Permeability of human oocytes to ethylene glycol and their survival and spindle configurations after slow cooling cryopreservation

Human Reproduction Vol.22, No.10 pp. 2776 2783, 2007 Advance Access publication on August 3, 2007 doi:10.1093/humrep/dem240 Permeability of human oocytes to ethylene glycol and their survival and spindle configurations after slow cooling cryopreservation L. De Santis 1, G. Coticchio 2,6, S. Paynter 3, D. Albertini 4, K. Hutt 4, I. Cino 1, M. Iaccarino 5, A. Gambardella 2, C. Flamigni 2 and A. Borini 2 1 IVF Unit, Vita-Salute University, H. S. Raffaele, Milan, Italy; 2 Tecnobios Procreazione, Via Dante 15, 40125 Bologna, Italy; 3 Cardiff University, Cardiff, UK; 4 University of Kansas Medical Center, Kansas City, KS, USA; 5 Clinica Mediterranea, Napoli, Italy 6 Correspondence address. Tel: þ39-051-2867511; Fax: þ39-051-2867512; E-mail: coticchio@tecnobiosprocreazione.it BACKGROUND: To develop novel cryopreservation methods, we estimated the permeability coefficients Lp (hydraulic conductivity) and P EG (cryoprotectant permeability) of mature human oocytes after exposure to ethylene glycol (EG) and tested the efficiency of a multi-step slow cooling protocol based on this cryoprotectant. METHODS: Oocytes were perfused with 1.5 mol/l EG for 10 min. Oocyte volume at each time point was calculated and normalized to the original volume. Slow cooling was conducted by exposing oocytes to increasing EG concentrations (0.5, 1.0 and 1.5 mol/l n 5 155) or 1.5 mol/l of propane-1,2-diol (PrOH) n 5 102. Oocytes which survived cryopreservation n 5 80 and fresh oocytes n 5 73 were prepared for confocal microscopy analysis of the meiotic spindle. RESULTS: During EG exposure, oocytes underwent an abrupt 50% volume reduction. Complete recovery of the initial volume was not observed. From the values of a best fit plot, the coefficients Lp 5 0.82 + 0.29 mm min 21 atm 21 (mean + SD) and P EG 0.10 + 0.01 mm s 21 were generated. Survival rates after freezing with EG were lower than with PrOH (51.6 versus 71.5%, respectively, P<0.05). The frequencies of normal spindle configuration were lower in frozen EG and frozen PrOH oocytes compared with fresh oocytes (53.8, 50.9 and 66.7%, respectively, P<0.05). CONCLUSIONS: The oocyte plasmalemma possesses limited permeability to EG and EG exposure causes considerable osmotic stress. However, post-thaw rates of survival and normal meiotic spindle organization may be preserved by protocols which are designed in order to minimize osmotic stress. Keywords: cryopreservation; oocytes; ethylene glycol; slow cooling; meiotic spindle Introduction In the last few years the number of pregnancies from cryopreserved human oocytes has increased considerably, suggesting that in the near future the storage of the mature female gamete could become an established option for the treatment of IVF patients. These experiences have principally resulted from the application of controlled rate freezing (slow cooling) protocols (Borini et al., 2004,2006a; Chen et al., 2005; Boldt et al., 2006; Levi Setti et al., 2006), although the potential of vitrification technology, still largely unexplored from a clinical standpoint, has recently gained the attention of cryobiologists and IVF specialists (Kuwayama et al., 2005; Lucena et al., 2006). Newly introduced protocols seem to have met the fundamental need of improving post-thaw survival rates, previously unacceptably low, but clinical experience acquired so far is in fact either contradictory or numerically insufficient to confirm that the oocyte developmental potential may be preserved substantially unaltered after cryopreservation. In effect, oocytes may be found apparently viable and without overt morphological anomalies after thawing, and yet unable to implant with high frequencies (Borini et al., 2006b; De Santis et al., 2007; Levi Setti et al., 2006). Objective evaluation of oocyte quality after storage is a difficult task (Coticchio et al., 2005), although a number of analytical methods, including polarized light and confocal microscopy assessment of the meiotic spindle, as well as ultrastructural analysis of the cell organelles, may provide important clues on possible damage caused by low temperature storage. In the attempt to improve the clinical outcome of oocyte cryopreservation, novel protocols have been derived from the application of a variety of approaches, including empirical change in the concentration of the extracellular cryoprotective agent (CPA) sucrose to achieve higher dehydration (Fabbri et al., 2001), replacement of sodium with the less toxic cation choline (Stachecki et al., 1998, 2004), or reduction in the toxicity of vitrification solutions through the application 2776 # The Author 2007. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Human oocyte cryopreservation in ethylene glycol of theoretical models (Fahy et al., 2004). Another possible option for the development of improved protocols is based on the study of the oolemma permeability to water and intracellular CPAs in order to minimize osmotic volume changes (Fuller and Paynter, 2004; Paynter, 2005), that are a recognized form of cellular stress, while ensuring appropriate dehydration to prevent intracellular ice formation during cooling. The slow cooling method based on propane-1,2-diol (PrOH) as an intracellular CPA has been electively employed for the storage of human cleavage-stage embryos. It is not surprising, then, that the same CPA has been adopted for the cryopreservation of mature oocytes. Experiments on permeability to PrOH have been generated recently in order to design oocyte-specific protocols (Paynter et al., 2001,2005), but in principle this does not rule out that other intracellular CPAs, perhaps less toxic, could play a role in the design of novel