Nuclear maturation

Oocyte nuclear maturation comprises nuclear modifications that take place during resumption of meiosis producing a haploid chromosome complement from the previous diploid state. Oocytes are arrested at the prophase I displaying a visible nucleus referred to as germinal vesicle (GV). When meiosis resumes, the oocytes undergo germinal vesicle breakdown (GVBD) i.e., dissolution of the nuclear envelope. Chromatin condenses where then homologues chromosomes are paired aligning on the meiotic spindle at metaphase I. During anaphase and telophase bivalents separate as the homologous chromosomes separate and separation is complete when metaphase II occurs, recognizable by presence of 1st polar body (PB). Meiosis is completed following successful fertilization visualized as the presence of 2nd PB. The legal nomenclature in several countries defines the resulting zygote or pre-embryo formation strictly to when the 2 pronuclei dissolves (i.e., the chromosomes become localized centrally at syngamy or synkaryosis).

Cytoplasmic maturation

The regulatory signals for the molecular processes involved in oocyte growth and maturation are concerted during a sophisticated 2-way cell communication between the gamete and the surrounding somatic cells of the follicle in response to circulating levels of gonadotropins and based on a developmental program intrinsic to the oocyte. Mammalian oocyte maturation is a very complex process involving the resumption of the meiotic cell cycle as well as cytoplasmic modulations including organelle migration and molecular changes. The acquisition of developmental competence involves the synthesis and storage of a wide range of molecules during oocyte growth followed by reprogramming and timely utilization of these stored messages during final maturation, fertilization and early embryogenesis.

The endpoint of oocyte maturation In Vivo is the release from the follicle of a mature (metaphase II) gamete, that upon sperm penetration, activation and maternal-embryonic transition of genome activation can support normal embryonic and fetal development to term and the birth of live offspring. Cytoplasmic maturation is shortly described as the unity of processes modifying the oocyte cytoplasm that are essential for fertilization and pre-implantation embryonic developmental competence. It comprises production and presence of specific factors, the relocation of cytoplasmic organelles and post-transcriptional modifications of mRNAs accumulated during oogenesis. Thus some elements of cytoplasmic maturation can be visualized (e.g. the line-up of cortical granules) whereas many other elements of cytoplasmic maturation are molecular and very challenging to visualize or monitor. Ultimately the quality and completeness of cytoplasmic maturation can be viewed as the developmental potential of the resulting pre-emplantation embryo.

FF-MAS

Among the substances that have been reported to influence oocyte maturation there is a promising endogenous signaling molecule: FF-MAS (Follicle Fluid-Meiosis Activating sterol, 4,4-dimethyl-5 α -cholest-8,14,24-trien-3 β -ol), an intermediate in the cholesterol biosynthetic pathway present in all cells. FF-MAS has been known for some time to be able to trigger the resumption of meiosis in oocytes from mice In Vitro. This review gives special focus to FF-MAS, the effects seen in animal and human studies so far and its potential use in treatment of human infertility is being discussed, including both the safety and efficacy issues that need to be addressed.

A. Ultrastructural changes and endocrine dynamics during meiotic maturation In Vivo in the equine oocytes (Paper I)

This study was undertaken to describe the ultrastructure and the endocrine dynamics of the equine maturing oocyte In Vivo. The In Vivo equine oocyte maturation has not previous been described in details presumably due to the difficulties associated with obtaining this material.

Thus, the ultrastructural characteristics of the resumption of meiosis In Vivo of equine oocytes was subject for our investigation and several unique findings have been documented for the equine species (Grøndahl, et al. 1995, I). Oocytes and follicular fluid was aspirated at different time-points throughout the follicle growth phase until very near ovulation to document the In Vivo maturation process. Oocytes were subjected to fine structure morphological investigation and follicle fluid to biochemical measurements. In summary, the final maturation of equine oocytes includes a series of well-defined nuclear changes (oocyte nucleus migration, dissolution of the nuclear envelope, extrusion of the first polar body) paralleled by defined cytoplasmic changes (disruption of cumulus-oocyte intermediate junctions and trafficking of mitochondria to a central position). Moreover, a massive build up of cortical granules was noted as a significant element of equine cytoplasmic maturation quite similar to what has been observed in human pre-ovulatory oocytes In Vivo (Sathananthan, 1994). In the equine follicle milieu a sharp rise in Progesterone (P4) was noted to be highly correlated with fully mature metaphase-II oocyte inside the follicle and a high degree of synchrony between nuclear and cytoplasmic markers of full maturity. The concentration of Estradiol-17-beta (E2) was observed to be constantly elevated throughout the period and did not drop prior to ovulation.

These findings are in contrast to the steroid content in pre-ovulatory follicles of sheep (Hay and Moor, 1975), cattle (Dieleman et al. 1983) and human (Lumsden et al. 1986) where marked declines in E2 are seen prior to ovulation. These findings could be important in the design of future composition of In Vitro culture media for equine and other species.

Fundamentally differences have been observed between mammalian species in the oocytes development, biochemistry and physiology (Zamboni et al. 1972; Dvoràk et al. 1982; Van Blerkom, 1989; 1991; Sathananthan, 1984; 1994), however, also many common themes have been identified. Of pivotal differences between mouse and human oocytes can be mentioned the paternal centrioles/centrisomes, which are inherited by the early embryo and actively participate in the mitotic cell cycle in human zygote which is quite unlike the mouse embryo (Sathananthan, 1994). Kinetics of oocyte maturation is also vastly different among species. Approximately 40% of rodent and bovine oocytes will resume meiosis within 1-2 hours as measured by visible initiation of nuclear envelope breakdown after release from the follicle (Anderiesz et al. 2000a), whereas human, equine and primate oocytes resume spontaneous meiosis at a very low rate and with a difference in kinetics compared to other species (e.g., the human egg has been reported to still be in GV stage after 24 hours of culture in sharp contrast to e.g. mouse oocytes) (Edwards, 1965; Trounson et al. 1994a; Grøndahl et al. 1995 (I); Grøndahl et al. 1997 (III); Grøndahl et al. 1998 (IV); Sathananthan, 1994; Anderiesz et al. 2000b; Grøndahl et al. 2000a (V)). A common theme however seems to be that the process of oocyte maturation In Vitro of mammalian oocytes once initiated progress faster than the physiological process In Vivo (as we have seen in both equine, mice and humans) which creates further requirements on the completeness of the cytoplasmatic processes In Vitro.

B. Fertilization and activation of the embryonic genome in equine zygotes (Paper II)

These studies were undertaken to characterize, in the equine species, one of the most important pre-implantation embryo development phases namely the timing of the maternal-embryonic transition (MET).

