Outlook Truths and myths of oocyte sensitivity to controlled rate freezing

Transcription

1 RBMOnline - Vol 15. No Reproductive BioMedicine Online; on web 22 May 2007 Outlook Truths and myths of oocyte sensitivity to controlled rate freezing Giovanni Coticchio was awarded a PhD in cell and developmental biology from the University of Palermo, Italy in In , he specialized in human IVF, achieving a Master of Medical Sciences Degree in assisted reproduction technology at the University of Nottingham, UK. Whilst there ( ), he developed his research and clinical embryology interests as post-doctoral fellow. In 1998 he joined Tecnobios Procreazione, Bologna, Italy, where he has been appointed research supervisor. In 1999 he was visiting scientist at the Department of Obstetrics and Gynaecology, University of Sydney, Australia. His major areas of interests are oocyte cryopreservation and oocyte maturation. Dr Giovanni Coticchio G Coticchio 1, MA Bonu, R Sciajno, E Sereni, V Bianchi, A Borini Tecnobios Procreazione, via Dante 15, Bologna, Italy 1 Correspondence: Tel: ; Fax: ; coticchio@tecnobiosprocreazione.it Abstract The mammalian oocyte is especially sensitive to cryopreservation. Because of its size and physiology, it can easily undergo cell death or sub-lethal damage as a consequence of intracellular ice formation, increase in the concentration of solutes and other undesired effects during the conversion of extracellular water into ice. This has generated the belief that oocyte storage cannot be achieved with the necessary efficiency and safety. However, many concerns raised by oocyte freezing are the result of unproven hypotheses or observations conducted under sometimes inappropriate conditions. For instance, spindle organization can undergo damage under certain freezing conditions but not with other protocols. The controversial suggestion that cryopreservation induces cortical granule discharge and zona pellucida hardening somehow questions the routine use of sperm microinjection. Damage to mouse oocytes caused by solute concentration is well documented but, in the human, there is no solid evidence that modifications of freezing mixtures, to prevent this problem, provide an actual advantage. The hope of developing oocyte cryopreservation as a major IVF option is becoming increasingly realistic, but major efforts are still required to clarify the authentic implications of oocyte cryopreservation at the cellular level and identify freezing conditions compatible with the preservation of viability and developmental ability. Keywords: calcium, cryopreservation, cryoprotectant, intracellular ice formation, oocyte, solute concentration Introduction 24 Over the last three decades, IVF has become an established procedure for the treatment of almost all types of human infertility. In this respect, cryopreservation has played an important role, allowing the postponed use of surplus embryos and thereby considerably increasing the chances of achieving a pregnancy per cycle of stimulation (Tiitinen et al., 2004). Nevertheless, embryo storage implies well-recognized legal and ethical complications that make it a less than ideal form of fertility preservation. In principle, oocyte cryopreservation would offer a better option, circumventing the major objections raised against embryo freezing, giving women increased control of their reproductive function and making oocyte donation safer and simpler. In 1977, Whittingham (1977) pioneered the field of oocyte cryopreservation, obtaining in the mouse relatively high postthaw survival rates (60 70%). Whittingham described a reduced fertilization ability of frozen thawed oocytes. However, he also reported that embryos from cryopreserved oocytes displayed high implantation and developmental ability after transfer to pseudopregnant recipients. The first attempts at introducing oocyte cryopreservation into the clinical routine date back to as early as 1986 (Chen, 1986; Al-Hasani et al., 1987). Having been performed with inadequate protocols and methodologies, it should not have been surprising that these studies did not result in significant progress. Human oocyte cryopreservation was not explored for the following decade, having been acritically assumed to be too difficult to achieve. More recently, viable pregnancies obtained from frozen oocytes have stimulated renewed interest. However, in general the IVF community still appears reluctant to replace more orthodox cryopreservation procedures. Reasonably, having been tested only in limited cases, oocyte storage should 2007 Published by Reproductive Healthcare Ltd, Duck End Farm, Dry Drayton, Cambridge CB3 8DB, UK