methods. High concentrations of ethylene glycol (EG) have been successfully tested for the vitrification of bovine (Bautista and Kanagawa, 1998) and mouse (Miyake et al., 1993) embryos and oocytes, suggesting good tolerance of reproductive cells to this intracellular CPA. So far, experiments on permeability and osmotic response to EG have been reported for mouse (Paynter et al., 1999), but not human, oocytes. In this study, in order to expand the possible options for the design of novel protocols, we estimated the permeability coefficients Lp (hydraulic conductivity) and P EG (permeability to EG) of mature human oocytes at room temperature (RT) from volume changes measured after exposure to a solution containing 1.5 mol/l EG for a period of 10 min. The resulting permeability coefficients were used to simulate the osmotic response of human oocytes on exposure to stepwise addition of EG and thus to design a cryoprotectant addition protocol aimed at minimizing osmotic stress. The volume change of oocytes exposed to stepwise addition of EG and subsequent exposure to EG and sucrose was also measured. Human mature oocytes were cryopreserved subsequent to addition of EG using the modified protocol and, for comparison, with a more established PrOH-based slow cooling method. Oocytes which survived cryopreservation were finally subjected to confocal microscopy for the assessment of the meiotic spindle, considering the special sensitivity of this cytoskeletal structure to low temperatures and its potential implication in oocyte viability loss after cryopreservation (Coticchio et al., 2006). Materials and Methods Source of oocytes Oocytes were obtained from consenting couples undergoing assisted reproduction treatment for male and unexplained infertility or, in one case, endometriosis. Controlled ovarian hyperstimulation was induced with long protocols using GnRH agonist and recombinant FSH, according to the standard clinical procedures routinely employed by the participating clinics (Borini et al., 2006a; De Santis et al., 2007). Ten thousand IU of HCG were administered 36 h prior to oocyte collection. After retrieval, oocytes were cultured in IVF media (Cook IVF, Brisbane, Australia, or Sage IVF Inc, Trumbull, CT, USA) for 3 4 h. Complete removal of cumulus mass and corona cells was performed enzymatically using hyaluronidase (20 40 IU/ml), and mechanically by using fine bore glass pipettes. Only oocytes devoid of any sort of dysmorphisms and showing an extruded first polar body (thus presumably at the metaphase II stage) were allocated to the study. After a total period of about 3 h following retrieval, according to their assignment, they were used for the permeability measurements n ¼ 44 or alternatively fixed for confocal analysis n ¼ 21. Frozen/thawed oocytes n ¼ 83 were cultured for further 3 h before fixation. Microperfusion The method for microperfusion of oocytes has been described previously (Paynter et al., 2001). Briefly, a single oocyte was placed in a 5-ml droplet of Dulbecco s phosphate-buffered saline (PBS) (Gibco, Life Technologies, Paisley, UK) supplemented with 20% (final concentration, 10 mg/ml) of plasma protein solution (PPS) (BAXTER AG, Vienna, Austria). The dish containing the oocyte was placed on the stage of a Nikon Diaphot 200 or alternatively of an Olympus 170 inverted microscope. A holding micropipette (Cook IVF) with a 1.5 mm diameter tip opening was used to hold the oocyte in the correct position during perfusion. The micropipette was filled with PBS and positioned adjacent to the oocyte using a Narishige micromanipulator. The pipette was then used to hold the oocyte by negative pressure generated by a Narishige IM-5A injector applied to the outer zona pellucida, care being taken not to deform the inner oolemma (measurement of the magnitude of negative pressure was not conducted in these experiments). The oocyte was then perfused by adding carefully 1 ml of perfusate by means of an air displacement pipette. The time taken between placing the oocyte in the 5-ml droplet and flushing it with the perfusate was minimized to reduce evaporation from the droplet. Perfusion was conducted with a freezing mixture containing 1.5 mol/l EG (Fluka, Milan, Italy) in PBS þ PPS. Oocyte volume changes were also monitored during stepwise addition of EG (in PBS þ PPS) by perfusing oocytes with 0.5 mol/l EG as described above. A separate group of oocytes that had been exposed to 0.5 mol/l EG for 5 min, were then perfused with 1 mol/l EG. A further group that had been exposed to 0.5 mol/l EG for 5 min followed by 1 mol/l EG for 5 min, were then perfused with 1.5 mol/l EG. Two further groups of oocytes were exposed to 0.5, 1.0 and 1.5 mol/l EG for 5 min each and were then perfused with 1.5 mol/l EG with either 0.2 or 0.3 mol/l sucrose. Data collection and analysis Each oocyte was observed before, during and after perfusion using the inverted microscope and the images were recorded by a video camera. Using computer software designed by N. Gallaghan of Wales College of Medicine, Cardiff University, three approximately equidistant points were marked on the oocyte membrane and the radius of the circle formed by joining these points was calculated. This was repeated at a set of three different positions on the oocyte membrane and the mean radius was taken as the radius of the oocyte at a given time point. The volume of the oocyte at each time point was calculated and was normalized to the volume of the oocyte immediately prior to perfusion. Best-fit plots to the measured data were generated using computer software (DIFFCHAM) designed by Professor John McGrath (McGrath et al., 1992). This software generated combinations of values for permeability of the cell to water (hydraulic conductivity) and permeability of the cell to solute, in this case EG, according to the Kedem Katchalsky model (K K model) (Kedem and Katchalsky, 1958) of the movement of solutes across cell membranes. 2777

