Hum. Reprod. Advance Access originally published online on April 11, 2007
Human Reproduction 2007 22(7):1959-1972; doi:10.1093/humrep/dem083
Osmotic responses and tolerance limits to changes in external osmolalities, and oolemma permeability characteristics, of human in vitro matured MII oocytes
1 Centre for Reproductive Medicine, Academic Hospital, Dutch-speaking Brussels Free University, Laarbeeklaan 101, 1090 Brussels, Belgium 2 Research Centre Reproduction and Genetics, Dutch-speaking Brussels Free University, Laarbeeklaan 103, 1090 Brussels, Belgium 3 German Institute for Reproductive Medicine, Hannoversche Strasse 24, 31848 Bad Muender, Germany 4 Comparative Medicine Centre, University of Missouri, Columbia, MO 65211, USA
5 Correspondence address. E-mail: Etienne.vandenabbeel{at}uzbrussel.be
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionConclusionsAcknowledgementsReferences |
|---|
BACKGROUND: Oocyte cryopreservation remains a realistic objective, provided that more systematic approaches are applied, such as thorough analysis of the oocyte oolemma permeability to water and diverse cryoprotectants.
METHODS: We prospectively investigated volume changes over time at different temperatures (30°C, 22°C and 8°C) of human metaphase II (MII) oocytes (obtained in stimulated ICSI cycles and matured in vitro from the germinal vesicle stage) when exposed to changes in external osmolality. We also investigated human in vitro matured (IVM) oocytes membrane permeability characteristics at 22°C to 1,2-propanediol (PG) and dimethylsulphoxide (DMSO) and at 30°C, 22°C and 8°C to ethylene glycol (EG), and calculated corresponding oocyte oolemma permeability coefficients (Lp and Pcpa). Furthermore, we investigated the osmotic tolerance limits of IVM oocytes exposed to changes in external osmolality as assessed by their developmental competence during the course of 72 h after ICSI.
RESULTS: The results of our studies describe human oocyte membrane permeability coefficients for EG at 30°C (2.85 ± 0.15 x 10–3 cm/min), 22°C (1.17 ± 0.60 x 10–3 cm/min) and 8°C (0.37 ± 0.15 x 10–3 cm/min). Furthermore, at 22°C the EG oolemma permeability coefficient was lower than that of PG and DMSO (1.17 ± 0.60 x 10–3 cm/min versus 2.15 ± 0.70 x 10–3 and 1.56 ± 0.38 x 10–3 cm/min, respectively). Our results also indicate, that human IVM MII oocytes tolerated exposure to solutions in the range of 39–2264 mOsmol/kg H2O as assessed by the oocytes' developmental competence after exposure.
CONCLUSIONS: The results of the present study may contribute to a better understanding of the biology and cryobiology of human oocytes, and to the design of better and more robust cryopreservation (freezing or vitrification) protocols.
Key words: Cryopreservation/Human/Oocyte/Cryoprotectants/Permeability
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionsAcknowledgementsReferences |
|---|
The cryopreservation of human oocytes is highly desirable to preserve a woman's fertility, however, oocyte freezing has only recently become a successful reality (Jain and Paulson, 2006
; Oktay et al., 2006).
Mammalian oocytes are generally more susceptible to injury during a cryopreservation procedure than embryos, and human oocytes are no exception. Among the factors responsible for this sensitivity is the presence of the metaphase II (MII) spindle, which can be affected by osmotic stress and chilling (Pickering et al., 1990; Zenzes et al., 2001; Mullen et al., 2004a,b). Furthermore, the human oocyte's large volume-to-surface ratio increases the probability of intracellular ice formation during freezing, which is lethal to most cell types (Leibo et al., 1978). The above-mentioned properties have collective detrimental effects on cryopreserved oocytes (Smith and Silva, 2004). Since the major concerns about spindle damage in oocyte freezing are all linked to the use of mature oocytes (MII), several attempts were made to freeze immature oocytes [germinal vesicle (GV)], in which the meiotic spindle is not yet formed (Ambrosini et al., 2006). Tucker et al. (1998) obtained the first birth of a healthy child from a cryopreserved GV oocyte. However, GV oocyte cryopreservation was not advantageous, since the method has been related to an increased incidence of chromosomal abnormalities despite the absence of the MII spindle (Park et al., 1997; Boiso et al., 2002). The reasons for this are unknown and therefore, most research on human oocyte cryopreservation has been focused on the MII oocytes.
By 2000, however, not more than about 20 children had been born from frozen-thawed oocytes (Van der Elst, 2003). Consequently, with the aim of improving the overall survival of oocytes post-cryopreservation, various modifications of this protocol have been tested (Trad et al., 1999; Fabbri et al., 2001; Boldt et al., 2003, 2006; Chen et al., 2005; Borini et al., 2006; Stachecki et al., 2006). However, the basic principle for slow cooling, remains the same: oocytes are suspended in a solution containing cryoprotectants (CPAs), cooled and then seeded to induce extracellular ice crystal formation promoting dehydration of the oocyte as the solution is cooled slowly at 0.3°C/min to –30°C or below. Finally, the oocyte is plunged into liquid nitrogen (LN2) for storage.
As a potential alternative to slow cooling methods, vitrification of human oocytes has also been successfully investigated (Kuleshova et al., 1999; Yoon et al., 2003; Kuwayama et al., 2005; Lucena et al., 2006). Since chilling injury to oocytes is time-dependent, the rationale is to prevent ice formation and injury by freezing at a rate fast enough to solidify the intracellular water before it can crystallize (Martino et al., 1996). However, there are several challenges when applying vitrification techniques, especially when applied to human oocytes. Important concerns with this technique are the increased toxicity of high levels of CPAs at room or higher temperatures, and the ability to freeze and warm fast enough to avoid crystal formation and devitrification (Pegg, 2005; Vajta and Kuwayama, 2006). Oktay et al. (2006) clearly demonstrated in a meta-analysis that when compared with slow cooling methods, vitrification of oocytes resulted in somewhat better outcomes however, the overall efficiency of human oocyte cryopreservation (children born/oocytes cryopreserved and thawed) remains deceptively low. Kuwayama et al. (2005) described a highly efficient ‘cryotop’ vitrification method using very fast cooling and warming rates. This method requires lower concentrations of CPA to achieve vitrification and prevent devitrification compared to that required when using other cryopreservation systems. The method as described by Kuwayama et al. (2005) is said to be a non-equilibrium vitrification method and as a result the procedure is not very robust (Pegg, 2005). Furthermore, the ‘open cryo-top system’ allows direct contact of the samples with LN2; consequently, there is a risk of transmission of disease through LN2.