Pre-implantation embryos were surgically obtained from mares where the insemination and ovulation timing were closely monitored. Subsequently, the pre-embryos were subjected to short In Vitro culture (labelled Uridine culture) as well as to ultrastructural morphological investigation.

In summary, we have found in the equine In Vivo derived embryo from naturally cycling mares the maternal to embryonic transition was first established to occur in 6-8 cell embryos of the 4th cell cycle and clearly paralleled with significant ultrastructural features of the nucleolus apparatus turning from the inactive stage of nucleolus precursor bodies to fully RNA-synthesis active reticulated nucleoli (Grøndahl et al. 1993; Grøndahl and Hyttel, 1996, Paper II). In humans this transition has similarly been observed to occur between the 4- to 8-cell stage, thus considerable different from mice (Tesarik et al. 1986; Braude et al. 1988; Sathananthan, 1994).

The ability to progress through the natural transition of maternally controlled development to that of embryonic governance is acquired during oocyte maturation and early development. Fertilization of the mammalian oocytes results in completion of the meiotic process that has been initiated at the time of oocyte formation in fetal life. Mammalian pre-implantation embryogenesis is initially dependent upon maternally-inherited molecules during early development (Mclaren, 1981). Upon sperm activation of the ooplasm inactivation of MPF occurs allowing the cell cycle to progress through the arrest at metaphase-II. Subsequently, the important cortical reaction (release of sub-zonal cortical granules) takes place that ideally prevent the situation of polyspermia to happen (Szöllösi, 1967; Szöllösi and Hunter, 1973; Ducibella et al. 1990; 1993; Ducibella, 1996). The acquired ability of the oocyte to release Ca 2+ in response to the fertilizing spermatozoa constitute an important competency step during oocyte maturation, for review see (Carroll et al. 1996). The frequency and amplitudes of calcium waves (and thereby the quality of the signal) are dependent upon the maturational stage of the oocyte and are determent for embryo quality (Swann and Lawrence, 1996; Herbert et al. 1997). As example the rate of calcium release in response to fertilization is four times greater in fully mature oocytes compared to maturing oocytes (Carroll et al. 1996).

The very important developmental check-point, transition from maternal to embryonic genome transcription (MET), occurs at species specific cell-cycle and is pivotally influenced by transcripts and factors during oocyte maturation and early embryonic development (Telford et al. 1990; Hwang et al. 2001). In the mouse there is evidence that a major burst of embryonic transcription occurs at the 2-cell stage and that these transcripts are essential for continued development (Kidder and McLachlin, 1985). Transcription may already occur at a very low rate from the 1- or 2-cell stage in other mammalian species as reported in cattle (Plante et al. 1994; Hay-Schmidt et al. 2001), however the major burst of transcription occurs later at species specific stages of development as reported in human (Tesarik et al. 1986), bovine (Camous et al. 1986; Grøndahl et al. 1990) and comparatively reviewed by Telford (Telford et al. 1990). Also in non-human primates (rhesus monkeys) MET has been identified to occur between the 4-8 cell-stage. In rhesus monkeys inadequate cytoplasmic maturation In Vitro has led to poor developmental potential and impairments of the embryonic genome activation (Schramm et al. 2003). Early cleavage arrest has been reported in most species studied including man and the need for adequate changes in protein transcripts around the time of maternal-embryonic transition has been firmly substantiated (Davis, 1985; Crosby et al. 1988; Frei et al. 1989; Heikinheimo et al. 1995; Moor et al. 1998).

Our findings in the equine species are the first to describe this process in this rarely studied species; findings that may well be important and used in designing the optimal culture conditions for embryo culture as well as guidance to timing of back-transfer to the uterus to best mitigate the risk of early cleavage arrest In Vitro.

C. ICSI fertilization of equine IVM oocytes
(Paper III)

This study was undertaken in order to improve and pioneer In Vitro fertilization methods of the equine oocytes since capacitation of the equine sperm has proven particular difficult hence normal IVF has been challenging in the horse and consequently the comparative knowledge in this species is limited.

Intracytoplasmic sperm injection (ICSI) was performed on equine oocytes matured In Vitro. Oocyte were aspirated from abattoir ovaries and matured In Vitro for 36 hours in TCM-199 medium according to method previously described by Grøndahl et al. (1995b).

In summary, we found that when employing ICSI to equine In Vitro matured oocytes it was noted that the pronuclear formation was obtained in a similar manner and with normal kinetics as observed In Vivo maturation and In Vivo fertilization (Grøndahl et al. 1993; Grøndahl et al. 1997, III). However, the oocyte activation was still incomplete and even though cleavage was obtained further embryonic development was largely compromised. This is thought to be a sign of inadequate quality M-II following the In Vitro maturation. In mice ICSI has been particularly troublesome (Markert, 1983; Ron-El et al. 1994) until Kimura and Yanagimachi (Kimura and Yanagimachi, 1995) reported a breakthrough using piezo-driven injection techniques highlighting that physical properties of the mouse oocyte make them leak after sperm injection.

Historically, the rapidly evolving techniques of IVF since the essential work of Steptoe and Edwards (Steptoe and Edwards, 1978) have been very successful in humans as well as moderate successful in several farm animals as reviewed by Trounson (Trounson et al. 1994a) and Edwards (Edwards, 1995). Intracytoplasmic injection of spermatozoa into the ooplasm of the oocyte (ICSI) has dramatically improved the chances of treatment of male infertility (Palermo et al. 1992; Van Steirteghem et al. 1993), despite the fact that this technique bypasses the normal gamete interactions like sperm-egg fusion and membrane hyper-polarisation. Nevertheless the ICSI injections spurs calcium oscillations to occur which is followed by activation of the egg leading to fairly high percentage of two-pronuclear eggs (Tesarik et al. 1994). This technique has had variable success in domestic species, cow (Younis et al. 1989; Goto et al. 1990), comparative studies in different species (Iritani, 1991) and equine (Squires et al. 1996). We contributed to the development of the ICSI technique in the equine species in the form of the present paper (III). Today ICSI is widely used in the human clinic and at some IVF centers up to 50% of procedures for infertile couples make use of ICSI. These techniques have noticeably never been through thorough safety investigations (e.g. chromosome soundness evaluation) before the widespread use in the human clinic and, consequently, the involved safety of the ICSI procedure and wellbeing of the resulting children has exclusively been and is still monitored as retrospective follow-up studies.

Only if the quality and success rates of IVM of equine oocytes dramatically improve it is likely that this ICSI technique will be of use for this variant of husbandry breeding.