De Santis et al. Cryopreservation solutions Oocytes were cryopreserved using a slow cooling method. All cryopreservation solutions were prepared using Dulbecco s PBS (Gibco) and a PPS (10 mg/ml, final concentration) (BAXTER AG). PrOH solutions The equilibration solution contained 1.5 mol/l PrOH þ 20% PPS in PBS and the loading solution consisted of 1.5 mol/l PrOH þ 0.3 mol/l sucrose þ 20% PPS in PBS, as described by Fabbri et al. (2001). The thawing solutions contained a gradually decreasing concentration of PrOH and a constant 0.3 mol/l sucrose concentration. They were prepared as follows: (i) 1.0 mol/l PrOH þ 0.3 mol/l sucrose þ 20% PPS; (ii) 0.5 mol/l PrOH þ 0.3 mol/l sucrose þ 20% PPS and (iii) 0.3 mol/l sucrose þ 20% PPS. EG solutions The solutions were 0.5, 1.0 and 1.5 mol/l EG þ 20% PPS in PBS (equilibration solutions) and 1.5 mol/legþ 0.2 mol/l sucrose þ 20% PPS in PBS (loading solution). The dilution solutions were (i) 1.0 mol/l EG þ 0.2 sucrose þ 20% PPS; (ii) 0.5 mol/l EGþ 0.2 sucrose þ 20% PPS and (iii) 0.2 sucrose þ 20% PPS and (iv) 20% PPS. PrOH freezing procedure Cumulus-free oocytes were incubated in the equilibration solution for 10 min at RT and then transferred to the loading solution for 5 min. The oocytes were finally loaded in plastic straws (Paillettes Crystal 133 mm; Cryo Bio System, France) individually or in small groups (maximum three oocytes per straw). The temperature was lowered through an automated Kryo 10 series III biological freezer (Planer Kryo 10/1,7 GB) from 208C to288c at a rate of 28C/min. Manual seeding was performed at 288C and this temperature was maintained for 10 min in order to allow uniform ice propagation. The temperature was then decreased to 2308C at a rate of 0.38C/min and finally rapidly brought to 21508C at a rate of 508C/min. The straws were then directly plunged into liquid nitrogen at 21968C and stored for later use. The thawing procedure was carried out at RT. Plastic straws were held in air at RT for 30 sec and then transferred into a waterbath at 308C for 40 s. Stepwise CPA dilution was performed by transferring the oocytes in thawing solution (i) for 5 min, then in solution (ii) for additional 5 min and finally in solution (iii) for 10 min before final dilution in PBS þ 20% PPS for 20 min (10 min at RT and 10 min at 378C). EG freezing procedure Oocytes were washed in PBS supplemented with 20% PPS. One to three oocytes were exposed subsequently at RT to increasing concentrations of EG, 0.5, 1.0 and 1.5 mol/l, 5 min for each equilibration step, followed by treatment for 5 min with the loading solution containing 1.5 mol/l of EG and 0.2 mol/l sucrose. Afterwards, oocytes were loaded into plastic straws (Paillettes Crystal 133 mm; Cryo Bio System, France), the total time of exposure to EG being 20 min. Details of controlled rate freezing programme are the same as those described for the PrOH protocol. After storage in liquid nitrogen, the straws were thawed rapidly in air for 30 s and afterwards in a 308C waterbath for 40 s. The CPA was removed at RT by stepwise dilution. Oocytes were expelled in the first thawing solution, 1.0 mol/l EGþ sucroseþ 20% PPS for 5 min, then equilibrated in 0.5 mol/l EGþ sucrose þ 20% PPS for additional 5 min. Finally they were placed in sucrose þ 20% PPS for 10 min before final dilution in PBS þ 20% PPS for 20 min (10 min at RT and 10 min at 378C). 2778 Immunofluorescence Only morphologically normal oocytes were used for this study and randomly assigned to the control (fresh) and treatment (frozen) groups. After freezing and thawing, oocytes showing swelling or shrinkage, vacuoles, membrane blebbing or other anomalies were not considered viable and were suitable for microscopy analysis. Prior to fixation, thawed oocytes were cultured for 3 h in 20 ml drops of glucose-free medium (Cook IVF, Brisbane, Australia) normally used for embryo culture under warm mineral oil at 378C inan atmosphere of 5% CO 2 in air. Oocytes were fixed and processed for microtubules, DNA and f-actin immunofluorescence as previously described (Combelles et al., 2002). Fixation and extraction was carried out for 30 60 min at 378C ina microtubule stabilizing buffer (100 mm PIPES, 5 mm MgCl 2, 2.5 mm EGTA, 2% formaldehyde, 0.1% Triton-X-100, 1 mm taxol, 10 U/ml aprotinin and 50% deuterium oxide) and stored until use in a blocking solution comprising PBS supplemented with 2% bovine serum albumin, 2% skim milk powder, 2% normal goat serum, 100 mm glycine, 0.01% Triton X-100 and 0.2% sodium azide. For immunostaining of microtubules, oocytes were first incubated with mouse monoclonal anti-ab tubulin (Sigma) diluted 1:100 in blocking solution for 1 h at 378C, followed by Alexa 488 labelled goat antimouse immunoglobulin G (Molecular Probes) diluted 1:800 in wash solution for 1 h at 378C. DNA was stained with Hoechst 33 258 (1 mg/ml in blocking solution) for 30 min and membrane integrity was analysed by staining actin with rhodamine labelled phalloidin (1 mg/ml in blocking solution) for 30 min. Oocytes were mounted under cover slips without compression in medium containing 50% glycerol and 25 mg/ml sodium azide. Oocytes were analysed using a Zeiss LSM Pascal confocal imaging system mounted on a Zeiss Axioscope II with UV (405 nm), HeNe (543 nm) and Argon (488 nm) laser excitation of probes. For every spindle, a complete Z-axis data set was collected at 0.5 0.7 mm intervals (20 sections/ spindle) using a 63 oil objective (na ¼ 1.4). Spatial restoration and three-dimensional projections for each Z-series data set were computed and analysed using LSM 5 Image Browser. Statistical analysis Statistical comparison between fresh and frozen groups was carried out on pooled data using chi square analysis. A P-value,0.05 was considered significant. Results are presented as mean+sd. Results Permeability to EG On initial exposure to 1.5 mol/l EG, as water left the cell, 15/ 27 oocytes shrank in a very irregular fashion, with massive distortions that compromised their original spherical shape (Fig. 1). Those oocytes were not used to predict permeability Figure 1: Mature oocyte immediately before (A) and shortly after (B) (~40 s) exposure to 1.5 mol/l EG