It is essential to develop an efficient, robust as well as safe equilibrium vitrification procedure for human oocytes. In order to develop an effective cryopreservation protocol, it has been argued that the determination of fundamental cryobiological characteristics of each cell type is essential (Critser et al., 1997). Coticchio et al. (2004) stated that oocyte cryopreservation remains a realistic objective, provided that more systematic approaches are applied, such as a thorough analysis of the oocyte oolemma permeability to water and the diverse CPAs. Furthermore, Paynter et al. (2005a) indicated that protocols should be designed to keep the oocyte's cell volume within tolerated limits or to achieve set levels of dehydration prior to freezing Therefore, we must extend our knowledge on the fundamental cryobiology of human oocytes. With the knowledge of both a cell's permeability to a CPA and the osmotic tolerance, an optimal method by which CPAs can be loaded into and unloaded from cells without exceeding the osmotic tolerance can be predicted using theoretical modelling (Gao et al., 1995; Agca et al., 1998b). Several authors have calculated permeability coefficients of human oocytes to water and the CPAs dimethylsulphoxide (DMSO) and propylene glycol (PG) (Fuller et al., 1992; Hunter et al., 1992a,b; McGrath et al., 1995; Bernard and Fuller, 1996; Newton et al., 1999; Paynter et al., 1999a; Parisi et al., 2000; Paynter et al., 2001). The CPA ethylene glycol (EG) is currently the most commonly used permeating CPA in vitrification of human oocytes, embryos and blastocysts (Liebermann et al., 2003; Yoon et al., 2003; Kuwayama et al., 2005). The permeability of MII human oocytes to EG has yet to be investigated and a prospective comparison of osmotic responses and permeability characteristics of human oocytes exposed to the CPAs DMSO, PG and EG, has so far not been performed. Furthermore, the osmotic tolerance limits of human MII oocytes as assessed by their further developmental potential have not been determined.
In most countries, the supply of human oocytes for research purposes is a major problem. However, in stimulated ICSI cycles it is well known that 10% of the collected oocytes are at the immature GV stage of development (Van Steirteghem et al., 1998). These oocytes are never used for the patients' infertility treatment and maturing these oocytes in vitro provides an important source of MII oocytes for research.
The current paper describes a series of studies to determine biophysical characteristics of IVM human oocytes obtained in stimulated ICSI cycles. We investigated the osmotic responses of MII oocytes when exposed to changes in external osmolality. We also investigated prospectively the permeability of human oocytes to DMSO, PG and EG and determined the permeability of IVM human oocytes to EG at different temperatures. Furthermore, we investigated the osmotic tolerance limits of human MII oocytes as assessed by their further developmental potential. Using the information obtained from this research, we should be able to improve slow cooling and vitrification technology for human MII oocytes.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionsAcknowledgementsReferences |
|---|
Source and collection of oocytes
The study was approved by the Commission of Medical Ethics of the Academic Hospital of the Dutch-speaking Brussels Free University (AZ-VUB). GV stage oocytes were obtained from patients, with informed consent, undergoing ICSI treatment. Female patients (mean age 33.8 years) underwent ovarian stimulation using urinary or recombinant FSH in combination with GnRH antagonist or agonist. Oocyte retrieval was carried out 36 h after hCG injection by vaginal ultrasound-guided puncture of ovarian follicles (Kolibianakis et al., 2004
). Cumulus–oocyte complexes (COCs) were collected in HEPES-buffered HTF medium (Cambrex, Belgium) supplemented with 10% synthetic serum substitute (SSS) (Irvine scientific catalogue no. 99193, Orange Medical, The Netherlands), further referred to as HEPES medium. COCs were denuded from their surrounding cumulus cells (Van Landuyt et al., 2005). Naked oocytes were checked for maturity. GV stage oocytes were used for this study and their quality was determined under the stereomicroscope (x40 magnification) or inverted microscope (x400 magnification). Typical GV oocytes without vacuoles (Veeck and Zaninovic, 2003) were transferred to a culture dish (Falcon) and cultured individually in 25 µl droplets of maturation medium (MM) under mineral oil.
In vitro maturation of oocytes, ICSI procedure and further embryo development
Media for IVM were prepared the day before oocyte retrieval for overnight equilibration. The MM consisted of a modification of the medium as described by Cekleniak et al. (2001). In-house made KSOMAA medium (Biggers et al., 2000; Biggers and Racowsky, 2002), using Sigma chemicals, was supplemented with SSS; recombinant FSH, 0.075 IU/ml (Gonal F; Serono Laboratories); estradiol, 1 µg/ml; hCG, 0.5 IU/ml (Serono Laboratories); taurine, 6.25 mg/ml; epidermal growth factor, 10 ng/ml. After approximately 20–24 h of culture, the oocytes were inspected for the presence of a first polar body. Only those oocytes with a clearly visible, morphologically normal, not fragmented, polar body, and with normal cytoplasmic aspects (clear cytoplasm and no vacuoles), were used in the studies. Furthermore, the oocytes had to be evenly sized and have diameters within defined ranges of 110 and 125 µm.
To study the developmental competence of IVM human oocytes, oocytes were microinjected with sperm cells. The ICSI procedure was done as described by Van Steirteghem et al. (1995). Sperm cells for ICSI were from several male patients with excellent sperm characteristics (=number of sperm cells, motility and morphology) and who consented to donate supernumerary sperm cells for scientific research. Sperm preparation for ICSI was done as described by Van Landuyt et al. (2005). After the ICSI procedure, oocytes were cultured individually in 25 µl KSOMAA medium supplemented with SSS under mineral oil for 72 h. Eighteen hours after injection, oocytes were inspected for two pronucleus formation (2PN), and after 72 h the developmental stage and the quality of the embryos obtained was evaluated. The developmental competence was studied during the course of 72 h only because in a preliminary experiment it was found that less than 5% of human IVM MII oocytes develop to the blastocyst stage. Embryos of good quality were embryos with less than 20% of their volume filled with anucleate fragments, with no multinucleation or vacuoles present in single blastomeres and with normal blastomere size and a clear cytoplasm.