2 still be considered experimental. For a number of reasons, the mature oocyte is undoubtedly more difficult to preserve than the cleavage-stage embryo. Notwithstanding, the assumption that the oocyte is especially sensitive to low temperature storage has often been based on mere prejudice. For example, concerns on possible damage to the meiotic spindle have been disproved by recent findings based on oocytes cryopreserved with slow cooling protocols, as discussed below. Recently, on the other hand, vitrification has stirred considerable enthusiasm (Kuwayama et al., 2005). The notion that this approach causes less harm to the oocyte and, as consequence, is able to generate a superior clinical outcome is quickly spreading among IVF specialists. In fact, at the moment the evidence in support of these opinions remains largely insufficient or absent. Clearly, the oocyte soft spots are diverse. Studies aiming at assessing possible cell damage after low temperature storage should comprehend objective and appropriate analyses of all the major oocyte attributes, to liberate oocyte cryopreservation from preconceptions or unproven arguments. This paper reviews some of the most significant issues that pertain to the sensitivity of the mature oocyte to the conditions occurring during the thermal transitions between physiological and liquid nitrogen temperatures according to the controlled-rate freezing (slow cooling) procedure. Vitrification, the alternative methodology for the low temperature preservation of cells and tissues, described for the first time by Rall and Fahy (1985), will not be discussed specifically in this review. Births from vitrified oocytes have been described by Kuleshova et al. (1999) and Katayama et al. (2003). Yoon et al. (2003) reported other births from vitrified oocytes, but the overall implantation rate per thawed oocyte (1.7%) was not superior to the one achieved with slow cooling protocols. Recently, Kuwayama et al. (2005) vitrified 64 human oocytes, obtaining very impressive survival, fertilization and blastocyst rates rate (91%, 90% and 62%, respectively). In this study, the pregnancy rate was 43%, an outcome that, if it were possible to reproduce, would represent a landmark progress in oocyte storage. Recently, oocyte cryopreservation with the cryotop vitrification technique has been described to generate high survival (87.2%), fertilization (89.2%), and pregnancy rates (56%), although it should be noted that the majority of pregnancies (9/13) were obtained from oocyte donation cycles (Lucena et al., 2006). Similar data were reported also by Antinori et al. (2007). Oocyte size During controlled rate freezing, the surface-to-volume ratio is a critical factor, influencing the rate of dehydration and therefore the risk of intracellular ice formation. This factor has been deemed to be responsible for the failures of the early attempts of freezing mature oocytes, in the light of the spherical shape and very large size of these cells. In fact, the oocyte geometry is unlikely to be the major reason that makes it not amenable to freezing. The conventional slow cooling protocol, which was originally designed for cleavage-stage embryos, is unable to ensure survival rates in excess of 40 45% when applied to human mature oocytes. Yet interestingly, with the same protocols, it is possible to achieve survival rates ranging around 80% after the freezing and thawing of fully grown germinal vesicle-stage oocytes (Sereni et al., 2000) or pronuclear-stage eggs (Senn et al., 2000). Clearly, during the journey from the breakdown of the germinal vesicle to the formation of pronuclei, some key attributes of the oocyte undergo important changes that can account for such different responses to the same cryopreservation conditions. In line with this hypothesis, Ford et al. (2000) showed that, in rat oocytes, the oolemma becomes less permeable to water during maturation. This is concomitant with the disappearance of the aquaporin-9 transcript. This mrna encodes for a member of the family of membrane-bound proteins described as a broad selective channels, responsible for solute and water transfer. Dynamic changes of the oocyte permeability are also suggested by previous studies in which it was observed that the permeability of mouse oocytes increases during fertilization (Jackowski et al., 1980). This change occurs rather rapidly, within 3 h from sperm penetration. So, although the purely physical characteristics of the oocyte definitely play a role in the exchange of water and cryoprotectants between the intra- and extracellular compartments, other more specific factors, such as membrane permeability, concur more decisively to determine cell survival after thawing. Influence of cryoprotectants on oocyte function Intracellular ice formation is perhaps the major, but surely not the exclusive, source of injury during cryopreservation. As a result of osmotic stress and/or chemical toxicity, cryoprotectants can inflict significant damage, for example, to the delicate mechanism of fertilization. Shaw and Trounson (1989) reported that, in mouse oocytes, exposure at 27 C for 5 min to 1.5 M propane-1,2-diol (PrOH), but not dimethyl sulphoxide (DMSO), causes oocyte activation at a rate (over 80%) comparable to that typically achieved with the classical activating agent ethanol. The difference in the effects of the two cryoprotectants is paradoxical considering that DMSO is generally believed to be more cytotoxic. Besides, the effect of PrOH should be considered in the light of the fact the oocytes used in that study were treated at 15 h after human chorionic gonadotrophin injection, i.e. at a stage in which ageing may have enhanced their sensitivity to activating conditions. Full activation caused by PrOH treatment does not occur when oocytes are treated at an earlier time (13 h) after induction (Stachecki et al., 1998a). Concerning human oocytes, Gook et al. (1995) were the first to conclude that culture and exposure to PrOH during the dehydration and rehydration procedure do not induce parthenogenetic activation. Such a conclusion is confirmed by more recent data, where the observed rates of one-pronucleate oocytes following cryopreservation and microinjection are not higher than the percentages normally obtained with fresh oocytes (Borini et al., 2006; Levi Setti et al., 2006). Nevertheless, the hypothesis that PrOH may trigger, at least to some extent, the activation process remains a matter of debate. One of the early cellular manifestations that occurs during fertilization, and is induced illegitimately during activation, is the release of cortical granules. Under physiological conditions, the enzymatic contents of the granules play the important role of modification of the zona pellucida (zona hardening) and prevention of polyspermic fertilization. In their studies, Gook et al. (1995) showed that fluorescence derived from specific cortical granule staining was comparable in fresh and 25