Human oocyte cryopreservation in ethylene glycol exposure to 1.5 mol/l EG and 0.2 mol/l or 0.3 mol/l sucrose oocytes shrank continuously achieving a volume of 62 and 55%, respectively, after 5 min of exposure. Figure 2: Mean+SD normalized volume of mature oocytes during exposure to 1.5 mol/l EG at RT to water and CPA because their irregular shrinkage was not compatible with the requirements of the model adopted in this study. For those oocytes in which sphericity was preserved throughout perfusion (n ¼ 12), their individual volume was plotted as normalized volume, that corresponds to the volume at a given time point following addition of CPA solution divided by the volume of the same oocyte in isotonic medium (PBS) immediately prior to perfusion. On average, oocytes underwent a 50% reduction in the original volume within about 40 s. As water and cryoprotectant entered, re-expansion occurred but full recovery of the initial volume was not observed by the end of the treatment period (Fig. 2). For each oocyte a best-fit plot was generated and, from this, values for the K-K coefficients Lp ¼ 0.82 + 0.29 mm min 21 atm 21 and P EG 0.10 + 0.01 mm s 21 were generated. The predicted response was in good agreement with the measured response in each case. The volume change of oocytes exposed to stepwise addition of 1.5 mol/l EG is depicted in Fig. 3. On exposure to 0.5 mol/l EG oocytes reached a minimum volume of 70% and recovered to 76% after 5 min exposure. On exposure to 1.0 mol/l EG oocytes shrank by a further 16%, giving an overall shrinkage from the volume at time 0 of 64%, and recovered to 72% of that volume after 5 min. When 1.5 mol/l EG was added oocytes shrank by a further 9% giving an overall shrinkage of 65% and recovered to 70% initial volume. On subsequent Survival after thawing A total of 155 oocytes were frozen-thawed with the three-step EG protocol designed on the basis of the permeability of the oolemma to this CPA. Immediately after thawing, all oocytes appeared intact, although a proportion of them showed irregular shrinkage. During rehydration and EG dilution some oocytes degenerated, but the majority of oocyte loss was observed during the following 3 h. After post-thaw culture, 80 oocytes were considered viable, i.e. showing no sign of swelling or shrinkage, vacuoles, membrane blebbing or other anomalies (Fig. 4), corresponding to a 51.6% survival rate. For comparison, another group of oocytes was frozen-thawed with the 1.5 mol/l PrOH, 0.3 mol/l slow cooling protocol, achieving a survival rate of 71.5% (73/102) (P, 0.05). A proportion of the survived oocytes were allocated to confocal analysis. Figure 4: Representative micrographs of oocytes considered viable (A) or degenerate (B and C) after thawing and incubation for at least 3 h Meiotic spindle and chromosome configurations In fresh and frozen-thawed oocytes, in most cases (50 67%) the chromosome segregation apparatus was found undisturbed, i.e. with barrel-shaped and bipolar spindles, microtubules converging at both poles and all chromosomes present and evenly aligned at the equatorial plate (Fig. 5a). However, to a certain degree in all groups both the microtubular and chromatin elements underwent moderate to major alterations. The diverse configurations were classified according to the following categories. Bipolar spindle/non-aligned chromosomes Oocytes in this group had barrel-shaped and bipolar spindles with microtubules meeting at both poles, but chromosomes were not evenly aligned on the equatorial plate (Fig. 5b). Figure 3: Mean normalized volume of mature oocytes during exposure to stepwise addition of EG0 300 s addition of 0.5 mol/l EG (n ¼ 10), 300 600 s addition of 1.0 mol/l EG(n ¼ 10), 600 900 s addition of 1.5 mol/l EG(n ¼ 12), 900 1200 s addition of 1.5 mol/l EG with 0.2 (open squares n ¼ 10) or 0.3 mol/l sucrose (closed squares n ¼ 10) Disarranged spindle/aligned chromosomes In this category, spindles deviated from the typical barrelshaped bipolar structure. Instead, clusters of disorganized microtubules were evident, as well as multi-polar spindles or spindles with microtubules not converging at one or both 2779

De Santis et al. Figure 5: Representative confocal images of meiotic spindle and chromosome organizationmeiotic spindles were classified as: (A) bipolar spindle with aligned chromosomes, (B) bipolar spindle with non-aligned chromosomes, (C) disarranged spindle with aligned chromosomes or (D) disarranged or absent spindle with dispersed or absent chromosomes. The asterisks indicates dispersed chromosome. Tubulin is green and DNA is red (changed from blue to enhance contrast). Scale bar represents 5 mm poles. Despite the failure to form a normal spindle structure, chromosomes were associated with microtubules, and were closely aligned and evenly spaced from one another (Fig. 5c). Disarranged or absent spindle/dispersed or absent chromosomes Oocytes in this category exhibited severe abnormalities, including absent or abnormal spindles, and in this group, chromosomes were dispersed unevenly or were not detected at all (Fig. 5d). Table 1 details the incidence of each of these categories in the fresh and frozen groups. In the unfrozen control, the majority of oocytes (66.7%) exhibited a typical