D. Pharmacological effect of MAS sterols on
In Vitro Maturation (Papers IV & VIII)

This study was primarily undertaken to investigate the pharmacology of synthetic FF-MAS on murine oocytes and establish a classical dose-response curve as well as testing related oxysterols effect on oocyte maturation.

Follicular fluid meiosis-activating sterol (FF-MAS, 4,4-dimethyl-5 α -cholest-8,14,24-trien-3 β -ol) has been known for some time to be able to trigger the resumption of meiosis in oocytes from mice In Vitro (Byskov et al. 1995), however, the mechanism is largely unknown. FF-MAS (and the close metabolite, Testis meiosis-activating sterol, T-MAS) are intermediates in the cholesterol biosynthetic pathway and present in all cholesterol-producing cells (see Figure 1 ).

Since this original observation, numerous studies using follicle derived or synthetic FF-MAS have been carried out to observe FF-MAS effectiveness to induce oocyte maturation.

We found that FF-MAS in a dose-dependant manner can induce resumption of meiosis of murine oocytes cultured In Vitro under different meiosis inhibiting principles (Hypoxanthine, IBMX, dbcAMP) and could be consistent with a physiological role for FF-MAS in oocyte maturation (Grøndahl et al. 1998 (IV)).

Moreover we published in a separate study that the range of meiotic active endogenous molecules are far greater than just FF-MAS and T-MAS and seen in this illustration from Grøndahl et al. (2001). We identified the chemical structure of 8 endogenous MAS molecules (see Figure 2 ) that are approximately equipotent in micromolar activity in their capacity to induce nuclear maturation (i.e. meiosis) in naked mouse oocytes.

Other groups have investigated the pharmacological effects (i.e. physiological or supra physiological doses interacting In Vitro) of FF-MAS (Hegele-Hartung et al. 1999; Hegele-Hartung et al. 2001; Downs et al. 2001; Tsafriri et al. 2001; Griffin et al. 2004) and confirmed the nuclear maturation activity in the low micromolar range (1-10 micromolar). Initially it was hypothesized that the primary effect of MAS was that of the pivotal meiosis (i.e. nuclear maturation) inducing signal (Byskov et al. 1995; Byskov et al. 1998; Byskov et al. 1999; Andersen et al. 1999). Many groups have confirmed the effect of MAS sterols on oocyte nuclear maturation. Our group has later focused on FF-MAS effect on oocyte quality and several studies are pointing towards the hypothesis that MAS sterols are important molecules for cytoplasmic maturation and the quality of the oocyte and subsequent embryo potential (a discussion that will be further substantiated in later sections). Even though FF-MAS in several models has been shown to induce nuclear maturation it is still debated whether FF-MAS constitute a pivotal meiosis mediating signal downstream from gonadotropins (Downs et al. 2001; Baltsen, 2001; Tsafriri et al. 2002; Grøndahl et al. 2003, X). Therefore, the potential therapeutic value of MAS based therapeutic are foremost thought to be in the field of treatment of human infertility. Possible uses could arise in the potential field of novel non-hormonal contraception if the postulation of MAS being an obligatory molecule for meiosis can be further substantiated and safe and non-toxic MAS antagonist can be discovered. In conclusion, a proven pharmacological effect on meiosis has been observed both by us and other groups establishing 0.1 micromolar as non-effective dose, 0.7-1 micromolar as the lowest effective dose (see Figure 3 ) and 20 micromolar as the highest efficacy dose with doses of 50 or 100 micromolar being supra-physiological and approaching toxicity limits (Grøndahl et al.1998; unpublished data Grøndahl et al.).

These findings have assisted us and others to guide the dose range of synthetic FF-MAS used for subsequent clinical studies in humans considering 0.1 micromolar FF-MAS as an ineffective dose and the most interesting active dose-range to be 1-10 micromolar FF-MAS. Fifty or 100 micromolar has been observed to have negative effects on oocyte homeostasis and on meiosis (IV; VIII, XIII)

E. Effect of FF-MAS on human oocytes from women diagnosed as having Polycystic Ovarian Syndrome (PCOS) (Paper V)

This study was undertaken as a pilot clinical study to get important knowledge on how the human oocytes reacted to FF-MAS and to get some initial data on both kinetics and pharmacology of synthetic FF-MAS formulated in an ethanol-based formulation. Hence our group was the first to investigate and report the use of synthetic FF-MAS for improving human IVM with the aim to optimize culture design and media with regards to In Vitro culture of immature human oocytes aspirated from small antral follicles in patients diagnosed with polycystic ovarian syndrome (PCOS) (V). Numerous groups have been reporting on successful IVM on PCOS patients, however still with low clinical efficiency (Trounson et al.1994b; Beckers et al. 1999; Chian et al. 1999; 2000; Cha et al. 2000; Chian et al. 2000, Mikkelsen and Lindenberg 2001; Lin et al. 2003).

In collaboration with a Danish IVF clinic that obtained approval from both the local (Copenhagen) and central (Danish) scientific-ethical committees to perform immature oocyte retrievals in patients diagnosed with polycystic ovarian syndrome (PCOS) the objective was formulated to optimize the culture conditions including exploring the use of synthetic FF-MAS. Importantly FF-MAS exposed oocytes were not attempted fertilized or otherwise being part of a clinical treatment of patients.

The patients were asked to give their written consent to donate (randomly chosen) half of the aspirated oocytes for the optimization of the culture protocol provided that 4 or more oocytes were isolated. The other half of the oocytes were used for the benefit of the patients pursuing pregnancy and remained outside the present study.

The immature follicles (sized 8-12 mm in diameter) were aspirated using a single barrel needle (COOK) typically on day 7-9 in the cycle and after the patients had received a low dose FSH priming for 3 days. The oocytes were cultured in a very simple and completely defined medium (TCM-199) without any supporting cells or serum, human serum albumin (hSA) being the only protein source added. Moreover, no FSH, LH or other hormones or growth factors were added - for two separate reasons: 1) the placebo control group should act as a true baseline of spontaneous oocyte maturation and 2) the exclusive effect of exposing with FF-MAS on nuclear maturation should be observed. Synthetic FF-MAS was made to high purity (> 99% pure) from Litocol acid as previously described (Murray et al. 2000) and formulated in final concentration of 0.2% v/v ethanol in TCM-199 medium. The control groups did also receive 0.2% v/v ethanol. Thirty-one treatment cycles were performed and 86 human oocytes were donated to enter the study and randomly assigned to 1 of 7 treatment protocols: 0 hour controls (fixed immediately upon isolation), 22 hours culture with or without FF-MAS (20 micromolar), 30 hours with or without FF-MAS (20 micromolar) and 40 hours with or without FF-MAS (20 micromolar), approximately 12 oocytes per group.