3 26 frozen oocytes, while it was massively reduced in oocytes that were activated deliberately. This evidence has been recently challenged by fluorescence and electron microscopy studies. By testing the same protocol, Ghetler et al. (2006) found that mere exposure to PrOH leads to a significant reduction in cortical granule staining in human oocytes. Ultrastructural analysis confirmed that freezing and thawing induces extensive loss of cortical granules from the cortex of mature oocytes. Transmission electron microscopy data converge on the hypothesis of a perturbing effect of cryopreservation on the mechanism of cortical granule release. Particularly observed was that the population of sub-cortical granules undergoes a decrease in oocytes frozen with PrOH (Nottola et al., 2006). The magnitude of this phenomenon appears to be influenced by the type of protocol. In particular, a low concentration (0.1 mol/l) of the extracellular cryoprotectant sucrose in the freezing solution is associated with a more conspicuous loss of cortical granules in comparison with a higher concentration (0.3 mol/l) of the same cryoprotectant. Data generated in the mouse model indicate that zona hardening as a consequence of cortical granule release may occur after cryopreservation. Carroll et al. (1990) showed that the low rate of fertilization (48.8%) observed in frozen thawed oocytes could be restored to a normal value (87.8%) after zona drilling. Collectively, this evidence lends credit to the notion that the insemination of cryopreserved human oocytes should be pursued via sperm microinjection rather than standard IVF. In effect, almost all the clinical and experimental experiences gained so far have implied the use of intracytoplasmic sperm injection (ICSI). Exception to this rule is represented by a recent report (Li et al., 2005), which suggests that, in oocytes frozen with a slow cooling method, high fertilization rates (in excess of 80%) can be achieved irrespective of the use of ICSI or standard IVF. Consistent with this, in the same study it is described that immunostaining of frozen thawed oocytes does not reveal evidence of the premature release of cortical granules. This work cannot be considered conclusive, especially because of the small number of oocytes subjected to analysis. Yet, it has the merit to challenge the credence that cortical granules are lost after cryopreservation, thereby stimulating novel studies on this matter. The release of cortical granules is one of the manifestations of the more comprehensive mechanism of fertilization. Central to this process is the regulation via intracellular calcium signalling. As soon as gamete fusion occurs, the sperm-derived factor phospholipase C ζ triggers, via the formation of inositol trisphosphate, a periodical release and uptake of Ca 2+ from the smooth endoplasmic reticulum. The oscillations of cytosolic [Ca 2+ ] thereby generated initiate a sequence of events including cortical granule release, exit from metaphase II (MII) arrest mediated by a decrease in maturation promoting factor and microtubule-associated protein kinase activities, extrusion of the polar body II, pronuclear formation and recruitment of maternal mrnas (Ducibella et al., 2002). In such a perspective, the hypothesis that cryopreservation may cause cortical granule release bears the important implication that [Ca 2+ ] i regulation may also be perturbed. This may not only influence the conversion of an oocyte into a zygote as discussed above but also, in principle, affect the expression of the normal pattern of fetal development. In particular it has been observed in the rabbit that amplitude and frequency of Ca 2+ oscillations experimentally applied to pre-ovulatory oocytes, apart from inducing egg activation, influence the morphology and size of the ensuing parthenogenetically developed fetuses (Ozil and Huneau, 2001). Recent evidence provides arguments for the possibility that the oocyte Ca 2+ release mechanism may be affected by cryopreservation. Measuring the relative fluorescence intensity emitted by the Ca 2+ -specific dye Fluo-3, Jones et al. (2004) found that the ability of cryopreserved human oocytes to release Ca 2+ from the intracellular stores after treatment with the Ca 2+ ionophore A23187 was significantly reduced in comparison to unfrozen controls. The authors speculated that such a defect in the Ca 2+ signalling could alter the normal course of fertilization and have downstream consequences for normal development. However, it should be noticed that the slow cooling protocol applied by Jones et al. appear rather inappropriate. In particular, dehydration