Microperfusion technique, osmotic response data collection and analysis, and calculation of permeability coefficients
A micropipette perfusion technique was conducted according to Gao et al. (1994). The micropipettes were made from 100 mm borosilicate glass capillary tubes (Drummond Scientific Company, Broomhall, PA, USA), using a horizontal microelectrode puller (Model P-97, Sutter Instrument Co., Novato, CA, USA) and a microforge (Model MF-9, Narishige Co. Ltd., Tokyo, Japan). The holding pipettes had a 20 µm diameter tip opening and an angle of about 25°. We used two holding pipettes filled with mineral oil to apply negative pressure to the zona pellucida. This allowed the oocyte to be held in place during the experimental treatment.
The culture dish (Falcon, type 1006, Becton Dickinson, Puurs, Belgium) with the oocyte was placed on the stage of an inverted microscope (Diaphot, Nikon Europe BV, Badhoevedorp, The Netherlands) equipped with a micromanipulator (IM6, Narishige). The temperature of the stage was maintained at 30°C, 22°C or at 8°C by a temperature controller (Type PE 60 Peltier, Linkam Scientific Instruments Ltd., Tadworth, Surrey, UK). We observed the oocytes with a DIC Achromat objective (High N.A.: LWD-40X C) (Nikon Europe BV). A Sony 3CCD camera (Nikon Europe BV) recorded the real-time images. The images were digitized and stored in the computer using the EZ 2000 software programme (Nikon Europe BV) developed for image acquisition and analysis. Experimental data were obtained by measuring the change in the oocyte area during static or kinetic experiments. The area of each analysed image was calculated with the software EZ 2000. The area was defined by means of related points around the oolemma contour. The value of the cell volume Vi at a given time point was calculated from the area (Ai) for each individual oocyte. The normalized cell volume was calculated as Vi/Viso where Viso is the iso-osmotic volume (302 mOsmol/kg H2O) immediately prior to perfusion. For calculation of membrane permeability coefficients and activation energies, data on cell volume changes in relation to time were entered into the parameter estimation portion of computer software (MLAB, Civilized Software Inc., http://www.civilized.com, Bethesda, USA). The Kedem and Katchalsky formalism was used as the theoretical model of the permeability of cell membranes to water and non-electrolytes (Kedem and Katchalsky, 1958) as described in detail by Agca et al. (1998a,b, 1999, 2000a).
Experiments
Experiment 1: morphology of oocyte shrinkage
The aim of this experiment was to study the morphology of IVM oocytes when exposed for 10 min at 22°C to a phosphate-buffered saline (PBS) solution supplemented with SSS and 1 M sucrose (PBS–sucrose). Before PBS–sucrose treatment, IVM MII oocytes were either treated with PBS also containing EGTA (Sigma) (1 mM for 15 min) (Younis et al., 1996; Albertini, 2004), or vortex-treated (Vortex Labinco L 46, max speed) in an Eppendorf vial for 1 min in 200 µl of PBS also containing 0.2 M sucrose (Sigma) (Agca et al., 2000b; Albertini, 2004), or treated with PBS also containing 10 IU/ml pronase (Protease, Sigma). Oocytes without any pre treatment were also exposed to the hyper-osmotic PBS–sucrose solution and served as controls. After a 10 min incubation in the hyper-osmotic solution, the morphology of the oocytes was determined and classified as spherical, when oocytes uniformly contracted from the zona pellucida, or as irregular when the oocytes remained in contact with the zona pellucida (at one spot or at several spots) (Younis et al., 1996), or as degenerated. After determination of the oocytes morphology, they were incubated in iso-osmotic PBS to evaluate their capacity to regain normal volume.
In a second experiment, we evaluated the developmental competence of vortex-treated IVM matured oocytes. One hundred and twenty-four MII oocytes were randomly allocated to two different groups. One group consisted of 62 MII oocytes that were vortex-treated as described above. After vortex treatment, oocytes were returned to PBS. A second group consisted of 62 oocytes without any treatment. In both groups, oocytes were prepared for ICSI. Injected oocytes were cultured in vitro in KSOMAA medium 72 h. We evaluated the percentage of 2PN formation, the percentage of good-quality embryos obtained and their developmental stage (=number of cells).
Experiment 2: osmotic responses of oocytes in solutions containing the non-penetrating solute galactose
In experiment 2a, osmotic inactive cell volumes (Vb) were determined by exposing the human oocytes to various galactose (Sigma) solutions (0.3, 0.9 and 1.6 M galactose made up in PBS). The experiments were done according to methods previously described by Songsasen et al. (2002) with some modifications. MII oocytes were sequentially exposed in 80 µl droplets under mineral oil to 0, 0.3, 0.9, 1.6 M solutions of galactose prepared in PBS medium for 5 min in each solution at 22°C and photographed using a SONY 3CCD camera attached to an inverted microscope. Solutions were prepared for the entire duration of the experiments and stored in aliquots at –20°C. The osmolality of each solution was measured immediately after preparation and before the initiation of an experiment using a calibrated vapour pressure osmometer and the values were found to be 302 (PBS), 598 (PBS and 0.3 M galactose), 1262 (PBS and 0.9 M galactose) and 2264 (PBS and 1.6 M galactose) mOsmol/kg H2O. Oocytes that remained circular in cross-section during exposure were used for the calculation of the osmotic inactive volume. Oocyte images were recorded and analysed as described above.
The equilibrium normalized cell volumes (100x) were plotted against the reciprocal of the extracellular osmolality of the solution, which is described by the Boyle van't Hoff relationship. The data were then fitted to a straight line by the least-squares method and the osmotic inactive volume fraction of the oocytes (Vb) was determined by extrapolating the best-fit line to an infinitely concentrated solution (Toner et al., 1990).
In experiment 2b, immobilized oocytes in a 5 µl iso-osmotic PBS solution were perfused at different temperatures (30°C, 22°C or 8°C) by adding 1 ml of a pre-equilibrated PBS solution also containing 0.9 M galactose (osmolality = 1262 mOsmol/kg H20). Pre-equilibration was either in a 30°C or 22°C warm water bath or in a water bath cooled to 8°C.
A randomization plan (www.randomization.com) was generated and used to allocate 15 oocytes that maintain spherical shape during the perfusion, to the three different temperatures of exposure. The data points used for analysis were obtained from: (i) image 0 until image 120, one image every 2 s; (ii) image 150 until image 390, one image every 30 s. Data points were put in graphs (normalized cell volume/time). Hydraulic conductivities at different temperatures were calculated.