We observed that the culture time of 30 hours was superior to both 22 hours (majority of eggs still immature (GV) or maturing (M-I)) as well as 40 hours, irrespective of FF-MAS treatment (potentially 40 hours was too long culture period under our culture condition, since the degeneration rate (morphological criteria) was observed to increase).

Furthermore, we observed a significant beneficial effect of FF-MAS (20 micromolar) at 30 hours culture both on the frequency of M-II oocytes (29% in controls vs. 67% in FF-MAS treated) as well as apparent effect on degeneration (29% in controls vs. 4% in FF-MAS treated, see Figure 4 ). Caution should be applied in concluding on these results since the entire study was based on a small oocyte number (86 oocytes in total, divided into 7 groups). Moreover these were oocytes from PCOS patients and it is known that the endocrine background and the oocyte physiology are different vs. the more normal infertile patient population.

Forty hours of culture was seen in our study to be too long and clear signs of oocytes aging/degeneration were noted and all the oocytes had lost their cumulus cells.

In comparison several research groups have reported beneficial effect upon human oocyte nuclear maturation In Vitro by applying even longer culture times e.g. 48-74 hours, in media containing sera, pituitary hormones, growth factors, steroids or cells (Lefevre et al. 1987) (48 hours), (Prins et al. 1987) (36-74 hours), (Cha et al. 1991) (32-48 hours), (Trounson et al. 1994a) (21-54 hours), (Barnes et al. 1995, 1996) (36-46 hours), and (Russell et al. 1997) (52 hours).

One other investigation has been published on the effect of exposing human oocytes In Vitro to meiosis activating sterol, however MAS in the form of extracted and purified FF-MAS from human follicle fluid (Cavilla et al. 2001). The study population was a mixture of PCOS patients and patients from couples undergoing ICSI treatment for male infertility, and the patients responded quite different. For the ICSI patients FF-MAS (in 10 and 30 micromolar concentration) was seen to increase nuclear maturation (to M-II) similar to what Grøndahl et al. had observed with synthetic FF-MAS (concentration 20 micromolar, Grøndahl et al. 2000a, V)), albeit in a PCOS population. Cavilla et al. observed the oocytes from PCOS patients to benefit from 30 micromolar by higher survival rates (lower degeneration rates), but noticeably no significant impact on nuclear maturation to M-II completion was noted. Finally, an apparent tendency towards improved subsequent development in the FF-MAS treated groups following ICSI by sperm donor with proven fertility was noted as encouraging,

In summary, we found that oocytes from PCOS patients tolerated to be cultured in a very basic TCM-199 based medium without any hormones, sera or other additives except hSA as protein source and pyrovate as energy source, that 30 hours were the optimal culture time to obtain M-II and that FF-MAS had beneficial effects towards nuclear maturation, cytoplasmic maturation and oocyte degeneration a knowledge we have utilized in subsequent studies. We have later adapted a more complex medium that contains a series of survival factors for the subsequent IVM trials necessary for culture of oocytes from non-PCOS patients which will be addressed later in section 6J (Smitz et al. 2007, XIV).

F. Signaling pathways of FF-MAS in mouse oocytes (Papers VI & VII)

These studies were undertaken to investigate the signaling pathways of FF-MAS, to corroborate or disprove the hypothesis that FF-MAS is part of the physiological maturation signal downstream of FSH and LH surges and to reveal any potential ligand-receptor mechanism.

We have hypothesized that FF-MAS and related endogenous sterols could act as a secondary messenger downstream from the gonadotropins surge inducing signals in mammals involved in meiosis signaling and oocyte maturation. This effect could be on the nuclear character (chromatin condensation etc) and on the cytoplasmic maturation (biochemical, MPF activation etc) as well as on processes interlinking nuclear and cytoplasmic maturation e.g. spindle assembly. Different line of evidence has been generated to support the existence of a receptor action in the induced-maturation pathway.

The FF-MAS induced maturation has been found to be pivotally dependent upon protein synthesis (Grøndahl et al. 2000b, VI). This is similar to progesterone mediated maturation in Xenopus oocytes that requires initiation of protein synthesis. In Xenopus following progesterone induction and preceding GVBD several proteins start to be synthesized from stored mRNA pools including Cyclin A and B1 and c-mos protein (Nebreda and Hunt, 1993).

We have furthermore established that the FF-MAS-induced (however not the spontaneous) oocyte maturation in mice can be blocked by even low concentrations of cholera toxin (Grøndahl et al. 2000b, VI). This could suggest that a G-coupled receptor mechanism is involved in the FF-MAS signaling. This is analogous to observations in Xenopus lavis oocytes where progesterone-induced oocyte maturation is sensitive to cholera toxin (Schorderet-Slatkine et al. 1978). Noticeably, Xenopus oocytes are non-responsive to FF-MAS and mammalian oocytes non-responsive to progesterone (Grøndahl and Ottesen, 1997).

Moreover to support the hypothesis of the existence of a MAS receptor we have published how labeled FF-MAS show displaceable binding to the oolemma of oocytes from mice, cows and monkeys (marmoset) in contrast to non-specific binding of close cholesterol analogues (Faerge et al. 2001b, VII).

In comparison in fish and amphibians the endogenous meiosis signaling has been identified and a firm receptor-paradigm of the signaling pathway is being constructed. A Japanese group has put forward a general model of MPF formation during oocyte maturation in amphibians (Yamashita et al. 2000). In this model progesterone (secreted from the supporting cells in the oocyte micro-environment upon gonadotropins regulation) mediates a membrane receptor induced signaling that via mos, MEK and MAP-kinase modulating pre-MPF (pre-maturation promoting factor) to MPF in Xenopus lavis oocytes. The responsible progesterone receptor on the Xenopus oocyte surface has recently been cloned (Bayaa et al. 2000) and it is noticeably the non-genomic signaling pathway that mediates meiosis. In fish and other amphibians than Xenopus the initiation of meiosis is mediated by 17α-20β-dihydroxy-4-pregnen-3-one (17, 20-DP) and again involves activation of this pathway. In mammals a putative receptor-mechanism is hypothesized to involve three pathways 1) the mos/MEK/MAP-kinase pathways leading to MPF formation and stabilization 2) direct translational activation of masked Cyclin B mRNA translation to Cyclin B and 3) a pathway of decreasing cAMP and A-kinase activation (Yamashita et al. 2000; Schmitt and Nebreda, 2002).