was conducted by sequential exposure to 1.5 mol/l PrOH for 20 min and 1.5 mol/l PrOH and 0.2 mol/l sucrose for 10 min, i.e. for time periods far in excess of those actually needed. In fact, Paynter et al. (2001) showed that, after 10 min, the dehydration effect of 1.5 mol/l PrOH is completed and further incubation is unnecessary or perhaps detrimental. In a successive study (Paynter et al., 2005), it was also concluded that the dehydration step achieved with 1.5 mol/l PrOH and 0.2 ml/ sucrose does not require more than 5 min. In addition, as far as is known, the post-thaw rehydration procedure, including the two steps in the presence of sucrose alone applied by Jones et al., are rather arbitrary and do not necessarily reflect conditions aimed at alleviating possible osmotic stress. So it may be entirely possible that the conditions tested compromise the oocyte ability to release Ca 2+ from the intracellular store. Nevertheless, the same conditions diverge significantly from those that should be applied in order to minimize potential damage from cryostorage. More recently, Larman et al. (2006) also investigated possible detrimental effects of cryoprotectants and cryopreservation on the physiology of mouse oocytes. By using fluorescence microscopy, intracellular [Ca 2+ ] changes were monitored using the signal emitted by the Ca 2+ -sensitive dye Indo-1 AM. They observed that treatment with DMSO, and to lesser extent ethylene glycol, caused an important and prolonged rise in intracellular [Ca 2+ ] in association with hardening of the zona. According to the authors, this could explain the poor efficiency of cryopreservation, considering the central role of intracellular [Ca 2+ ] i in the fertilization process as well as in much more advanced developmental stages, as discussed above. Yet, similar to the case of the study of Jones et al., the conditions under which the DMSO effects were tested by Larman et al. do not coincide with those identified as more appropriate for mouse oocyte cryopreservation. In particular, these authors monitored changes in [Ca 2+ ] i caused by DMSO exposure at room temperature but, in fact, slow-cooling storage of mouse oocytes with this cryoprotectant requires incubation at 0 4 C. Over two decades ago, Carroll et al. (1993) successfully cryopreserved mouse oocytes with a slow-cooling method involving exposure to DMSO pre-cooled at 0 C. After thawing, these oocytes underwent fertilization after standard IVF and pre- and post-implantation development with rates comparable to those of non-frozen controls, suggesting that, under the conditions of the study, crucial events that are regulated by the [Ca 2+ ] i signalling system, such as cortical granule release and expression of epigenetic information, are not perturbed.

4 In conclusion, while it is conceivable that cryopreservation conditions may influence the mechanism of [Ca 2+ ] i regulation, it remains to be established whether, in all cases, this effect may compromise oocyte physiology, either neutrally or paradoxically beneficial. The experiences conducted by Jones et al. (2004) and Larman et al. (2006) are certainly interesting and stimulating. However, they provide information of limited importance because they are derived from experimental conditions that do not reflect those applied in the most advanced and efficient protocols. On the other hand, the work conducted on the mouse (Carroll et al., 1993) bears the important lesson that, given appropriate conditions, cryopreserved oocytes can be fertilized successfully with standard IVF in the absence of any sort of micromanipulation-assisted insemination. There is no reason to believe that the human oocyte is governed by a different physiology of fertilization, an assumption that should encourage novel studies aiming at adopting standard IVF as the method of choice and circumventing the potentially negative implications of ICSI. Cytoskeletal sensitivity to cryopreservation The effects of low temperatures on the oocyte cytoskeletal organization (Pickering and Johnson, 1987; Pickering et al., 1990; Wang et al., 2001) and the physical stresses to which the whole cell is subjected during the dehydration rehydration and phase transitions of extracellular water (Leibo et al., 1978) are well known. Direct observation of the freezing process has unveiled that the growing ice front distorts the oocyte, an effect that is independent from the osmotic stress imposed by the presence of cryoprotectants in the freezing solution and remains until thawing. This predisposes the oocyte to blebbing of the oolemma and physical damage of intracellular structures (Ashwood-Smith et al., 1988). For such reasons, preservation of the MII spindle in frozen thawed oocytes has raised concerns since the very early days of oocyte cryopreservation. This matter has remained contended over a number of years, with several studies for and against the hypothesis of spindle disruption after cryopreservation (Coticchio et al., 2005). Recent evidence extends the dispute. Mullen et al. (2004) exposed mature human oocytes to anisosmotic conditions generated by using buffer solutions including different concentrations of sucrose, with the aim of simulating the osmotic stress occurring during slow cooling and ascertaining possible consequences to the MII spindle structure. The authors found that both hypo- and hyperosmotic conditions caused a decrease in the proportion