The EMBO workshop ‘Oocyte maturation and fertilization: lessons from canonical and emerging models’ was held at the Oceanologic Observatory of Banyuls in France in June 2013 and was organized by Anne-Marie Geneviere, Olivier Haccard, Peter Lenart and Alex McDougall. A total of 78 participants shared their research on germline formation, oocyte development, sperm, fertilization and early development. Here, we report the highlights of this meeting.
This European workshop brought together the oocyte maturation and fertilization communities. Leaders of these fields, as well as younger researchers, could compare their studies on eggs and sperm in a wide range of animal models including protostomes, such as cnidarians (jellyfish), nematodes, insects (Drosophila), and deuterostomes (echinoderms – starfish and sea urchin), urochordates (ascidian) and vertebrates (Xenopus, Zebrafish, mouse, bovine). Forty talks were spread between ten sessions and provided the attendees with the opportunity to learn more about germ line determination, asymmetric spindle positioning, cell cycle regulation or egg activation. In addition, two emerging models in these fields were introduced in this third meeting in the series that started in 2008. They are the nematode Caenorhabditis elegans and the fruit fly Drosophila melanogaster, both genetic models, which will bring invaluable insights into the mechanisms of oocyte development and egg activation.
Germ line determination
The first talk of the meeting presented how the germ line is determined. The germ line is either predetermined or induced later in development depending on the organism. Echinoderm embryos, which are amenable to both biochemical and live imaging studies, are especially well suited to study the processes of germ cell specification during early development. Sea urchins appear to use an inherited mechanism to form the germ line, a mechanism that is very different to the closely related starfish. In sea urchin embryos, the small micromeres contribute to the germ line (Yajima and Wessel, 2011) and express germ line factors, such as vasa and nanos. In his talk, Gary Wessel (Brown University, Providence, USA) explained how such germ line factors are expressed and maintained specifically in the germ line. He showed, for example, that the post-translational regulation of vasa by gustavus contributes to the selective accumulation of vasa in the germ line (Gustafson et al., 2011). S. Zachary Swartz (Brown University, Providence, USA) presented further insights from sea urchin based on his recent findings that the primordial germ cells downregulate a specific deadenylase to stabilize the transcripts inherited by the germ line.
Cell cycle regulation
Many invertebrate and vertebrate oocytes are blocked in prophase of the first meiotic division until ovulation in the sexually mature animal. Resumption of meiosis in prophase-arrested oocytes is triggered by the activation of maturation/M-phase-promoting factor (MPF), the classic biochemical activity that induces the entry into M phase. It had been known for a long time that MPF and cyclin-B–Cdk1 were not equal, but it was unknown what (in addition to Cdk1) constituted the missing part of MPF. However, using starfish oocytes, the keynote speaker Takeo Kishimoto (Tokyo Institute of Technology, Yokohama, Japan) demonstrated that cyclin-B–Cdk1 activity is not synonymous with MPF; instead, the MPF also requires Greatwall kinase, which phosphorylates the 19 kDa cAMP-regulated phosphoprotein (ARPP19) to suppress the activity of the phosphatase PP2A-B55 subunit, thus maintaining cyclin-B–Cdk1 substrates in their phosphorylated state (Hara et al., 2012). In Xenopus oocytes, activation of MPF and Cdk1 follows a two-step mechanism; a catalytic amount of Cdk1 is generated in a protein kinase A (PKA)-dependent manner, which then initiates a MPF auto-amplification loop to promote the full activation of Cdk1 and the resumption of meiosis. This second step is independent of PKA activity. Using frog oocytes, Olivier Haccard (CNRS, UPMC UMR7622, Paris, France) demonstrated that the phosphorylation of ARPP19 by Greatwall, which is controlled by basal levels of Cdk1, promotes the binding of ARPP19 to PP2A-B55, thereby inhibiting PP2A. This phosphorylation event renders the MPF auto-amplification independent of PKA (Dupré et al., 2013). The involvement of Greatwall and ARPP19 in MPF activation is conserved between starfish and frog, and it will be interesting to assess whether they also participates in MPF auto-amplification in other species, such as invertebrates (Cnidarian or C. elegans) or simple marine chordates (urochordates).