Activation of mos/MAP-kinase during meiotic maturation has been established for many species first in Xenopus oocytes (Nebreda and Hunt, 1993) later in mammals e.g. mice (Verlhac et al. 1994) and human (Heikinheimo et al. 1995; 1996).

A study by Coticchio and Flemming observed that spontaneous and FSH-induced maturation in mouse oocytes is mediated by different intracellular pathways (Coticchio and Flemming, 1998). The authors found overt differences via investigating the IP3 pathway and interfering with Ca2+ dynamics in maturing oocytes. In a later study both FSH, epidermal growth factor and FF-MAS was observed to differentially influencing nuclear maturation and polar body extrusion (Coticchio et al. 2004).

We and others have been demonstrating that the signaling pathway of spontaneous and induced meiotic maturation is different observing that for instance the mos /MAP kinase pathway is involved in both the FF-MAS and the FSH signaling In Vitro in contrast to spontaneous resumption of meiosis where GVBD can occur without MAP kinase activation (Grøndahl et al. 2000b, VI; Faerge et al. 2001a, VII; Leonardsen et al. 2000b; Sun et al. 1999a;1999b).

This observation could be key in the mechanistic understanding of the role of FF-MAS in oocyte maturation. Since FF-MAS clearly signals through mos and MAP-kinase activity this could elicit a plethora of important cytoplasmic and nuclear events including regulation of microtubule dynamics, spindle assembly and chromosome condensation all determinants for oocyte quality and embryo developmental potential (Verlhac et al. 1993; 1994; 1996; Dedieu et al. 1996; Sun et al. 1999a; 1999b). Microtubule dynamics and spindle quality was observed to be beneficially influenced by FF-MAS in murine oocytes (Hegele-Hartung et al. 1999). Furthermore FF-MAS has been observed to protect oocytes from precocious chromatid segregation in mice (Cukurcam et al., 2003) which could be associated with positive influence on the spindle structure and microtubule configurations related to the mos-gene family/ MAP kinase signaling. Also in human oocytes mos/MAP-kinase has been identified (in the specific isoform p42ERK2) and although not widely studied in humans yet, MAP-kinase is thought to play a similar role in oocyte maturation as in other vertebrates (Sun et al. 1999a).

In conclusion we have found strong evidence for that the signaling pathway of FF-MAS involves the activation of the mos-gene family /MAP-kinase signaling cascade. Further studies are needed to investigate the eventual ligand-receptor interaction on this activation. Finally, we hypothesize that LXRalpha and especially LXRbeta receptors are involved in FF-MAS signaling since we have observed binding and conformational changes to these receptors of FF-MAS and other meiosis activating compounds, presently published in the form a WO patent application (Grøndahl et al. WO 200404459). Janowski reported that oxysterols made use of LXRalpha in their signaling (Janowski et al. 1996) however our group could not confirm that the oxysterols in that publication had any meiosis activity (Grøndahl et al. 1998, IV).

Hence the full signaling pathway of FF-MAS and the influence of gonadotropins and growth factors still remain to be fully elucidated.

G. The endogenous formation and biochemical synthesis of FF-MAS (Papers IX & X)

These studies were undertaken to investigate 1) the physiological nature of FF-MAS and 2) to identify the regulation under which FF-MAS is governed including the relationship to the gonadotropins FSH and LH.

Furthermore the synthetic generation of FF-MAS has been pioneered as well as alteration of the molecule to discover more drug-like analogues i.e. by increasing the potency as well as improving the formulation and stability characteristics of MAS analogues.

The first isolation and identifications of FF-MAS (and the closely related meiosis active metabolite T-MAS) were obtained from samples of human follicle fluids from IVF patients and crude testicular extracts from bulls (Byskov et al. 1995). FF-MAS is the product of demethylation of lanosterol by the action of the enzyme lanosterol 14alpha-demethylase (LDM) and converted to T-MAS by sterol delta14-reductase (Aoyama and Yoshida, 1986). T-MAS is further converted to zymosterol and later cholesterol in several steps under the action of among others delta7-reductase (Schroepfer, 1982; Yoshida and Aoyama, 1984). LDM is a key enzyme in cholesterol biosynthesis and the gene encoding for LDM is identified to be CYP51 (Rozman et al.1996; Strömstedt et al. 1996) a member of the cytochrome P450 gene family and highly conserved among yeast, plants and animals (Aoyama et al. 1996). The human and rat CYP51 gene has recently been cloned (Strömstedt et al. 1998) and show high degree of homology.

Thus, FF-MAS and T-MAS are produced in all cholesterol-producing cells by this conversion of lanosterol, however, they accumulate only in the gonads probably due to the tissue-specific blocking of the enzymes converting these sterols downstream to zymosterol (Leonardsen et al. 2000a).

Quite a range of endogenous sterols that are meiosis active have been described in human follicular fluid with similar potency as FF-MAS (Grøndahl et al. 2001) as earlier discussed (see section 6.D and Figure 2). Synthetic FF-MAS and MAS analogues have been engineered and a preliminary structural-activity relationship of synthetic sterol analogues has been modeled (Murray et al. 2000; Murray et al. 2002, IX; Murray & Blume unpublished ). In summary, lipophilic substituents on the 25 and 26 position of the sterol side chain are generally tolerated. Thus, these analogues have retained efficacy to induce meiosis in the oocyte assay (see Figure 5 ).

Moreover, it was observed that in compounds bearing hydroxy-group on position 16, the delta 8 or the delta8,14 double bond moieties is not essential for activity. Again it was substantiated that only very minor chemical alteration completely changed the meiotic activity as seen in a previous report (Grøndahl et al. 1998, IV). Synthetic analogues that are potent antagonists to the meiotic resumption of meiosis have also been generated (Grøndahl et al. Int. Patent WO98/28323) as have analogues that possess both a superior efficacy as well as superior potency profile vs. the endogenous sterols (i.e. compounds that are active in the nano-molar range see Marin Bivens et al. 2004b, XII). Highly surprising is has been possible to screen on the simple parameter of meiosis activating activity (i.e. inducing meiosis) and to get very potent candidates that are highly efficacious in inducing cytoplasmic maturation in both outbreed (fertile) and inbreed (sub-fertile) mouse models as later discussed in details (see section 6.I)

The endogenous regulation of FF-MAS and the influence of gonadotropins on FF-MAS dynamics has been the focus of two reports, one paper in mice (Baltsen, 2001) and one paper in rabbits (Grøndahl et al. 2003, X), respectively, however with somewhat different kinetic findings. In mice, Baltsen observed that the ovaries responded to an ovulatory stimulation of FSH and hCG by increasing their content of 4, 4 di-methyl-sterols. This increase was dose-dependently correlated to gonadotropins administrated. The peak of FF-MAS was observed to build up 10-15 hours after hCG administration peaking in the time before ovulation and therefore the author questioned whether this could support the role of FF-MAS for mediating the resumption of meiosis. In pre-pubertal rabbits, we found that the stimulation of FSH was not seen to elicit any significant burst of FF-MAS synthesis. However as early as 1 hour after hCG stimulation (following 48 hours of FSH priming) a significant 4-times increase of FF-MAS was observed i.e. much faster than was reported in mice by Baltsen. This increased FF-MAS content in the rabbit ovaries (concentration in total ovarian tissue, for detailed M&M see Grøndahl et al. 2003, X) was maintained throughout the observation periods (4 and 12 hours after hCG administration) ( Figure 6 ).