of oocytes with a normal spindle, in extreme cases with a total loss of a recognizable microtubular structure. However, these authors did not include penetrating cryoprotectants in their treatments, which may influence microtubule dynamics (Pickering and Johnson, 1987). Recently, evidence from the authors study centre was reported that argues against the loss of normal spindle organization after cryostorage (Coticchio et al., 2006). In fact, by using confocal microscopy, partial disruption of the MII spindle and associated chromosomes was found only in coincidence with slow-cooling conditions that were showed to be inadequate for oocyte storage (Borini et al., 2004). However, it is known that a protocol adopting a higher sucrose concentration in the freezing solution (Fabbri et al., 2001) promotes the retention of an intact chromosome segregation apparatus comparable in incidence to freshly collected oocytes. Ultrastructural analysis (Nottola et al., 2006), on oocytes frozen with the same high sucrose concentration protocol, established also that the MII spindle was regularly barrel-shaped, anastral, with slightly pointed poles and properly positioned in the ooplasm of both fresh and frozen thawed oocytes. The suggestion that the structure of the MII spindle can remain unaltered after cryopreservation given appropriate conditions converges with clinical experience. Should the MII spindle be significantly affected by the freezing procedure, flawed chromosome segregation at meiosis II would occur at high frequency and the aneuploidy condition produced in the ensuing embryo would doubtless increase the proportion of early pregnancy loss. In fact, the same cryopreservation conditions can generate pregnancies whose abortion rate is comparable to the one normally observed in pregnancies from fresh cycles (Borini et al., 2006). Low abortion rates in pregnancies from frozen oocytes have been confirmed in relation to the use of another slow-cooling protocol (Bianchi et al., 2007), whose effects on the MII spindle are currently under investigation. Another important component of the oocyte cytoskeletal apparatus is the cortical meshwork of actin filaments, whose integrity is essential for the cytokinesis events, including polar body II extrusion and the first cleavage after fertilization. Evidence of possible effects of cryostorage on actin filaments is virtually lacking. However, the study of Younis et al. (1996) that described a loss of polymeric actin in rhesus mature oocytes after exposure to glycerol at 20 C should be considered with some concern, although this particular cryoprotectant is not normally used for oocyte cryopreservation. Should cryopreservation give rise to disruption to cortical actin, the predictable consequences would correspond to retention of the polar body II and inhibition of cleavage after fertilization. In effect, an increased frequency in three-pronuclear fertilization in frozen thawed mouse oocytes (Carroll et al., 1989) and a rise in cleavage arrest in fertilized human eggs (Hunter et al., 1995) have been reported after storage with inappropriate protocols. L De Santis and others (personal communication) using fluorescence microscopy and phalloidin staining, have observed that, in oocytes cryopreserved with a high sucrose concentration protocol (Fabbri et al., 2001), the sub-cortical actin meshwork was unchanged in comparison to control oocytes. Also, data on the developmental performance of frozen thawed oocytes are in agreement with the conservation of a fully functional subcortical actin ring after cryopreservation. In two separate series of treatments with frozen oocytes (Borini et al., 2006; Bianchi et al., 2007), a low incidence of one- or three-pronuclear fertilization (5.8% and 6.3% in the two studies, respectively) and a high cleavage rate (90.2% and 93.7%, respectively) have been reported. These frequencies are indistinguishable from those normally generated with fresh oocytes. Sensitivity to high solute concentration during slow cooling During slow cooling, intracellular ice formation is essentially prevented by cell dehydration. This can be achieved in a controlled fashion by the strength of the osmotic gradient 27

5 28 imposed by the concentration of cryoprotectant in the freezing solution. Another, not less important, phase of dehydration occurs during the progressive conversion of liquid water into ice, a process that leads to the concentration of solutes, mainly sodium, in the unfrozen fraction of the system, a condition that draws further water out of the cell. Sufficient dehydration at this stage is guaranteed, provided that the cooling rate and consequent ice formation are sufficiently slow. However, apart from such a beneficial effect, high solute concentration may also be the source of biochemical toxicity. In particular, the stability of intracellular membranes and the plasmalemma are believed to be particularly sensitive to high solute concentration. It should be noticed, nevertheless, that cleavage-stage embryos can be safely cryopreserved with sodium-rich solutions. Because sodium is the most represented solute in freezing mixtures, it has been suggested that replacing this element with the cation choline, which does not diffuse through the plasmalemma, could significantly reduce the detrimental effects of high solute concentration on oocyte viability. On this basis, in a slow-cooling study conducted on