Experiment 3: tolerance limits of oocytes to changes in external osmolality
Different solutions were prepared to test the tolerance limits of oocytes to changes in external osmolality. Hypo-osmotic solutions were prepared by diluting the PBS solution with high purity grade MilliQ/Milli RO water. A 2x, 4x and 8x dilution of PBS was prepared. The concentrations of all salts in PBS other than NaCl are so low that one can assume with insignificant error that their molality and osmolality decrease in a linear fashion with volume dilution (Mazur and Schneider, 1986). Hyper-osmotic solutions were prepared by supplementing PBS with 0.3, 0.9 or 1.6 M galactose. The osmotic values of hypo- and hyper-osmotic solutions were measured using a calibrated vapour pressure osmometer and values were 39, 76 and 151 mOsmol/kg H2O for 8x, 4x and 2x diluted PBS, respectively, and 302, 598, 1262 and 2264 mOsmol/kg H2O for PBS supplemented with 0, 0.3, 0.9 and 1.6 M galactose. Solutions were prepared for the entire duration of the experiments and stored in aliquots at –20°C. The osmolality of each solution was measured immediately after preparation and before initiation of an experiment.
IVM MII oocytes were exposed to seven different treatments for 5 min at 37°C. The seven different treatments consisted of 8x, 4x and 2x diluted PBS and PBS supplemented with 0, 0.3, 0.9 or 1.6 M galactose. A randomization plan was generated (a block randomization of seven treatments and seven blocks) and used to allocate 343 oocytes to the different treatments.
Following exposure, oocytes were put in iso-osmotic PBS solution for 5 min and then cultured individually in 25 µl KSOMAA medium supplemented with SSS under mineral oil for 2–4 h. The oocytes were then prepared for ICSI and their further developmental competence was assessed as described above. The 2PN formation and percentage of good-quality embryos obtained after 72 h of culture and their developmental stage (=number of cells) was recorded.
Experiment 4: osmotic responses of oocytes to different CPA solutions
In experiment 4a, immobilized oocytes in a 5 µl iso-osmotic PBS solution were perfused by adding 1 ml of a pre-equilibrated CPA solution (solution of 1.5 M DMSO, 1.5 M EG or 1.5 M PG in PBS). DMSO, PG and EG were from Sigma. Pre-equilibration was in a 22°C warm water bath. The CPA was carefully added with a 1000 µl pipette tip to prevent the oocyte from becoming displaced from the holding pipettes.
The temperature of the CPA solutions was measured before and after perfusion by a copper-bar sound thermometer and the initial and final temperature did not vary substantially For these experiments, we programmed the software to capture one image per second, and a total of 1000 images in one experiment. Thus, the total time of exposure was 16 min 40 s. For each group of CPA solution, tentatively nine oocytes in total were analysed. A randomization plan was generated (a block randomization of three treatments and three blocks) and used to allocate 27 oocytes that maintain spherical shape during the perfusion, to the three different CPA solutions. The data points used for analysis were obtained from: (i) image 0 until image 120, one image every 2 s; (ii) image 150 until image 720, one image every 30 s; (iii) image 750 until image 960, one image every 60 s. For each oocyte and CPA, volume change data from the image analysis was imported into Excel spreadsheets and compared with a theoretical model of volume change data calculated and the permeability coefficients were determined.
In experiment 4b immobilized oocytes in a 5 µl iso-osmotic PBS solution were perfused by adding 1 ml of a pre-equilibrated solution of EG solution in PBS (1.5 M EG at 30°C or 8°C). Pre-equilibration was either in a 30°C or in an 8°C cooled water bath.
Temperature measurements and experimental conditions for data acquisition were the same as for experiment 4a. At each temperature of exposure, tentatively eight oocytes that maintain spherical shape during the perfusion were analysed.
Statistical analysis
In the experiment on the comparison of membrane permeability coefficients, the values for Lp and Ps were analysed using a standard analysis of variance with the general linear model procedures of the statistical analysis system (SAS, Cary NC, USA), and multiple comparisons were performed with the Turkey–Kramer procedure. The significance level (
) was chosen to be 0.05.
In the experiments on osmotic tolerance limits of oocytes exposed to changes in external osmolality, standard analysis of variance approaches were used to compare percentages normal fertilization (2PN) and percentages good-quality embryos obtained. The significance level () was chosen to be 0.05.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionsAcknowledgementsReferences |
|---|
In vitro maturation of GV oocytes
GV oocytes (n = 1385) were obtained from 586 patients. After 24 h of culture, we obtained 851 MII oocytes (61.4% of GV oocytes). We also obtained 271 MI oocytes. 62 GV oocytes degenerated after a 24 h culture period and 201 oocytes remained at the GV stage. MII oocytes (n = 601) were used for the different experiments.
Morphology of oocyte shrinkage
When control IVM MII oocytes (n = 37) and EGTA-treated oocytes (n = 37) were exposed to PBS–sucrose, all oocytes shrank irregularly. When oocytes were treated with pronase (n = 37) on the other hand, 67.6% shrank spherically but when exposed to PBS–sucrose, the remaining oocytes degenerated (32.4%). When oocytes were vortex-treated (n = 37), 70.3% shrank spherically after exposure to PBS–sucrose, and the remaining oocytes (29.7%) showed irregular shrinkage. All irregular and spherically shrunken oocytes regained normal morphology when they were brought back from hyper-osmotic to iso-osmotic conditions.
We also analysed the developmental competence of vortex-treated oocytes (n = 62 MII oocytes). When compared with a control group without any treatment (n = 62 MII oocytes), the percentage of 2PN embryos obtained (72.2 versus 72.2%, respectively) did not differ between the two groups. After 72 h of culture the percentage of good-quality embryos obtained (54.3 versus 56.5%, respectively) and the number of cells in the good-quality embryos (5.7 versus 5.7) did not differ between the two groups. Therefore, it can be assumed that the vortex treatment did not affect normal oocyte development and function.
Subsequently, for determination of osmotic response to changes in external osmolality, and for determination of oocyte permeability characteristics, all oocytes were vortex-treated in a sucrose solution.