bipolar spindle and aligned chromosomes. In the frozen PrOH group, normal spindle and chromosome configuration were found in about half of the analysed oocytes (50.9%). The other samples of this group displayed microtubule and/or chromosome abnormalities falling to varying extents into the other three classification categories. In the EG group, bipolar spindles and aligned chromosomes were found in 53.8% of oocytes, whilst the second most represented category was the one including severe anomalies, i.e. disarranged or absent spindles, with dispersed or absent chromosomes (38.5%). The difference in the frequency of oocytes with normal bipolar spindle between the control and PrOH/EG groups was significant (P, 0.05). The actin cytoskeleton was found as an uninterrupted meshwork of cortical filaments without significant distortions in the large majority of oocytes in the fresh, PrOH and EG groups (85.7, 73.7 and 88.5%, respectively) (Table 1). Discussion Oocyte cryopreservation has progressively become an area of major interest in human IVF, having the potential to compete with embryo storage for achieving maximal efficiency of clinical treatment and at the same time to expand the assisted reproduction options by preserving oocyte quality irrespective of female age. Recent studies have supported these hopes, by reporting high oocyte survival and clinical pregnancy rates (Chen et al., 2005; Kuwayama et al., 2005). Nevertheless, from a methodological and biological standpoint various issues remain either unresolved or open to further progress. In this study, we designed a cryoprotectant addition protocol based on increasing EG concentrations (0.5, 1.0 and 1.5 mol/l) after assessing the oolemma permeability to this CPA. We also ascertained that under the conditions tested, slow cooling storage with EG is able to preserve to a limited extent the configuration of the meiotic spindle and associated chromosomes. From previous vitrification studies, it was known that various cell types and organs, including mouse embryos (Kasai et al., 1990; Kasai and Mukaida, 2004), exhibit relatively higher tolerance to brief treatments of very high concentrations to EG, in comparison to other CPAs. EG has been also adopted successfully for the formulation of mixtures used for the slow cooling storage of mouse embryos (Miyamoto and Ishibashi, 1977). More Table 1: Frequencies of meiotic spindle configuration, chromosome organization and membrane integrity in fresh oocytes or in the surviving morphologically normal fraction after cryopreservation Treatment (Number of oocytes) Meiotic spindle and chromosome configurations (%) Cortical actin (%) Bipolar spindle, aligned chromosomes Bipolar spindle, non-aligned chromosomes Disarranged spindle, aligned chromosomes Disarranged or absent spindle, dispersed or absent chromosomes Normal Abnormal Fresh (21) 14 (66.7)a 5 (23.8) 2 (9.5) 0 (0.0) 18 (85.7) 3 (14.3) Frozen, PrOH (57) 29 (50.9)b 6 (10.5) 10 (17.5) 12 (21.1) 42 (73.7) 15 (26.3) Frozen, EG (26) 14 (53.8)b 0 (0.0) 2 (7.7) 10 (38.5) 23 (88.5) 3 (11.5) a versus b: P, 0.05. EG, ethylene glycol; PrOH, propane-1,2-diol. 2780

Human oocyte cryopreservation in ethylene glycol recently, slow cooling cryopreservation with EG of mature cat oocytes was reported to produce higher survival, cleavage and proportion of 8-cell stage embryos compared with dimethylsulphoxide (Luvoni and Pellizzari, 2000). Until now, the permeability of human oocytes to EG, and the possible stress after exposure to this CPA, has not been investigated. In the attempt to develop alternative slow cooling protocols, firstly we monitored the osmotic response of these cells to 1.5 mol/l EG over a period of 10 min, i.e. under osmotic conditions normally applied during slow cooling. The sharp decrease in the cell volume on initial exposure to EG was a manifestation of a much higher permeability to water, i.e. the initial loss of water was not compensated by an equivalent intake of CPA. After initial shrinkage, the cell volume increased as a result of CPA penetration, but without fully recovering the original size over the entire observation period. In the majority of oocytes (55.6%) such a volume excursion occurred with a transient loss of sphericity. We do not know if and to what extent this response may affect post-thaw viability. It is interesting to note, however, that germinal vesical-stage oocytes cryopreserved with the classical PrOHbased slow cooling protocol undergo a high rate of non-spherical shrinkage (G. Coticchio and S. Paynter, unpublished data) but their post-thaw degeneration rate remains low Sereni E, Bonu MA, Borini A, Sciajno R, Trevisi MR and Flamigni C. (2000) High survival rate after cryopreservation of human prophase I oocytes. Fertility and Sterility 74 No. 3S,161. A very similar osmotic response curve was previously described in the case of mouse oocytes (Paynter et al., 1999), suggesting that this animal model is a good surrogate for the human for the type of experiments reported in the present work. The response of oocytes to the presence of EG is in contrast to that in the presence of PrOH where, under similar conditions, a minimum volume of only 70% was measured and recovery was faster, achieving 90% of initial volume within 5 min of exposure (Paynter et al., 2001). Considering that following exposure to 1.5 mol/l EG the abrupt and pronounced volume change was indicative of a high risk of osmotic stress, we ruled out the possibility to test a slow cooling protocol following direct exposure to such a CPA concentration. Using the permeability coefficients generated in the presence of EG, we simulated the volume response of stepwise addition of EG at 258C. Using steps of 0 0.5, 0.5 1.0 and then 1.0 1.5 mol/l, we predicted the minimum volume of the cell at each