mouse oocytes, Stachecki et al. (1998b) investigated the effects of replacing increasing proportions of sodium with equimolar amounts of choline. The authors found that the rates of survival, fertilization and development were very poor in the presence of high sodium concentration, but were maximal after replacing sodium with choline. The association between the use of choline-based media and high post-thaw survival has been subsequently expanded and confirmed with other experiments in the mouse (Stachecki and Willadsen, 2000). Choline-based media proved themselves objectively superior to conventional freezing formulation. Nevertheless, the real importance of the solution effect in oocyte cryopreservation is not entirely clarified. In particular, it is interesting to note that, by using sodium-rich media for cryopreserving mouse oocytes, Carroll et al. (1993) obtained remarkable rates of survival, fertilization and pre- and postimplantation, although overall moderately lower than those achieved by Stachecki et al. (2002) with choline-based freezing solutions. This suggests that the detrimental effects of high sodium concentration during slow cooling do not occur systematically but rather depend on, or are enhanced by, other conditions. For instance, a major difference in the protocols employed by Carroll et al. (1993) and Stachecki et al. (1998b) is the type of penetrating cryoprotectant, DMSO in the former case and PrOH in the latter. Therefore, while the solution effect is a well-recognized drawback of the slow-cooling strategy, nevertheless its relative importance in determining post-thaw cell viability may vary. Recently, low-sodium media have been tested by Stachecki et al. (2006) on human oocytes, with results that are difficult to assess and that also raise some questions. The author reported that media based on choline and 0.2 mol/l sucrose can guarantee survival rates as high as 90%. However, it is difficult to appreciate such an outcome, considering that the treatment groups of this study included no more than oocytes. In addition, no control groups frozen with sodium-rich media were included. Therefore, it cannot be concluded that cholinebased media have a genuinely beneficial effect on the viability of human oocytes. Lastly, after thawing and fertilization, even the choline-based freezing mixture that produced the highest survival rate generated a low rate of cleavage (78%). On the contrary, as discussed above, various conventional sodium-rich freezing media ensure cleavage rates (90 93%) that appear unchanged compared with unfrozen control (Borini et al., 2006; Levi Setti et al., 2006). Sodium-depleted media have also been recently applied in clinical IVF (Boldt et al., 2006). Survival rates were suboptimal (61% and 59%), irrespective of the adoption of phosphate - buffered saline (PBS) or HEPES as ph buffers in separate treatment groups. In the PBS group, fertilization after ICSI was also inadequate (55.8%), while cleavage rates were not reported for either group. Based on implantation rates of 10.6% and 15.9% obtained with the PBS and HEPES groups, respectively, the authors concluded that sodium-depleted media is beneficial to the cryopreservation of human oocytes. It should be considered, however, that the freezing solution formulation adopted by Boldt et al. (2006) was not the same as the one successfully developed with mouse oocytes by Stachecki et al. (2002). In fact, the cryopreservation method applied by Boldt et al. (2006) included other changes in comparison with more traditional slow-cooling protocols. Firstly, dehydration was performed in a single step with a freezing mixture containing PrOH and sucrose. Secondly, post-thaw rehydration was carried out with a single solution containing a high sucrose concentration (0.5 mol/l), but no decreasing concentrations of PrOH. Individually or in association, these two protocol modifications, rather than the use of choline, may have produced the observed outcome. Thawing conditions So far, attempts to improve oocyte cryopreservation protocols have involved changes mainly limited to the dehydration steps required before freezing or the composition of the freezing mixtures. Limited attention has been given to the optimization of post-thaw rehydration conditions (Tao et al., 1995; Stachecki and Willadsen, 2000). In reality, rehydration is likely to be as important as other steps of the entire procedure, or perhaps even more. Removal of the intracellular cryoprotectant and reestablishment of the original water content is an obvious source of osmotic stress. This occurs after exposure of the oocyte to the very challenging conditions of freezing and thawing. Such treatments could sensitize the cell and make it particularly vulnerable. Nothing is known on the magnitude and dynamics of the oocyte volume excursions imposed by the diverse thawing solutions. Currently, applied dehydration conditions for oocyte freezing are generally derived from protocols designed for cleavage-stage embryos and are likely to be rather inappropriate. It is not a coincidence that an important proportion of oocytes that appear intact immediately after thawing and release from the storage device afterwards degenerate during rehydration. Therefore, rehydration conditions should be put under rigorous scrutiny in a fashion similar to the investigations conducted on the pre-freezing dehydration phase (Paynter et al., 2001, 2005). Conclusions For many years, oocyte cryopreservation has been considered with little enthusiasm and even scepticism. The oocyte characteristics are surely unique, making this cell not amenable to freezing. Critical changes in the plasmalemma occurring during the various functional stages regulate the passage of water and small molecules (including cryoprotectants) between the intra- and extracellular compartment, influencing the rate