Osmotic responses of oocytes to changes in external osmolality
As a kind of validation experiment, the aim of this first series of experiments was to study osmotic inactive volume, hydraulic conductivity (Lp) and the activation energy (Ea) of human IVM MII oocytes obtained in stimulated ICSI cycles. For the determination of the oocytes fraction of osmotic inactive volume (Vb/Viso), 21 oocytes were vortex-treated and 13 of them (61.9%) showed regular shrinking when exposed to the different hyper-osmotic conditions, and were further analysed. The mean normalized equilibrium cell volumes reached (range) in solutions containing 0.3, 0.9 and 1.6 M galactose were 59% (50–67), 39% (30–43) and 31% (24–35), respectively. Analysis of the cell volume in hyper-osmotic conditions indicated that human oocytes were linear osmometers in the range of concentrations used (r2 = 0.99) and that a mean of 19 ± 4% (range 8–28%) of the total normalized cell volume was the osmotic inactive fraction (=Vb/Viso), (total solids plus osmotic inactive water). A Boyle van't Hoff plot of this relationship is shown in Figure 1.
|
When oocytes were perfused with a solution containing 0.9 M galactose, oocytes responded by loosing water, and the speed was influenced by the temperature of exposure. For determination of oocytes' hydraulic conductivity at different temperatures, 26 oocytes were vortex-treated and subsequently perfused with a 0.9 M galactose solution. During perfusion 16 oocytes (61.5%) showed regular shrinkage. The osmotic response (normalized cell volume changes/time) of 15 individual oocytes is shown in Figure 2 and these data were fitted to calculate Lp values at 30°C (five oocytes), 22°C (five oocytes) and 8°C (five oocytes). The corresponding individual Lp values as well as the mean Lp values are shown in Table 1. The activation energy of Lp is shown in Figure 3.
|
|
|
As expected, the mean values of Lp differed across the temperature range (from 0.24 µm/min/atm. at 8°C to 1.25 µm/min/atm. at 30°C). At each temperature, Lp values also differed among oocytes as could be depicted from the individual Lp values obtained at each temperature. The activation energy for Lp of IVM oocytes was 10.8 kcal/mol.
There were also wide differences among oocytes for the normalized equilibrium volumes reached at the end of the hyper-osmotic exposures.
Tolerance limits of oocytes to changes in external osmolality
The percentage of oocytes showing normal fertilization (2PN) and developing into good-quality embryos after 72 h of culture and the mean number of cells in embryos after exposure of oocytes to changes in external osmolality is shown in Table 2. The percentage of normal fertilization and the percentage of good-quality embryos after 72 h of culture were not different between the iso-osmotic condition and the six different aniso-osmotic conditions.
|
Osmotic responses of oocytes in different CPA solutions
For determination of the oocytes' permeability characteristics to DMSO, PG and EG, 38 oocytes were vortex-treated, and subsequently perfused with either a 1.5 M DMSO, 1.5 M PG or 1.5 M EG solution. During perfusion, 27 oocytes (71.0%) showed regular shrinkage. Figure 4 shows the normalized cell volume changes of 26 IVM human oocytes following one-step addition of 1.5 M DMSO (nine oocytes), PG (nine oocytes) and EG (eight oocytes) at 22°C. As expected, the oocytes initially shrank and gradually expanded as CPA and water entered the oocyte and for each CPA, the volumes reached differed substantially among oocytes. The changes in oocyte volume in the presence of DMSO, PG and EG were measured over time, and the data were fitted to calculate Lp and PCPA at 22°C. One oocyte treated with EG was excluded from the analysis because of unusual morphological features at the end of the exposure time as visualized at the time of data analysis. The mean value for the coefficients for each CPA and the individual values are listed in Table 3.
|
|
The value for LpDMSO was significantly different from that for LpPG (P < 0.05). The mean value of PPG was significantly higher than that for PEG (P < 0.05). None of the other values were statistically different although it can be depicted from the table that the value for PPG was the highest, followed by PDMSO and PEG. For each CPA, it is also clear that Lp and PCPA values differed widely among oocytes. The mean reflection coefficient, sigma, was different for PG when compared with EG and DMSO (0.56 versus 0.84 and 0.76), and for each CPA, sigma values differed widely among oocytes.
For determination of a temperature effect on oocyte permeability characteristics to EG, 25 oocytes were vortex-treated and subsequently perfused with a 1.5 M EG solution at either 30°C or 8°C. During perfusion, 17 oocytes (68.0%) showed regular shrinkage. Figure 5 shows the normalized cell volume changes of 24 IVM human oocyte following a one-step addition of 1.5 M EG at 30°C (eight oocytes), 22°C (eight oocytes) and 8°C (eight oocytes) (the figure at 22°C was the same as reported in Fig. 4). As expected, the oocytes initially shrank and gradually expand as water and CPA entered the oocyte. The degree of shrinkage and time for swelling was influenced by the temperature of the solution. Furthermore, at each temperature of exposure the normalized cell volume changes differed widely among oocytes. The changes in oocyte volume in the presence of EG were measured over time, and these data were fitted to calculate Lp and PEG at different temperatures. The mean value for these coefficients and individual values are listed in Table 4. The activation energy for Lp in the presence of EG and of PEG is given in Figure 6A and B.
|
|
|
The mean values of Lp differed widely across the temperature range (from 0.32 µm/min/atm. at 8°C to 1.39 µm/min/atm. at 30°C) and the differences between temperatures were statistically different (P < 0.05). At each temperature of exposure, Lp values in the presence of EG differed substantially among oocytes. For Lp in the presence of EG, the activation energy was calculated to be 11.2 kcal/mol (Fig. 6A).
The mean values of PEG differed widely across the temperature range (from 0.37 x 10–3 cm/min at 8°C to 2.85 x 10–3 cm/min at 30°C) and these differences were statistically different (P < 0.05). At each temperature, PEG values differed substantially among oocytes. The activation energy for PEG was calculated to be 15.3 kcal/mol (Fig. 6B). The mean reflection coefficient, sigma was not temperature-dependent and at each temperature sigma values differed widely among oocytes.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionsAcknowledgementsReferences |
|---|
Despite the potential benefits of oocyte cryopreservation for women, the relatively poor success in comparison to non-frozen oocytes continues to make this procedure investigational in nature, and further improvements in the post-thaw viability are needed to make this procedure a standard clinical option (Practice Committee ASRM, 2006). In order to develop optimal oocyte cryopreservation methods, it is important to understand the cryobiological relevant parameters of these cells such as the plasma membrane permeability characteristics to water (Lp) and CPAs (Ps), as well as the effects of volume excursions on the oocyte's ability to survive cryopreservation and develop into normal young after in vitro fertilization/ICSI and embryo transfer (Critser et al., 1997
; Paynter et al., 2005a; Paynter, 2005b). The present study was designed to investigate important biophysical characteristics of IVM human oocytes. We will discuss the relevance of these findings for the design of cryopreservation procedures for this particular cell type and for human oocytes in general. In the present study, we aimed to study osmotic tolerance limits of human IVM oocytes to changes in external osmolality and to study oocyte permeability characteristics. Therefore, human GV oocytes obtained in stimulated ICSI cycles were matured in vitro to the MII stage. After about 24 h of culture, 61.4% of GV oocytes reached the MII stage. This rate was comparable to results obtained by Cekleniak et al. (2001) using a similar approach to mature human GV oocytes obtained in stimulated ICSI cycles.