step to not exceed 75% of initial volume. When the response was measured, the initial step gave a minimum volume of 70% but the subsequent steps remained within the predicted range. The five minute exposure time at each step meant that oocytes were shrunken when additional cryoprotectant was added but, even when this was taken into account, the minimum volume reached during exposure to EG alone was 64%, hence considerably less shrinkage occurred than with one step addition of EG. Another interesting feature of the stepwise addition protocol is that the incidence of nonspherical shrinkage during cryoprotectant addition was decreased to,25%. Whether non-spherical shrinkage is an indicator of greater stress is unknown but, whereas with the one step addition of EG half of the oocytes used in this study could not be measured, with the multi-step protocol most of the oocytes were used and hence gave an accurate representation of what can be considered to be the normal response in these circumstances. With this approach, we were able to obtain a survival rate of 51.6% that was significantly lower compared with the alternative PrOH protocol (71.5%). It is possible that such a difference may have been generated by intrinsically different sensitivity of the oocyte to the two CPAs, in terms of osmotic response and overall toxicity. Otherwise, the lower sucrose concentration in the loading solution of the EG protocol (0.2 instead of 0.3 mol/l) may have given rise to insufficient dehydration, with a consequent increased risk of intracellular ice formation. Measurement of human oocyte volume on one step exposure to 1.5 mol/l PrOH has been performed previously (Paynter et al., 2001) and showed that oocytes would be expected to return to initial volume after 10 min exposure at 258C. Measurement of oocytes during exposure to 1.5 mol/l PrOH and 0.2 or 0.3 mol/l sucrose, having previously been exposed to 1.5 mol/l PrOH for 10 mins at 258C, gave volumes of 72 and 62%, respectively (Paynter et al., 2005). Hence the degree of shrinkage and cellular dehydration at the time freezing commences is the same for the two protocols used in this study. Thus the extent of ice formation could be expected to be equivalent with these two protocols. In this respect, it is important to note that by adopting different mixtures of 1.5 mol/l PrOH and sucrose (0.1, 0.2 and 0.3 mol/l) in the loading solution, Fabbri et al. (2001) reported survival rates that were proportional to the concentrations of the non-penetrating CPA (39, 58 and 83%, respectively). The beneficial effect of high sucrose concentration on the survival rate has been confirmed by other authors (Chen et al., 2005; Borini et al., 2006b; Levi Setti et al., 2006). It is striking that, irrespective of the use of EG or PrOH, very similar survival rates may be obtained employing the same sucrose concentration (0.2 mol/l) in the loading solution. A further factor for consideration is the use of an identical sucrose concentration (0.2 mol/l) in the freezing and thawing solutions. This may have resulted in insufficient control of osmotic stress during post-thaw rehydration. In fact, Jericho et al. (2003) have shown that, after freezing with a 1.5 mol/l PrOH, 0.2 mol/l sucrose solution, a higher sucrose concentration (0.3 instead of 0.2 mol/l) during postthaw rehydration is able to significantly improve the survival rate of biopsied Day 3 human embryos. Therefore, in terms of survival rate the specific merit of the strategy of exposing human oocytes to increasingly higher concentrations of EG during the application of a slow cooling protocol remains open to further investigation and improvement. On the other hand, it is possible that an efficient slow cooling protocol based on EG could be perfected by adjusting the concentrations of sucrose during dehydration and rehydration, similar to previous experiments conducted with PrOH. In effect, recent data indicate that differential sucrose concentration during dehydration (0.2 mol/l) and rehydration (0.3 mol/l) increases overall oocyte viability (Bianchi et al., 2007). This prompted us to plan further EG experiments adopting the same criterion of sucrose use during dehydration and rehydration. The fact that, in oocytes frozen with EG, post-thaw degeneration occurred during rehydration or in the following 3 h of culture may suggests that in fact the critical factor for survival is 2781

De Santis et al. represented by the rehydration conditions rather than the formation of intracellular ice during freezing/thawing. Another option for protocol improvement could involve EG exposure at a higher temperature to achieve the same effect of dehydration within a shorter period (this strategy has been used with vitrification). Indeed, in the EG protocol oocytes were exposed for a longer time to dehydration conditions in comparison with the PrOH protocol (20 and 15 min, respectively), a factor that may have an effect on cell survival. In consideration of the importance for normal development and specific sensitivity of the meiotic spindle to lower, suboptimal temperatures (Pickering et al., 1990; Wang et al., 2001), we allocated a proportion of the surviving oocytes to analysis of cytoskeletal and chromatin configurations by using confocal microscopy. Deformation of the spindle of human oocytes has been reported following exposure to hypo and hyperosmotic conditions (Mullen et al., 2004). Studies on the status of the meiotic spindle in human oocytes after cryopreservation have often been conflicting (Gook et al., 1993; Park et al., 1997; Cobo et al., 2001; Boiso et al., 2002), leaving uncertainty on the question of a possible increased risk of aneuploidy. However, more recently, it has been shown that protocols that ensure high survival rates can also preserve