6 of survival after freezing and thawing in final analysis. Under particular conditions, cryoprotectants make storage possible, but are themselves a potential source of cell damage. Their toxicity may depend on the species considered, making findings obtained in animal models not always applicable to the human. For example, PrOH causes substantial activation in aged mouse oocytes, but can activate aged human oocytes only to a rather limited extent. This should be taken into account to appreciate the relative importance and implications of activation-related events, such as [Ca 2+ ] i changes, which will most likely occur in all mammalian oocytes following cryoprotectant treatment but whose outcome may differ between species. Cortical granules are not always released, indicating that insemination via microinjection could be avoided. Observations made here argue against a high risk of chromosome and meiotic spindle disorganization, under appropriate storage conditions. This should clear the ground from one of the major arguments that has fuelled concerns about an increased risk of aneuploidy in frozen oocytes. Cryodamage has been postulated to be derived from the increase in solute concentration during freezing. This appears entirely plausible and efforts aiming at clarifying such a possibility should be encouraged. However, the experiences reported in the human so far have not demonstrated that the solution effect has a major impact on oocyte survival and viability. Protocol optimization should be intended as an extensive endeavour. It is unlikely that decisive improvements will originate from individual changes. Rather, multiple approaches should be undertaken to establish better overall freezing conditions. In such a perspective, rehydration after thawing should be assessed with particular interest. References Al-Hasani S, Diedrich K, van der Ven H et al Cryopreservation of human oocytes. Human Reproduction 2, Antinori M, Licata E, Dani G et al Cryotop vitrification of human oocytes results in high survival rate and healthy deliveries. Reproductive BioMedicine Online 14, Ashwood-Smith MJ, Morris GW, Fowler R et al Physical factors are involved in the destruction of embryos and oocytes during freezing and thawing procedures. Human Reproduction 3, Bianchi V, Coticchio G, Distratis V et al Differential sucrose concentration during dehydration (0.2 mol/l) and rehydration (0.3 mol/l) increases the implantation rate of frozen human oocytes. Reproductive BioMedicine Online 14, Boldt J, Tidswell N, Sayers A et al Human oocyte cryopreservation: 5-year experience with a sodium-depleted slow freezing method. Reproductive BioMedicine Online 13, Borini A, Sciajno R, Bianchi V et al Clinical outcome of oocyte cryopreservation after slow cooling with a protocol utilizing a high sucrose concentration. Human Reproduction 21, Borini A, Bonu MA, Coticchio G et al Pregnancies and births after oocyte cryopreservation. Fertility and Sterility 82, Carroll J, Wood MJ, Whittingham DG 1993 Normal fertilization and development of frozen thawed mouse oocytes: protective action of certain macromolecules. Biology of Reproduction 48, Carroll J, Depypere H, Matthews CD 1990 Freeze thaw-induced changes of the zona pellucida explains decreased rates of fertilization in frozen thawed mouse oocytes. Journal of Reproduction and Fertility 90, Carroll J, Warnes GM, Matthews CD 1989 Increase in digyny explains polyploidy after in-vitro fertilization of frozen thawed mouse oocytes. Journal of Reproduction and Fertility 85, Chen C 1986 Pregnancy after human oocyte cryopreservation. Lancet 1, Coticchio G, De Santis L, Rossi G et al Sucrose concentration influences the rate of human oocytes with normal spindle and chromosome configurations after slow-cooling cryopreservation. Human Reproduction 21, Coticchio G, Bonu MA, Bianchi V et al Criteria to assess human oocyte quality after cryopreservation. Reproductive BioMedicine Online 11, Ducibella T, Huneau D, Angelichio E et al Egg-to-embryo transition is driven by differential responses to Ca 2+ oscillation number. Developmental Biology 250, Fabbri R, Porcu E, Marsella T et al Human oocyte cryopreservation: new perspectives regarding oocyte survival. Human Reproduction 16, Ford P, Merot J, Jawerbaum A et al Water permeability in rat oocytes at different maturity stages: aquaporin-9 expression. Journal of Membrane Biology 176, Ghetler Y, Skutelsky E, Ben Nun I et al Human oocyte cryopreservation and the fate of cortical granules. Fertility and Sterility 86, Gook DA, Osborn SM, Johnston WI 1995 Parthenogenetic activation of human oocytes following cryopreservation using 1,2-propanediol. Human Reproduction 10, Hunter JE, Fuller BJ, Bernard A et al Vitrification of human oocytes following minimal exposure to cryoprotectants; initial studies on fertilization and embryonic development. Human Reproduction 10, Jackowski S, Leibo SP, Maxur P 1980 Glycerol permeabilities of fertilized and infertilized mouse ova. Journal