Morphology of oocyte shrinkage
When exposed to PBS–sucrose, IVM MII oocytes contracted in an eccentric manor, in a way as also described by Hunter et al. (1990) for human IVM oocytes; Ruffing et al. (1993) for bovine oocytes; Le Gal et al. (1994) for goat oocytes; Younis et al. (1996) and Songsasen et al. (2002) for monkey oocytes. To study human oocyte osmotic responses to changes in external osmolality or to study oocyte membrane permeability characteristics, it is a prerequisite that oocytes shrink spherically and uniformly. This is necessary to apply geometric formulas for the calculation of volume changes based on measurements of areas. Hunter et al. (1990) and Paynter et al. (1999, 2001) described very clearly that not all human IVM MII oocytes shrank uniformly. In the present study, none of the IVM human oocytes shrank uniformly when exposed to a PBS–sucrose solution. The irregular shrinkage of oocytes may be related to transzonal projections that persist through IVM. Transzonal projections are specialized cytoplasmic extensions derived from granulosa cells that breach the zona pellucida and establish contact with the oolemma (Albertini, 2004). Normally, transzonal processes retract during meiotic maturation in concert with mucification and expansion of the cumulus oophorus (Albertini, 2004). Uniform shrinkage could be obtained after vortex treatment of oocytes or after removal of the zona pellucida through enzymatic treatment. In the present study and in contrast to the publication by Younis et al. (1996) on rhesus monkey oocytes, EGTA treatment failed to induce regular shrinkage after hyper-osmotic treatment. Vortex treatment in a hyper-osmotic sucrose solution probably disengages the structural contacts that anchor the plasma membrane to the zona pellucida, thereby liberating them and preventing the collapse of the plasma membrane. The observation that EGTA pre-treatment was unsuccessful in disengaging these contacts might suggest that the contacts in IVM human oocytes are rather adhesion-like junctions (Albertini, 1992, 2004). Interestingly, the present study showed that when compared with IVM oocytes without any treatment, vortex treatment had no influence on their further developmental competence.
Oocytes' osmotic response to changes in external osmolality
The cell surface-to-volume ratio and the osmotic inactive cell volume (Vb) are important factors related to the formation of lethal intracellular ice during freezing (Mazur, 1963). The oocytes of most mammalian species studied to date behaved as ‘ideal osmometers’, in which their cell volume changes are linearly related to the reciprocal of the extracellular non-permeating solute osmolality (Leibo, 1980; Ruffing et al., 1993; Le Gal et al., 1994; Newton et al., 1999; Songsasen et al., 2002). In the present study, the monosaccharide galactose was used to determine the osmotic response of IVM human oocytes to changes in external osmolalities. It has been reported for human oocytes that mono- and disaccharides can be used effectively for the determination of osmotic behaviour of human oocytes and that the monosaccharide galactose showed superior characteristics when compared with the disaccharide sucrose (McWilliams et al., 1995). The use of saccharides for this purpose eliminates the deleterious effects observed when oocytes/embryos are exposed to very concentrated NaCl solutions (Mazur and Schneider, 1986; Lawitts and Biggers, 1992). The present study demonstrated that IVM human oocytes are also ideal osmometers. For human IVM oocytes, we determined a mean normalized osmotic inactive volume (Vb) value of 19%. However, we also reported that Vb values differed substantially among 13 IVM oocytes. The Vb values varied from 8 to 28%. The mean Vb value of 19% in this study was identical to in vivo matured human oocytes (Newton et al., 1999), and similar to IVM bovine (25%; Ruffing et al., 1993), in vivo matured mouse (18%; Leibo, 1980), in vivo matured goat (20%; Le Gal et al., 1994) and in vivo matured rhesus monkey oocytes (17%; Songsasen et al., 2002). The Vb value is important for enabling mathematical modelling of the dynamic cell volume changes during the series of aniso-osmotic exposures inherent in cryopreservation, and the huge range of variation in osmotic inactive volume raises a difficult problem since it will have a drastic impact on the calculations in the mathematical modelling (Gao et al., 1994).
Permeability to water and its activation energy are among the principal determinants of the response of all types of cells to freezing and thawing (Leibo, 1986). Coefficients of water permeability in the absence of CPA have been determined for oocytes of several species including human (for review see Critser et al., 1997). In general, the osmotic responses of human IVM oocytes to changes in external osmolality were similar to those of in vivo matured human oocytes and to oocytes of other species (Leibo, 1980; Hunter et al., 1990, 1992a,b; Ruffing et al., 1993; Benson and Critser, 1994; Pfaff et al., 1998). The oocytes shrank when they were exposed to a hyper-osmotic solution of a non-penetrating solute to reach a minimal volume after a certain time. The time needed to reach the minimal volume was influenced by the temperature of exposure.
At each temperature of exposure, substantial differences were observed in shrinking pattern among human IVM oocytes. The corresponding ranges for the Lp's were large. These observations confirm and extend earlier observations on human in vivo matured oocytes by Hunter et al. (1990, 1992a,b), bovine oocytes by Ruffing et al. (1993) and ICR mouse oocytes by Benson and Critser (1994). As pointed out by Hunter et al. (1992a), when considering these observations in relation to developing a cryopreservation protocol, a variable Lp for individual cells becomes important since the membrane water permeability is critical in predicting the response of the cell to osmotic stress during the freeze/thaw process (Mazur, 1963). To account for such large individual differences, a successful protocol would need to be very robust. Litkouhi et al. (1997) described that differences in Lp and Vb values between oocytes might be a reflection of differences in oocyte maturity, and Le Gal et al. (1994) discussed that differences in oocyte cellular structure and microfilament organization of the cytoskeleton may be responsible as noted in a study on goat oocytes. In the present study we used IVM oocytes that within 24 h of culture extruded the 1st polar body and that were morphologically normal, however, small differences, in the meiotic maturation between the oocytes, which influence their permeability cannot be ruled out. However, Hunter et al. (1992a,b) stated, that there does not appear to be a correlation between those oocytes with extremely high or low values for the osmotic inactive volume and those with membrane water permeability values at the extremes of the range. In their experiments, individual oocytes from the same donor investigated on the same day produced results that varied widely for both Lp and inactive volume. This observation is consistent with the hypothesis that human oocytes may reflect an inherent biological variability. In addition to that, Benson and Critser (1994) reported that among ICR mice, the water permeability of oocytes collected from different females differed significantly. In contrast, the water permeability of oocytes from different females of golden hamster did not differ significantly (Benson and Critser, 1994), neither did water permeability of different females of the genetically homogeneous hybrid strain, B6D2F1, differ significantly (Leibo, 1980). Benson and Critser (1994) concluded that the data on ICR mouse oocytes supported the hypothesis that there may be a genetic influence for the expression of important cryobiological characteristics of mammalian oocytes, that of Lp.
