stress (Khalili et al., Reference Khalili, Maione, Palmerini, Bianchi, Macchiarelli and Nottola2012) that may also occur in human oocytes cryopreserved using different protocols of slow freezing or vitrification (Palmerini et al., Reference Palmerini, Antinori, Maione, Cerusico, Versaci, Nottola, Macchiarelli, Khalili and Antinori2014a; Nottola et al., Reference Nottola, Familiari, Micara, Aragona and Motta2016; Khalili et al., Reference Khalili, Maione, Palmerini, Bianchi, Macchiarelli and Nottola2012; Reference Khalili, Shahedi, Ashourzadeh, Nottola, Macchiarelli and Palmerini2017).
In conclusion, the data obtained in the present study highlighted the effect of vitrification on the ultrastructure of CCs collected from mature COCs taken from women who were undergoing ICSI.
We found that vitrification can preserve the fine structure of major organelles of human CCs. Nevertheless, an increase in lipid droplets and vacuoles was observed, thus confirming the cryosensitivity of these cells.
This study revealed that CC cell patches may be considered a simple and effective biological model to test cell damage during cryopreservation. In addition, based on our results, we can speculate that the maintenance of CCs around the oocyte during vitrification could decrease the extent of cryodamage to the oocyte, thus contributing to the preservation of oocyte competence for fertilization. Retained CCs could in fact create a sort of natural, protective shield around the oocyte against physico-chemical insults due to the cryopreservation procedure. Further studies are needed on vitrified COCs, if available, to have a complete, parallel view of both CC and oocyte ultrastructural morphology with the final aim of contributing to the optimization of the current vitrification protocols.
This work was supported by the Department of Life, Health and Environmental Sciences, University of L'Aquila, Italy (RIA 2015–2018) and the Department of Anatomy, Histology, Forensic Medicine and Orthopaedics, La Sapienza University of Rome, Italy (University grants 2015–2017).
Conflicts of interest
The Ethics Committee of Yazd Reproductive Sciences Institute, Shahid Sadoughi University of Medical Sciences, Yazd, Iran, approved this study (ethical code: IR. SSU. Medicine Rec. 1392.300). All experiments involving human participants were performed in accordance with the 1964 Helsinki declaration and relevant guidelines. Written informed consents were obtained from all patients who participated in our research study.