of Experimental Zoology 212, Jones A, Van Blerkom J, Davis P et al Cryopreservation of metaphase II human oocytes effects mitochondrial membrane potential: implications for developmental competence. Human Reproduction 19, Katayama KP, Stehlik J, Kuwayama M et al High survival rate of vitrified human oocytes results in clinical pregnancy. Fertility and Sterility 80, Kuleshova L, Gianaroli L, Magli C et al Birth following vitrification of a small number of human oocytes: case report. Human Reproduction 14, Kuwayama M, Vajta G, Kato O et al Highly efficient vitrification method for cryopreservation of human oocytes. Reproductive BioMedicine Online 11, Larman MG, Sheehan CB, Gardner DK 2006 Calcium-free vitrification reduces cryoprotectant-induced zona pellucida hardening and increases fertilization rates in mouse oocytes. Reproduction 131, Leibo SP, McGrath JJ, Cravalho EG 1978 Microscopic observation of intracellular ice formation in unfertilized mouse ova as a function of cooling rate. Cryobiology 15, Levi Setti PE, Albani E, Novara PV et al Cryopreservation of supernumerary oocytes in IVF/ICSI cycles. Human Reproduction 21, Li XH, Chen SU, Zhang X et al Cryopreserved oocytes of infertile couples undergoing assisted reproductive technology could be an important source of oocyte donation: a clinical report of successful pregnancies. Human Reproduction 20, Lucena E, Bernal DP, Lucena C et al Successful ongoing pregnancies after vitrification of oocytes. Fertility and Sterility 85, Mullen SF, Agca Y, Broermann DC et al The effect of osmotic stress on the metaphase II spindle of human oocytes, and the relevance to cryopreservation. Human Reproduction 19, Nottola SA, Macchiarelli G, Coticchio G et al Ultrastructure of human mature oocytes after slow cooling cryopreservation using different sucrose concentrations. Human Reproduction 22, Ozil JP, Huneau D 2001 Activation of rabbit oocytes: the impact of the Ca2+ signal regime on development. Development 128, Paynter SJ, Borini A, Bianchi V et al Volume changes of mature 29

7 human oocytes on exposure to cryoprotectant solutions used in slow cooling procedures. Human Reproduction 20, Paynter SJ, O Neil L, Fuller BJ et al Membrane permeability of human oocytes in the presence of the cryoprotectant propane-1,2- diol. Fertility and Sterility 75, Pickering SJ, Johnson MH 1987 The influence of cooling on the organization of the meiotic spindle of the mouse oocyte. Human Reproduction 2, Pickering SJ, Braude PR, Johnson MH et al Transient cooling to room temperature can cause irreversible disruption of the meiotic spindle in the human oocyte. Fertility and Sterility 54, Rall WF, Fahy GM 1985 Ice-free cryopreservation of mouse embryos at 196 degrees C by vitrification. Nature 313, Senn A, Vozzi C, Chanson A et al Prospective randomized study of two cryopreservation policies avoiding embryo selection: the pronucleate stage leads to a higher cumulative delivery rate than the early cleavage stage. Fertility and Sterility 74, Sereni E, Bonu MA, Borini A et al High survival rate after cryopreservation of human prophase I oocytes. Fertility and Sterility 74 (Suppl. 1), S161. Shaw JM, Trounson AO 1989 Parthenogenetic activation of unfertilized mouse oocytes by exposure to 1,2-propanediol is influenced by temperature, oocyte age, and cumulus removal. Gamete Research 24, Stachecki JJ, Willadsen SM 2000 Cryopreservation of mouse oocytes using a medium with low sodium content: effect of plunge temperature. Cryobiology 40, Stachecki JJ, Cohen J, Garrisi J et al Cryopreservation of unfertilized human oocytes. Reproductive BioMedicine Online 13, Stachecki JJ, Cohen J, Schimmel T et al Fetal development of mouse oocytes and zygotes cryopreserved in a nonconventional freezing medium. Cryobiology 44, Stachecki JJ, Cohen J, Willadsen SM 1998a Cryopreservation of unfertilized mouse oocytes: the effect of replacing sodium with choline in the freezing medium. Cryobiology 37, Stachecki JJ, Cohen J, Willadsen S 1998b Detrimental effects of sodium during mouse oocyte cryopreservation. Biology of Reproduction 59, Tao J, Du J, Kleinhans FW et al The effect of collection temperature, cooling rate and warming rate on chilling injury and cryopreservation of mouse spermatozoa. Journal of Reproduction and Fertility 104, Tiitinen A, Hyden-Granskog C, Gissler M 2004 What is the most relevant standard of success in assisted reproduction?: The value of cryopreservation on cumulative pregnancy rates per single oocyte retrieval should not be forgotten. Human Reproduction 19, Wang WH, Meng L, Hackett RJ et al Limited recovery of meiotic spindles in living human oocytes after cooling rewarming observed using polarized light microscopy. Human Reproduction 16, Whittingham DG 1977 Fertilization in vitro and development to term of unfertilized mouse oocytes previously stored at 196 degrees C. Journal of Reproduction and Fertility 49, Yoon TK, Kim TJ, Park SE et al Live births after vitrification of oocytes in a stimulated in-vitro fertilization embryo transfer program. Fertility and Sterility 79, Younis AI, Toner M, Albertini DF et al Cryobiology of non-human primate oocytes. Human Reproduction 11, Paper based on contribution presented at the Tecnobios Procreazione Symposium 2006 and 2nd International Conference on the Cryopreservation of the Human Oocyte in Bologna, Italy, 5 7 October Received 19 February 2007; refereed 15 March 2007; accepted 30 April