The timing of resumption of meiosis In Vivo is known to species dependent. It is believed that the mid-cyclic LH rise in most if not all species (either in the form of a peak or an LH build-up) triggers the oocyte maturation to occur In Vivo (Adashi, 1994). Interestingly, it has been observed that recombinant LH acts as a survival and differentiating factor that increases oocyte maturation when acting on the entire follicle in mouse pre-antral follicle culture In Vitro (Cortvrindt et al. 1998). Moreover, Zhang et al. (1993) reported enhanced fertilization and cleavage rate of IVM human oocytes when cultured in the presence of menopausal gonadotropin comprising mainly LH activity.

In conclusion, we postulate that the FF-MAS rise following hCG injection could be interpreted as some physiological role of MAS in relation to LH and therefore MAS could be argued to have the possibility to influence the final maturational steps in particular cytoplasmic maturation that occurs post GVBD a hypothesis that both our own data and data published by Baltsen support.

H. FF-MAS effect on oocyte maturation in Porcine Oocytes - doses and species specific features (Paper XIII)

It has been documented in a range of studies that FF-MAS increases the quality of oocyte maturation In Vitro in mice, and thus the developmental potential of oocytes exposed to FF-MAS during In Vitro maturation is improved.

This study (Paper XIII) was undertaken to investigate the effect of FF-MAS in porcine oocyte maturation and pronuclear formation to obtain data to guide the use of FF-MAS in humans or even to be deployed as a niche in biomedical animal model production. It should be stressed that it is highly unlikely that IVM, IVF and IVC ever will find use in commercial pig reproduction.

Porcine cumulus-oocyte complexes (COCs) were obtained by follicle aspiration of abattoir ovaries and cultured In Vitro for 48 hours in medium with or without 10% porcine follicle fluid (pFF) and supplemented with 1, 3, 10, 30, or 100 micromolar FF-MAS. The resulting matured oocytes were In Vitro fertilized and evaluated after 18 hours of In Vitro culture.

The results demonstrated that porcine COCs could be matured and fertilized in the absence of pFF and benefited from the presence of FF-MAS. Previous studies in porcine oocytes have observed an inhibitory effect on meiosis of both co-culturing granulosa cells and also by addition of follicular fluid (Tsafriri and Channing, 1975).

FF-MAS lowered the degeneration rate of zygotes (i.e. one-cell fertilized eggs) and lowered the errors of fertilization (polyspermia rate) regardless of the presence or absence of pFF. However this effect on cytoplasmic maturation is highly dose dependent and positive effect of FF-MAS (lowering polyspermia rate; lowering degeneration rate) was only seen in the dose range 1-10 micromolar and not in 30 and 100 micromolar concentrations.

This is the first report on effect of FF-MAS in porcine oocytes and gives important information to doses and effects in non-rodent species.

In comparison in mice micromolar concentration of synthetic FF-MAS (range 0.1-20 micromolar, optimal range 1-10 micromolar (thus the same concentration window), EC 50 ~ 1 micromolar, Emax (maximal efficacy) = 86% GVBD at 7 micromolar) has robustly been observed to overcome the meiotic inhibition from hypoxanthine (Hx, endogenous purine), or isobuthyl-methyl-xanthine (IBMX) or dbcAMP and significantly induce GVBD and PB extrusion in the C57black/ DBA F1 mice. These findings and the effective dose-range have been confirmed by several groups (Downs et al. 2001; Tsafriri et al. 2002), although some discrepancies were noted on the specific meiosis inhibitors and the effect in cumulus-enclosed oocytes. Recently, choice of mice strains has been observed to influence the action of Hx and the specific effect of FF-MAS (Griffin et al. 2004). We have been hypothesizing that the spontaneous maturation vs. (MAS)-induced maturation are vastly different, both with regards to requirements of protein synthesis inside the oocyte, the signaling messages and the resulting cytoplasmic maturation as well as the quality of the oocyte and the developing zygote post-fertilization.

Also a range of studies has been published investigating the influence on the meiotic process that enzymatic inhibition of the cholesterol biosynthesis may cause, thus aiming for inducing an endogenous accumulation of meiosis active sterols (Leonardsen et al. 2000a; Vaknin et al. 2001; Tsafriri et al. 2002).

FF-MAS has also been published to signal in rat oocytes In Vitro (Hegele-Hartung et al. 2001). Importantly, entire ovaries have been cultured as ex vivo perfused organ culture ad modum Brännström (Brännström et al. 1995) and FF-MAS has been observed to overcome this purely physiological meiosis inhibition exerted by the full functional follicle in rats in dose-dependent manner (Hegele-Hartung et al. 2001).

In conclusion we have found that FF-MAS has been effective in influencing the quality of oocytes maturation and thus the developmental potential of oocytes exposed during maturation in a range of species including porcine and murine oocytes in effective doses ranging from 1-10 (20) micromolar. Thirty (30) micromolars has been established to be too high and 0.1 micromolar has been established to be a non-effective dose in all species tested at least with regards to inducing nuclear maturation.

I. The effect of FF-MAS and MAS Analogue mediated oocyte maturation on fertilization and pre-embryo development in normal
and sub-fertile mice strains (Papers XI & XII)

Most previous studies of FF-MAS have been made on oocytes that have been arrested in their meiotic development and the observations have been focused to overcome the meiotic inhibition from hypoxanthine (Hx, endogenous purine), or isobuthyl-methyl-xanthine (IBMX) or dbcAMP. The present studies (XI and XII) were designed primarily to study cytoplasmic maturation and to evaluate the effect of FF-MAS and MAS analogues on cytoplasmic maturation in cumulus enclosed oocytes in different mice strains without any attempts of inhibiting meiosis (Marín-Bivens et al. 2004a, XI; 2004b, XII). Moreover, focus was to look at germinal vesicle (GV) to metaphase-II (M-II) transition in models where this process is known to be impaired.