In the present study, the mean value for Ea of human IVM oocytes was 10.8 kcal/mol (r2 = 0.88) and this value corresponds to the value of 8.61 obtained by Hunter et al. (1992b) for fresh human MII oocytes. However, it may become apparent that it will not be possible to predict accurately an Ea value for human oocytes as a single population of cells when individual differences in Lp and inactive volumes are so large (Hunter et al., 1992a).
Mean Lp values were 1.25 µm/atm./min at 30°C, 1.01 µm/atm./min at 22°C and 0.24 µm/atm./min at 8°C. These values were twice as high as those reported by Hunter et al. (1992a) for fresh human oocytes: Lp values (0.55 µm/atm./min at 30°C, 0.40 µm/atm./min at 20°C and 0.40 µm/atm./min at 10°C). It should be noted that the latter authors exposed oocytes to hyper-osmotic NaCl solutions and that a microscope diffusion chamber was used to allow direct observation of volumetric alterations following changes in the external solution, and that in our study, oocytes were exposed to hyper-osmotic galactose solutions and a microperfusion approach was used. Furthermore, vortex treatment of oocytes was used to make oocytes shrink spherically. These differences in methodology might explain the difference in the mean Lp value between IVM human oocytes and in vivo matured oocytes. However, Ruffing et al. (1993) determined clear differences between Lp for in vivo matured (0.45 mm/min/atm.) and in vitro matured (0.84 mm/min/atm.) bovine oocytes in the presence of a non-permeating solute (NaCl) at 20°C. Therefore, we think that our own study on human IVM oocytes and the study by Ruffing et al. (1993) with bovine oocytes document the effects of the in vitro culture conditions on the permeability characteristics of developing oocytes. It has been previously reported that the osmotic permeability of human GV oocytes was significantly higher than in vivo MI and MII oocytes (Parisi et al., 2000). Others (Ruffing et al., 1993; Le Gal et al., 1994; Agca et al., 1998a, 1999) reported that Lp values were higher for IVM goat and bovine MII oocytes than for GV oocytes. Parisi et al. (2000) suggested that a water channel (aquaporin) could be lost during the transition along the different maturity stages of human oocytes (GV, MI and MII) analogous to what was found in rat oocytes where maturation in vivo or in vitro resulted in plasma membrane permeability decrease, probably as a result of the suppression at the transcriptional level of genes encoding for broad selective channels, such as aquaporin-9, that regulate the transmembrane transport of water (Ford et al., 2000). We speculate that during IVM of human oocytes, the expression of aquaporin mRNA is disturbed when compared with that in in vivo matured oocytes. Understanding the activities of aquaporins in oocytes could lead to improvements in the methods used for oocyte cryopreservation. Aquaporins-3 and -7 mRNAs have been shown to be expressed in mouse oocytes (Edashige et al., 2000). This is important in cryopreservation of oocytes because water transport (and also CPA transport) may be mediated by aquaporins-3 and -7 (Pedro et al., 2005). Induced expression of aquaporin 3 in mouse oocytes has been shown to improve water and glycerol permeability as well as oocyte survival after cryopreservation (Edashige et al., 2003).
Oocytes' tolerance limits to changes in external osmolality
To the best of our knowledge, in the present study we report for the first time on human oocytes osmotic tolerance limits as assessed by their developmental competence. Often, the methods used for CPA addition and removal are empirically derived, and not based upon the cell's tolerance to osmotic stress. Determining the osmotic tolerance limits for cells is a common approach to optimizing cryopreservation protocols (Mazur and Schneider, 1986; McWilliams et al., 1995; Songsasen et al., 2002; Fuller and Paynter, 2004; Mullen et al., 2004a). Most vitrification solutions contain high concentrations of CPAs; these may create osmotic stress and cause osmotic injury upon dilution. It has been shown before that human oocytes retain integrity of the cell membrane following shrinkage to about 40% of their initial volume in the presence of saccharides (McWilliams et al., 1995). In contrast, however, deformation of the microtubular spindle has been reported following the exposure of human oocytes to anisotonic conditions (Mullen et al., 2004b).
Our results indicate that human IVM MII oocytes tolerated exposure to solutions in a range between 39 and 2264 mOsmol/kg H2O, and furthermore, tolerated irregular shrinkage under hyper-osmotic conditions, as assayed by oocytes' developmental competence during 72 h of culture in vitro. From the present study it can be depicted that IVM human oocytes tolerate shrinkage to about 30% of their initial volume. In contrast, Agca et al. (2000b) reported that, mature bovine oocytes have a reduced blastocyst formation rate after a 10 min exposure to hyper-osmotic conditions and they indicated that blastocyst formation might be a more sensitive assay to study oocytes osmotic tolerance limits. Consequently, for determining osmotic tolerance limits of human oocytes, blastocyst formation might be a better and more sensitive parameter.
Oocytes' osmotic response in CPA solutions
When oocytes are exposed to permeating CPA solutions, they undergo volume changes such that they first shrink in response to an extracellular hypertonic CPA solution, then return to slightly greater than their initial isotonic volume as the permeating CPA enters the cell and water follows to maintain its chemical potential (Paynter, 2005b). The extent of the initial shrinkage and the subsequent time course for the cell to return to near isotonic volume are directly related to the cell's Lp and PCPA coefficients. Paynter (2005b) stated that, in general, a higher CPA permeability relative to the water permeability will result in a reduction in total volume excursion experienced by the cell, and a CPA with a higher permeability can be loaded into cells more quickly than a CPA with a lower permeability. The extent of the shrink/swell response varies also with temperature. The higher the temperature the less shrink response of the oocytes and the less the osmotic stress. However, the higher the temperature, the greater are the potential effects of CPA toxicity. Thus, the two factors need to be balanced in designing protocols for exposure to CPA.
