An epigenomic biography of the mammalian oocyte

Successful fertilisation and early development depend on the quality of the ovulated oocyte. Even though this is endorsed by both the experimental and clinical practice, our still scarce knowledge of the biology of the mammalian oocyte makes it difficult to identify parameters that define the quality or 'developmental competence' of an oocyte. The many attempts to establish markers of the oocyte's developmental competence have produced divergent results or have worked only in specific experimental contexts. To this respect, the oocyte morphology, the concentration of various factors in the follicular fluid, the role of the oocyte's mitochondria, the telomere length and the transcription profile of cumulus cell-specific genes are some of the most studied aspects of the oocyte and of the ovarian follicle biology. The analysis of the level of cumulus cell-specific transcripts has identified groups of genes that are directly (PTGS2, HAS2, GREM1 and PTX3) [1-3] or inversely (GPX3, CXCR4, CCND2 and CTNND1) [4-5] correlated to human embryo preimplantation quality and pregnancy outcome. Recent studies analysed the whole cumulus cells transcriptome in human [6]and bovine [7] cumulus cell-oocyte complexes bringing up a new set of putative marker transcripts. The concentration in the follicular fluid of myo-inositol (a serum trophic factor), inhibin B [8-9] or AMH have been used for their predictive value of human preimplantation embryonic development, with the latter suggested as a better predictor of oocyte fertilisability [10] and pregnancy rate [11]. Some authors have proposed the presence of a high level of estradiol on the day of hCG administration as a candidate marker of low pregnancy rate [12-13]), but these data are conflicting with others that describe no correlations with the final pregnancy outcome [14-15]. Similarly, a reduction at the time of oocyte collection in the level of progesterone receptor in human cumulus cells was associated with morphologically good oocytes [16]. Mitochondria have been the subject of a large number of studies, but how and whether they contribute to the determination of the oocyte developmental competence is still unclear. The whole preimplantation period is sustained by mitochondria produced during oogenesis and only when the embryo begins implantation, their production is resumed. Therefore, an unbalanced number of these organelles, an incorrect distribution or an altered function may have negative effects on the early stages of development [for a review see 17]. The number of mitochondria in mouse primordial germ cells (PGCs) is very small, being approximately 10-100 per cell, then, by the mature oocyte they sum up to ~ 90.000 [18]. The total number of mitochondria seems to be critical to the developmental competence of an oocyte, since subnormal levels of these organelles correlate with premature maturation arrest of the oocyte and early death of the preimplantation embryo [19-20]. A low mitochondrial complement may determine a bioenergetic/metabolic shortage with consequences on the oocyte’s ability of meiotic resumption, fertilisation and to sustain the early phases of development [reviewed in 20-23]. Along with these studies, ATP values have also been associated with the oocyte’s developmental competence; an ATP content of > 2 pmol seems a threshold to distinguish between developmentally competent and incompetent human oocytes [24]. During folliculogenesis, mitochondria are located in different regions of the oocyte [25-28] and by the mature oocyte they will have an asymmetric polar distribution that will be maintained through segmentation resulting in blastomeres that will own a different number of mitochondria with a different spatial patterning. A number of observations substantiate the involvement of these organelles not only as powerhouse, producing most of the ATP in the cell, but they may also regulate development by modulating Ca2+ signalling, reactive oxygen species (ROS) and intermediary metabolites and through their control of apoptosis. Oxidative stress and intracellular redox potential (IRP) have been shown to regulate the function of a number of transcription factors important in early development. For example, NF-KB (nuclear factor kappa-light-chain-enhancer of activated B cells) and GSK3β (glycogen synthase kinase-3β), expressed during preimplantation development, are activated by mitochondrial ROS production [29-30]. S-glutathionylation of many transcription factors [31] occurs after the oxidation of the IRP, thus, the ability of mitochondria to modulate the IRP will possibly change the activity of these proteins. Altered mitochondrial Ca2+ cycling and ROS production are determinant for the oocyte to enter and accomplish the apoptotic programme [32]. Oxidative stress induces mitochondrial dysfunction that triggers apoptosis in the mouse oocyte and zygote [33]. Specific oocyte morphological features are used to select female gametes of good quality. Oocytes are graded as good, when they posses a cytoplasm with fine granules, a single polar body, a narrow perivitelline space and a ring shaped zona pellucida [34]. On the contrary, they are correlated with low fertilisability and developmental competence, when they own vacuoles and refractile bodies within the ooplasm. Although used in some laboratories that practice human assisted reproduction, these morphological markers have been the subject of many criticisms and are not widely accepted [34-40]. As for nuclear morphological features, a large literature as demonstrated the possibility to select between developmentally competent and incompetent antral oocyte, depending on their chromatin configuration (see below). This selection, however, relies on the use of fluorochromes whose use, for obvious reasons, is not advisable in our species. Up to date, our knowledge lacks of links between the morphological aspects described and the genomic and epigenomic features that work in the backstage. Bringing to light these links would permit the identification of molecular markers of the oocyte developmental competence. In this article, we will review our current understanding of the changes that occur to the oocyte epigenetic signature during folliculogenesis and in mature oocytes