The idea was to zoom in on the influence on cytoplasmic maturation observed as the resulting developmental potential of oocytes exposed exclusively during In Vitro maturation. Some of the strains used have an excellent fertility and display normal progression and kinetics of meiosis (e.g. C57BL/6J × SJL/J)F 1 , whereas other strains have a phenotype of disturbed meiosis and reduced fertility (e.g. inbred strains, that exhibit retarded kinetics of germinal vesicle breakdown (GVBD) and low incidence of progression to M-II (Eppig and Wigglesworth, 1994)). The design involved In Vitro maturing cumulus cell-enclosed oocytes in medium supplemented with FF-MAS and assessing the progression of meiosis to M-II. This was subsequently followed by observation of the M-II and their competence to undergo development to the 2-cell stage and blastocyst stage after In Vitro fertilization.

We found the results to be very interesting and de facto constitute a Proof-of-Concept for FF-MAS as promoting cytoplasmic maturation and increasing embryo quality. On a background of normal fertility FF-MAS had no effect on the kinetic of meiosis (i.e. from GV to GVBD), in contrast to previous studies in arrested oocytes (Hegele-Hartung et al. 1999); however FF-MAS significantly increased the frequency of M-II. No effect was noted in the frequency of 2-cells in these models of outbred mice (see Figure 7 ); however a significant beneficial effect was noted in developmental potential of 2-cells to blastocyst. As a consequence of increased frequencies of maturation to M-II and completion of the 2-cell to blastocyst transition, the percentage of oocytes (total starting number) that developed all the way to the blastocyst stage increased dramatically from 42% (controls) to 75% with maturation promoted by FF-MAS (in concentration 10 and 20 micromolar). Moreover the percentage of 2-cells that developed to blastocysts increased significantly from 60% in controls to 85-90% with 10 and 20 micromolar FF-MAS (see Figure 7A, B). Thus, both nuclear and cytoplasmic maturation were improved by treatment with FF-MAS in normal fertile F1 mice.

Oocytes and embryos from C57BL/6J inbred mice generally are much less successful in undergoing In Vitro fertilization and pre-implantation development than those from hybrid or outbred mice (Roudebush and Duralia, 1996). Only 32% of the oocytes cultured in control medium matured to M-II, whereas FF-MAS (10 micromolar) increased the M-II rate to 79%. Frequency of 2-cell stage embryos increased from 38% in the control group to 60% in the FF-MAS-treated group (in inbred sub-fertile mice), paralleled by an increase in blastocyst transition after oocyte maturation in medium containing FF-MAS. This study concluded that the overall percentage of the total number of oocytes that developed to the blastocyst stage was significantly increased (4% to 17%) by exposure of FF-MAS (10 micromolar) exclusively during the pivotal time of final oocyte maturation.

In summary, FF-MAS was observed to induce both nuclear as well as important elements of cytoplasmic maturation that could be observed as readouts in embryo development both in fertile mouse models and in sub-fertility models.

In human infertility both age related and/or meiosis related mechanisms are thought to be involved. Moreover, dramatic stimulatory protocols have been seen to upset the synchrony of the oocyte and somatic cell compartment and therefore hamper the synchrony between nuclear and cytoplasmic maturation (Hunter, 1977; Eppig et al. 1992; Moor et al. 1998; Eppig et al. 2002).

The efficacy results documented in the present study pertaining to both fertile and sub-fertile mouse strains with significant effect on blastocyst rates and overall efficacy from GV stage oocytes to viable blastocysts (Marín-Bivens et al. 2004a, XI) present a strong pre-clinical proof-of-concept. If similar data in the future were to be reproduced, in a controlled clinical setting with human oocytes and embryos, then a solid clinical proof-of-concept would be established.

The objective of the following study (Marín-Bivens et al. 2004b, XII) was to investigate whether analogues of FF-MAS initially identified on the basis of their greater potency to overcome meiotic inhibition in medium cultured with hypoxanthine could also promote the completion of meiotic maturation and improve developmental competence to complete pre-implantation development.

Analogues were prepared by sterol chain introduction of amines to generate more water soluble, more potent, drugable and patentable analogues. MAS analogue `933 was found to be active in the nanomolar range and at least 10 × more potent than FF-MAS i.e. active at 0.1 micromolar concentration significantly influencing M-II, 2-cell rates and blastocyst rates. In a separate aim of this study, FF-MAS and MAS analogue `933 ability to promote meiotic competence in immature oocytes from very young research animals (in mice of only 18 or 20 days of age respectively) was investigated.

Both FF-MAS and MAS analogue `933 promoted the completion of the physiological processes that would have occurred later when the oocyte was growing in a large follicle - a development that gonadotropin treatment (eCG) In Vivo failed to promote. This could be clinically very relevant since many reports exist of failure of human oocytes to complete meiotic maturation during clinical IVF protocols (Rudak et al., 1990; Levran et al., 2002; Neal et al., 2002).

In conclusion, considering the low frequency of successful fertilization and development following In Vitro maturation of human oocytes, the addition of FF-MAS or MAS analogue `933 to the maturation medium may prove highly beneficial.

J. Clinical Phase I safety testing of applying synthetic FF-MAS in Human IVM (paper XIV)

Given the beneficial effect seen by exposure of the endogenous signaling molecules FF-MAS in all pre-clinical models tested, The Fertility Team at Novo Nordisk together with leading clinical IVM researchers set out to move into the human clinic with this molecule based on the following key observations: 1) FF-MAS (and related MAS sterols) induces meiosis in mouse oocytes In Vitro in both denuded and cumulus enclosed oocytes, shown by an increased onset of germinal vesicle breakdown and an increased incidence of metaphase II stage oocytes after exposure to FF-MAS (Grøndahl et al. 1998, IV ; 2000b, VI; Grøndahl, 2002, VIII; Marín-Bivens et al. 2004a, XI). 2) FF-MAS stimulation of oocytes during In Vitro maturation mediates a higher fertilization rate in mouse oocytes In Vitro after IVF compared to In Vitro spontaneously matured oocytes and In Vivo matured oocytes (Hegele-Hartung et al. 1998; Marín-Bivens et al. 2004a, XI and 2004b, XII). 3) FF-MAS stimulates the In Vitro resumption of meiosis and its corresponding cytoplasmic maturation in murine and human oocytes (Hegele-Hartung et al. 1999, Grøndahl et al. 2000a, V; Marín-Bivens et al. 2004a, XI and 2004b, XII). 4) FF-MAS when exposed to oocytes during In Vitro maturation mediates both a higher fertilization rate and yield significant higher blastocyst rates in mice in both fertile and sub-fertile models (Marín-Bivens et al. 2004a, XI and 2004b, XII). 5) FF-MAS when exposed to oocytes during In Vitro maturation mediates a higher fertilization rate after IVF and yields higher numbers of live born pups after 2-cell transfer in mice (Hegele-Hartung et al. 1998; Hegele-Hartung and Grøndahl unpublished data; Grøndahl, 2002, VIII).