the organization of the meiotic spindle. In particular, normal spindle and chromosome configurations have been found in a similar proportion (~70%) in fresh and frozen oocytes, after storage with a slow cooling choline-based protocol (Stachecki et al., 2004). Recently, we confirmed this result after storage with a slow cooling protocol involving high sucrose concentration (Coticchio et al., 2006). In reality, the spindle microtubules undergo depolymerization during dilution of the penetrating CPA after thawing (Rienzi et al., 2004). Nevertheless, shortly after that the oocyte is fully rehydrated, and under certain conditions, the tubulin fibres reorganize into a well organized structure (Bianchi et al., 2005). We found a normal barrel-shaped bipolar spindle and chromosomes orderly aligned on the equatorial plane in 70% of fresh control oocytes. This result is in agreement with previously published works (Stachecki et al., 2004; Coticchio et al., 2006), confirming that even in the absence of cryostorage an important proportion of human oocytes are at risk of defective chromosome segregation or other dysfunctions, such as failed fertilization or irregular cleavage, attributable to irregular or absent spindle organization. After storage with EG, the percentage of normal oocytes was moderately but significantly lower (53.8%) compared with fresh oocytes. In the PrOH group, the proportion of oocytes with normal spindle and chromosome configuration was 50.9%, i.e. comparable to the EG group but also in this case significantly reduced with respect to the control group. Recently, we reported that, after thawing with the same 1.5 mol/ PrOH, 0.3 mol/l protocol, the frequency of oocytes displaying a normal microtubular apparatus and chromosome alignment is unchanged compared with fresh oocytes. A possible explanation of this discrepancy may lie in the fact that while the oocytes included in our previous work were obtained from patients younger than 36 years, the oocytes used in the present study, although randomized, were donated from patients of unselected age. Because it 2782 has been established that advanced maternal age affects spindle function (Battaglia et al., 1996), it is possible that the spindle of oocytes of older patients is partially compromized in its ability to repolymerize after thawing. This hypothesis clearly requires further investigation, also in consideration of the obvious clinical implications. In conclusion, in the present study we observed for the first time that the oocyte plasmalemma possesses limited permeability to EG. We tested the hypothesis that a sequential dehydration protocol could reduce the osmotic stress generated by exposure to this CPA at a concentration normally adopted in slow cooling. While in terms of survival and preservation of the meiotic spindle organization this multi-step EG-based protocol does not appear to offer advantages in comparison to a method based on PrOH, nevertheless the well-documented low toxicity of EG warrants further investigation on the efficiency of protocols involving this CPA. Acknowledgements The present study was supported by grants from the Italian National Health Institute and from the Italian Ministry of Education, University and Research (PRIN Program), years 2005 2007. We are very grateful to Professor Barry Fuller for critical reading of manuscript and useful comments. References Battaglia DE, Goodwin P, Klein NA, Soules MR. Influence of maternal age on meiotic spindle assembly in oocytes from naturally cycling women. Hum Reprod 1996;11:2217 2222. Bautista JA, Kanagawa H. Current status of vitrification of embryos and oocytes in domestic animals: ethylene glycol as an emerging cryoprotectant of choice. Jpn J Vet Res 1998;45:183 191. Bianchi V, Coticchio G, Fava L, Flamigni C, Borini A. Meiotic spindle imaging in human oocytes frozen with a slow freezing procedure involving high sucrose concentration. Hum Reprod 2005;20:1078 1083. Bianchi V, Coticchio G, Distratis V, Di Giusto N, Flamigni C, Borini A. Differential sucrose concentration during dehydration (0.2 mol/l) and rehydration (0.3 mol/l) increases the implantation rate of frozen human oocytes. Reprod Biomed Online 2007;14:64 71. Boiso I, Marti M, Santalo J, Ponsa M, Barri PN, Veiga A. A confocal microscopy analysis of the spindle and chromosome configurations of human oocytes cryopreserved at the germinal vesicle and metaphase II stage. Hum Reprod 2002;17:1885 1891. Boldt J, Tidswell N, Sayers A, Kilani R, Cline D. Human oocyte cryopreservation: 5-year experience with a sodium-depleted slow freezing method. Reprod Biomed Online 2006;13:96 100. Borini A, Bonu MA, Coticchio G, Bianchi V, Cattoli M, Flamigni C. Pregnancies and births after oocyte cryopreservation. Fertil Steril 2004;82:601 605. Borini A, Lagalla C, Bonu MA, Bianchi V, Flamigni C, Coticchio G. Cumulative pregnancy rates resulting from the use of fresh and frozen oocytes: 7 years experience. Reprod Biomed Online 2006a;12:481 486. Borini A, Sciajno R, Bianchi V, Sereni E, Flamigni C, Coticchio G. Clinical outcome of oocyte cryopreservation after slow cooling with a protocol utilizing a high sucrose concentration. Hum Reprod 2006b;21:512 517. Chen SU, Lien YR, Chen HF, Chang LJ, Tsai YY, Yang YS. Observational clinical follow-up of oocyte cryopreservation using a slow-freezing method with 1,2-propanediol plus sucrose followed by ICSI. Hum Reprod 2005;20:1975 1980. Cobo A, Rubio C, Gerli S, Ruiz A, Pellicer A, Remohi J. Use of fluorescence in situ hybridization to assess the chromosomal status of embryos obtained from cryopreserved oocytes. Fertil Steril 2001;75:354 360. Combelles CM, Cekleniak NA, Racowsky C, Albertini DF. Assessment of nuclear and cytoplasmic maturation in in-vitro matured human oocytes. Hum Reprod 2002;17:1006 1016.