In the present study, we describe for the first time osmotic responses and permeability coefficients and activation energies of human oocytes to the CPA EG. The results obtained also clearly point out that at 22°C, EG has a lower permeability coefficient relative to PG and DMSO, however, only the difference between PG and EG reached statistical significance. The higher permeability coefficient for PG when compared with DMSO and for DMSO when compared with EG would probably result in statistical difference when more oocytes are allocated to a randomization. The present study is the first randomized comparison on the permeability characteristics of three CPAs commonly used for the cryopreservation of human oocytes (DMSO, PG and EG). To the best of our knowledge, the present study is also the first to document on human oocyte oolemma permeability characteristics to the CPA EG. Our results are in line with previous observations by Paynter et al. (1997) and Paynter (2005b) and Pedro et al. (2005), who showed that the oolemma in mouse oocytes has a lower permeability to EG relative to PG and DMSO; by Agca et al. (1998a) who showed that bovine IVM MII oocytes had a lower permeability to EG when compared with DMSO; and by Songsasen et al. (2002) who showed that rhesus monkey oocytes have a lower permeability to EG relative to DMSO. Interestingly, Paynter et al. (1999b) reported that murine oocyte permeability to EG was lower than DMSO at 20°C and that the difference was more marked at 30°C, and at 10°C, however, the permeability to the two CPAs was identical. Fuller et al. (1992) and Paynter et al. (2001) determined mean permeability characteristics of in vivo-derived MII human oocytes for PG and Paynter et al. (1999a) and Newton et al. (1999) have determined permeability characteristics of in vivo-derived MII human oocytes for DMSO. The values for DMSO are very similar to those reported in the present study. The permeability value from the present study for PG is somewhat in between the values obtained by Fuller et al. (1992) and Paynter et al. (2001), higher when compared with the values obtained by Paynter et al. (2001) but lower when compared with the values obtained by Fuller et al. (1992). The different sources of human oocytes used in the different studies may explain the differences in oocyte permeability differences to PG. Alternatively, effects of the vortex treatment as well as the in vitro culture conditions used could also explain the differences seen.
The selectivity of the membrane for water and the properties of the CPA used determine the Sigma value, which is constrained between zero and one: the larger the sigma value; the more selective the membrane is to a given solute. In the present study, the mean reflection coefficient Sigma was lower for PG than for EG and DMSO. This means that at 22°C the membrane is more selective for EG and DMSO than for PG. It has to be noted that for calculation of permeability values the Kedem–Katchalsky (K–K) model was used, which uses water and solute permeability and the solute–solvent interaction coefficient (the reflection coefficient, Sigma) as parameters. The interaction between water and solute during their transport across cell membranes is thought to be negligible especially given the discovery in molecular biology of aquaporins, or water channels, which are very selective to water. From this discovery it can be argued that the reflection coefficient and hence the K–K model are not necessary for the calculation of the water and CPA transport across the cell membrane and that a two-parameter model might be more suitable (Chuenkhum and Cui, 2006). The relevance of all this remains to be evaluated properly.
Our results and the results by Paynter et al. (2001) and Paynter (2005b) suggest that PG might be a superior CPA compared with the others studied due to the expected reduction in shrinking and swelling during its addition and removal. In fact, PG has been the most widely used CPA for human oocyte cryopreservation by slow controlled-rate cooling procedure (Leibo, 2004).
When exposure is at 22°C, the difference in CPA permeability suggests that EG may be an inferior permeating agent for the cryopreservation of human oocytes. Furthermore, Coticchio et al. (2004) described unusual shrinkage patterns when human MII oocytes were exposed to EG solutions. Despite these observations, EG is the predominant CPA used for the vitrification of mammalian oocytes including human (Liebermann, 2002; Liebermann et al., 2003; Kuwayama et al., 2005). Therefore, the volume changes associated with permeating CPAs are only one of the many important factors to consider when designing cryopreservation procedures; inherent toxicity of the permeating CPA may be a more important consideration (Liebermann, 2002; Liebermann et al., 2003; Coticchio et al., 2004). Interestingly, substantial research work has been done recently to develop a better understanding of CPA toxicity on mammalian oocytes (Litkouhi et al., 1999; Takahashi et al., 2004; Pogorelov et al., 2006; Larman et al., 2007) and these developments have led to improved vitrification solutions for mammalian oocytes (Larman et al., 2006).
Another aspect of permeability characteristics of human oocytes that deserves further consideration is the observation that for all three CPAs investigated, there was considerable variation in osmotic response among oocytes. Consequently, for each of the three CPAs investigated, the permeability coefficients of individual oocytes varied substantially. For human fresh oocytes exposed to DMSO, McGrath et al. (1995) described the same phenomenon. Paynter et al. (2005a) and Paynter (2005b) described that the concept of storing cells in a frozen state is deceptively simple, but in reality requires an appreciation of a variety of biophysical events, and a vigilance in applying the various steps required during a cryopreservation protocol. It is well known that when using theoretical modelling, predictions of optimal freezing and vitrification can be performed, provided the values are available for the membrane coefficients for water and CPA (Agca et al., 1998b). Therefore, the observations on the high variation in membrane permeability characteristics among denuded oocytes are not without consequences, especially when one wants to cryopreserve human oocytes in a reproducible and consistent way. When using mean values of oocytes permeability coefficients in theoretical modelling in order to develop a standard protocol for freezing or vitrification protocol of human oocytes, the protocol could be optimal for one oocyte and suboptimal for the other oocyte. It can thus also be concluded that to optimize freezing or vitrification protocols of human oocytes, the use of theoretical modelling is debatable.
Paynter (2005b) stated that studies of human oocyte cryopreservation show variability in success both within and between studies. She explained this variability to some degree to the diversity of protocols used for CPA exposure, however, she pointed the oocyte quality prior to cryopreservation as the main factor, determining outcome. Therefore, it can be argued whether the observed large variations regarding the different parameters (Lp, Pcpa, Ea, Vb) studied were not related to the non-physiological nature of the research model used; this is human cumulus cell-denuded GV stage oocytes, matured in vitro and obtained after hormonal stimulation. Suboptimal culture conditions might be responsible for the differences in Lp values between in vitro and in vivo matured oocytes and suboptimal in vitro culture conditions may be responsible for some of the variability seen in the parameters studied. However, quality of in vivo matured human oocytes obtained after hormonal stimulation and quality of oocytes obtained from several other mammalian species is also not uniform as evidenced by variations among these oocytes in important cryobiological parameters (Hunter et al., 1990, 1992a</