The design of this study was a bi-centric, parallel-group, controlled, partially blinded and prospective clinical trial (Smitz et al. 2007, XIV). The study objective pursued was In Vitro safety and efficacy evaluation of applying FF-MAS In Vitro to immature human oocytes based on the primary endpoint: incidence of Metaphase-II (M-II) oocytes with numeric chromosome abnormalities using full or partial karyotyping. The secondary endpoint was to assess the efficacy of FF-MAS in obtaining M-II oocytes after In Vitro culture compared to FSH/hCG-stimulated In Vivo matured human oocytes. No fertilization was attempted on FF-MAS exposed oocytes (or any oocytes) in this trial.

The very basic study design is depicted in Figure 8 .

Numeric chromosome abnormalities were assessed using three different methods of karyotyping: Giemsa count, Partial Karyotyping using FISH (Fluorescent in situ Hybridization using 7 commercial probes from MultiVision) or full karyotyping using SKY (Spectral Karyotyping) as shown in Figure 9 .

In summary of this 248-patient study investigating 1627 human oocytes (from non-PCOS IVF patients and healthy volunteers, age range 25-37 years), IVM with or without the presence of FF-MAS was found to yield M-II oocytes that did not display any increase in the rate of numeric chromosome abnormalities as compared to In Vivo FSH/ hCG stimulated controls. Furthermore, similar numbers of fully mature MII oocytes could be obtained from IVM protocols with MAS as could from FSH/hCG protocols (Smitz et al., 2007, XIV). It was confirmed that 30 hours of culture was optimal as seen previously in oocytes from PCOS-patients (Grøndahl et al. 2000a, V) based on the chromosome abnormality endpoint. However, due to the utilization of a complex defined culture medium and the inclusion of oocyte survival factors like rFSH, Insulin, IGF-1, nuclear maturation to M-II was high (> 70%) in all groups and this secondary endpoint gave no possibility for differentiation between groups, however could further substantiate the safety objective of the study.

In the IVM groups the short low-dose rFSH protocol (modified from Wynn et al. 1998) induced follicle growth in all patients. The overall oocyte retrieval rate was 61% and a mean number of 6.2 cumulus-oocytes complexes per patient were available for IVM.

Overall high chromosome normality rates were found in both the IVF and IVM groups after 30 hours of culture (e.g. 85% up to 93% of oocytes were observed to be chromosomal normal haploid cells). Tendencies of reduced aneuploidy rate were seen in the 1 and 10 micromolar FF-MAS groups after 30 hours of culture (down to 7% aneuploidy rate vs. 15% in controls), however no significant differences were observed.

Other studies have focused on the safety of FF-MAS added to In Vivo matured FSH stimulated oocytes retrieved after normal IVF stimulation and subsequent cultured with FF-MAS 4 hours prior to fertilization (Loft et al. 2004; Loft et al. 2005). In these studies FF-MAS was found to be safe as a medium additive exposed before fertilization based on chromosome abnormalities in the resulting pre-embryos. Moreover, the albumin formulation was observed to be superior to the ethanol based formulation. In the albumin formulated study the chromosome aberration rate on the resulting pre-embryos ranged between 39-53% with no significant difference among controls or FF-MAS groups.

In contrast FF-MAS was seen to cause chromosome aberrations when acting on FSH stimulated In Vivo maturated oocytes both before and 24 hours following fertilization (Bergh et al. 2004). Hence it can be concluded that FF-MAS exclusively should be exposed to oocytes prior to fertilization and not after zygote formation when the pre-embryo is transforming to make mitosis divisions.

Direct studies of human gametes have previously shown that as many as 20-25% of oocytes, but only 3 to 4% of sperm cells, are chromosomal abnormal (Martin et al. 1986; Pellestor, 1991; Hunt and Hassold, 2002). Our findings in the present paper displaying lower rates of chromosome abnormal oocytes could partly be due to the study population of both mainly younger patients and also younger fertile healthy volunteers.

Aneuploidy as such has been named the biggest single problem in human reproduction, being the single most common cause of miscarriage and affecting up to 50% of transferred embryos in the clinic (Cohen, 2002; Munne and Cohen, 1998; Munne et al. 2002; Ziebe et al. 2003).

A large proportion of the IVM or FSH/hCG primed IVF oocytes were observed to be chromosomally normal (between 85 to 93% normal haploid MII oocytes) in the present 248-patient study (age range 25-37 years) of IVF and IVM human oocytes (Smitz et al. 2007, XIV). In contrast, less than half of the IVF embryos (exactly 31% uniformerly normal embryos out of 144 human randomly selected IVF embryos) was observed to be normal in the study with material from 4 large hospital IVF clinics in Scandinavia (age range 25-37 years) (Ziebe et al. 2003).

Thus, surprisingly by combining the observations from the two different FF-MAS safety programmes on the one hand the present IVM study (Smitz et al. 2007) and on the other hand the parallel IVF studies we have conducted (Ziebe et al. 2003; Bergh et al. 2004; Loft et al. 2004, 2005), respectively, it appears that it is primarily during the fertilization process and the early pre-implantation cell divisions that a major part of the chromosome errors are introduced.

Moreover, a great deal of mosaics was noted in Ziebe et al. 2003 which questions the validity of using pre-implantation diagnostics on a single blastomeric cell as a accurate tool, when very often more than 2 cell lines exist in the embryo, given the possibility of both false positive (the biggest clinical challenge) and potentially false negative outcomes. Finally, it needs to be further substantiated to what degree an early cleaving human embryo can deal with a low grade of aneuploidy in a subset of the blastomers.

In conclusion, further clinical studies in human IVM involving fertilization and embryo transfer are needed to evaluate whether FF-MAS also acts at the human oocytes by contributing favorably to cytoplasmic maturation and oocyte quality or developmental potential, analogously to the data that have been obtained in several models of mouse oocytes (Marin-Bivens et al. 2004a, XI and 2004b, XII). This is now possible to pursue in subsequent clinical trials based on the safety assessment of human IVM documented herein.