Human oocyte cryopreservation in ethylene glycol Coticchio G, Bonu MA, Bianchi V, Flamigni C, Borini A. Criteria to assess human oocyte quality after cryopreservation. Reprod Biomed Online 2005;11:421 427. Coticchio G, De Santis L, Rossi G, Borini A, Albertini D, Scaravelli G, Alecci C, Bianchi V, Nottola S, Cecconi S. Sucrose concentration influences the rate of human oocytes with normal spindle and chromosome configurations after slow-cooling cryopreservation*. Hum Reprod 2006;21:1771 1776. De Santis L, Cino I, Rabellotti E, Papaleo E, Calzi F, Fusi FM, Brigante C, Ferrari A. Oocyte cryopreservation: clinical outcome of slow-cooling protocols differing in sucrose concentration. Reprod Biomed Online 2007;14:57 63. Fabbri R, Porcu E, Marsella T, Rocchetta G, Venturoli S, Flamigni C. Human oocyte cryopreservation: new perspectives regarding oocyte survival. Hum Reprod 2001;16:411 416. Fahy GM, Wowk B, Wu J, Paynter S. Improved vitrification solutions based on the predictability of vitrification solution toxicity. Cryobiology 2004;48:22 35. Fuller B, Paynter S. Fundamentals of cryobiology in reproductive medicine. Reprod Biomed Online 2004;9:680 691. Gook DA, Osborn SM, Johnston WI. Cryopreservation of mouse and human oocytes using 1,2-propanediol and the configuration of the meiotic spindle. Hum Reprod 1993;8:1101 1109. Jericho H, Wilton L, Gook DA, Edgar DH. A modified cryopreservation method increases the survival of human biopsied cleavage stage embryos. Hum Reprod 2003;18:568 571. Kasai M, Komi JH, Takakamo A, Tsudera H, Sakurai T, Machida T. A simple method for mouse embryo cryopreservation in a low toxicity vitrification solution, without appreciable loss of viability. J Reprod Fertil 1990;89:91 97. Kasai M, Mukaida T. Cryopreservation of animal and human embryos by vitrification. Reprod Biomed Online 2004;9:164 170. Kedem O, Katchalsky A. Thermodynamic analysis of the permeability of biological membranes to non-electrolytes. Biochim Biophys Acta 1958;27:229 246. Kuwayama M, Vajta G, Kato O, Leibo SP. Highly efficient vitrification method for cryopreservation of human oocytes. Reprod Biomed Online 2005;11:300 308. Levi Setti PE, Albani E, Novara PV, Cesana A, Morreale G. Cryopreservation of supernumerary oocytes in IVF/ICSI cycles. Hum Reprod 2006;21:370 375. Lucena E Bernal DP, Lucena C, Rojas A, Moran A, Lucena A. Successful ongoing pregnancies after vitrification of oocytes. Fertil Steril 2006;85: 108 111. Luvoni GC, Pellizzari P. Embryo development in vitro of cat oocytes cryopreserved at different maturation stages. Theriogenology 2000;53: 1529 1540. McGrath JJ, Hunter J, Bernard A, Fuller BJ. On the use of microdiffusion chamber methods to determine the coupled transport of water and cryoprotective agents across biological membranes. In: Diller KR, Shitzer A (eds). Macroscopic and Microscopic Heat and Mass Transfer in Biomedical Engineering. Amsterdam: Elsevier Press, 1992. 271 296. Miyake T, Kasai M, Zhu SE, Sakurai T, Machida T. Vitrification of mouse oocytes and embryos at various stages of development in an ethylene glycol-based solution by a simple method. Theriogenology 1993;40:121 134. Miyamoto H, Ishibashi T. Survival of frozen-thawed mouse and rat embryos in the presence of ethylene glycol. J Reprod Fertil 1977;50:373 375. Mullen SF, Agca Y, Broermann DC, Jenkins CL, Johnson CA, Critser JK. The effect of osmotic stress on the metaphase II spindle of human oocytes, and the relevance to cryopreservation. Hum Reprod 2004;19:1148 1154. Park SE, Son WY, Lee SH, Lee KA, Ko JJ, Cha KY. Chromosome and spindle configurations of human oocytes matured in vitro after cryopreservation at the germinal vesicle stage. Fertil Steril 1997;68:920 926. Paynter SJ. A rational approach to oocyte cryopreservation. Reprod Biomed Online 2005;10:578 586. Paynter SJ, Borini A, Bianchi V, De Santis L, Flamigni C, Coticchio G. Volume changes of mature human oocytes on exposure to cryoprotectant solutions used in slow cooling procedures. Hum Reprod 2005;20: 1194 1199. Paynter SJ, Fuller BJ, Shaw RW. Temperature dependence of Kedem- Katchalsky membrane transport coefficients for mature mouse oocytes in the presence of ethylene glycol. Cryobiology 1999;39:169 176. Paynter SJ, O Neil L, Fuller BJ, Shaw RW. Membrane permeability of human oocytes in the presence of the cryoprotectant propane-1,2-diol. Fertil Steril 2001;75:532 538. Pickering SJ, Braude PR, Johnson MH, Cant A, Currie J. Transient cooling to room temperature can cause irreversible disruption of the meiotic spindle in the human oocyte. Fertil Steril 1990;54:102 108. Rienzi L, Martinez F, Ubaldi F, Minasi MG, Iacobelli M, Tesarik J, Greco E. Polscope analysis of meiotic spindle changes in living metaphase II human oocytes during the freezing and thawing procedures. Hum Reprod 2004;19:655 659. Sereni E, Bonu MA, Borini A, Sciajno R, Trevisi MR, Flamigni C. High survival rate after cryopreservation of human prophase I oocytes. Fertil Steril 2000;74: p.161. Stachecki JJ, Cohen J, Willadsen SM. Cryopreservation of unfertilized mouse oocytes: the effect of replacing sodium with choline in the freezing medium. Cryobiology 1998;37:346 354. Stachecki JJ, Munne S, Cohen J. Spindle organization after cryopreservation of mouse, human, and bovine oocytes. Reprod Biomed Online 2004;8: 664 672. Wang WH, Meng L, Hackett RJ, Odenbourg R, Keefe DL. Limited recovery of meiotic spindles in living human oocytes after cooling-rewarming observed using polarized light microscopy. Hum Reprod 2001;16:2374 2378. Submitted on September 1, 2006; resubmitted on May 29, 2007; accepted on June 7, 2007 2783