Asymmetric spindle positioning
During oocyte maturation, the oocyte undergoes two asymmetric cell divisions to extrude two polar bodies; this process requires that the oocyte spindle relocates from the center of the egg to the cortex and relies on the intracellular actin network. Three presentations, by Melina Schuh, Marie-Helene Verlhac and Guillaume Halet, showed live imaging of the actin network in mouse oocytes using utrophin-CH–GFP, a powerful fluorescent probe based on the calponin homology domain of utrophin, which binds F-actin (Burkel et al., 2007), whereas Masashi Mori presented the use of the same technology applied to germinal vesicle breakdown (GVBD) in the starfish oocyte. Melina Schuh (MRC Laboratory of Molecular Biology, Cambridge, UK) showed that modulation of the actin network by Rab11a-positive vesicles is essential for the asymmetric spindle positioning during meiosis I. Her beautiful live images captured Rab11a-positive vesicles, which recruit actin nucleation factors and mediate their transport to the plasma membrane (Schuh, 2011). Using dominant-negative forms of Rab11a and myosin Vb, Melina Schuh showed that they are both necessary for vesicle transport and for the meiotic spindle to migrate to the egg cortex (Holubcová et al., 2013). On the basis of these and previous data, she suggested a model in which myosin-dependent pulling from the spindle poles couples the spindle to the outwards-directed movement of the vesicles and their associated actin filaments. Marie-Helene Verlhac (College de France, Paris, France) then explained that this asymmetric division also requires changes in the stiffness of the egg cortex. Using micropipette aspiration (Larson et al., 2010), she elegantly demonstrated that cortical tension decreases during meiosis I. Softening of the egg cortex relies on the exclusion of myosin II from the cortex and is triggered by Mos/MAPK. Interestingly, cortex softening is accompanied by an Arp2/3-stimulated thickening of the actin-rich cortex, which reaches a maximum thickness during polar body extrusion. Marie-Helene Verlhac suggested that cortex softening, owing to both a decrease of cortical tension and of cortical plasticity that is induced by the thickening of actin, is essential for the migration of the first meiotic spindle to the cortex (Chaigne et al., 2013). After migration to the egg cortex and first polar body extrusion, the peripheral chromosomes promote differentiation of the nearby cortex, resulting in an actin-rich cortical cap that contains activated Cdc42 (Dehapiot et al., 2013). Similarly, with the help of beautiful live images, Guillaume Halet (Institute of Genetics and Development of Rennes, Rennes, France) demonstrated a flow of actin filaments that radiate from this actin cap. The polarization of activated Cdc42 and actin flow were also observed during anaphase II, when the two clusters of segregated chromatids promote the formation of two actin-rich membrane protrusions, one of which eventually forms the second polar body. Finally, Masashi Mori (European Molecular Biology Laboratory, Heidelberg, Germany) presented imaging of the utrophin-CH probe in maturing starfish oocytes, which revealed a dynamic actin network within the germinal vesicle that is surrounded by an actin-rich ring in the nuclear cortex. He showed that both the actin network and nuclear cortex contract after GVBD, which, as he explained, might collect the chromosomes that are scattered within the 80-µm-wide germinal vesicle towards the animal pole so they can be caught by spindle microtubules (Mori et al., 2011). Taken together, this illuminating session illustrated that, despite extensive studies in the mouse since the development of in vitro maturation and in vitro fertilization protocols by R. G. Edwards in the 1960s (Nobel Prize 2012, Medicine), only the recent live imaging of GFP-based molecular probes has been able to grasp the dynamic (and unforeseen) behavior of chromosomes, vesicles or microfilaments that support the complex reorganizations of the egg during meiotic divisions.
Spindle assembly checkpoint
Meiosis I is a cell division that is particularly susceptible to the missegregation of chromosomes, potentially leading to aneuploidy. The spindle assembly checkpoint (SAC) prevents errors during meiosis by stalling the activity of the anaphase-promoting complex/cyclosome (APC/C) until all chromosomes have aligned at the spindle equator. In mitosis, complete congression of all sister chromatids onto a metaphase plate is coupled with the initiation of APC/C activity. Simon Lane (University of Southampton, Southampton, UK) presented measurements of APC/C activity by monitoring the degradation of cyclin B1 in mouse oocytes that undergo meiotic maturation. He observed that the timing of APC/C activation is dictated by the attachment of kinetochores to microtubules, but not by the alignment of bivalents on the metaphase plate nor by the tension on the kinetochores (Lane et al., 2012). He found that during meiosis, the initial attachment of chromosomes to kinetochores by microtubules partially activates the APC/C, independently of the biorientation status of the individual chromosomes. This persistent, partial APC/C inhibition mediated by the SAC extends meiosis I to allow any initial incorrect attachments to become corrected and so reduces aneuploidy.
Oocyte developmental competence
One talk addressed the issue of what renders an oocyte capable of developing. A mammalian, ovulated metaphase II oocyte is not always a good egg, it might resist fertilization or not be competent to sustain development. Thus, eggs have been studied with the aim of identifying markers that would enable the selection of good gametes to fertilize (Zuccotti et al., 2011). Maurizio Zuccotti (University of Parma, Parma, Italy) showed that immature mouse oocytes could be classified into two main groups depending on their chromatin configuration in meiotic prophase. An oocyte can be classified as ‘surrounded nucleolus’ (SN) if it possesses a ring of Hoechst-positive heterochromatin around its nucleolus, or as ‘non-surrounded nucleolus’ (NSN) if it lacks the ring and the chromatin is more dispersed. Fully grown antral SN oocytes complete full-term development. By contrast, fully grown antral NSN oocytes arrest development at the two-cell stage. By comparing the transcriptome of SN (developmentally competent) and NSN (incompetent) oocytes, Maurizio Zuccotti was able to characterize the molecular signature for developmental competence (Zuccotti et al., 2012). For instance, OCT4 is downregulated in NSN oocytes and thus can be considered as a marker for the developmental competence of mouse oocytes.
After ovulation, oocytes must be able to initiate embryogenesis if they are fertilized. Egg activation involves the resumption and completion of meiosis, translation of proteins from stored maternal mRNAs, degradation of other maternal mRNAs and changes to the vitelline envelope. Most events during egg activation can be triggered by a transient rise in cytosolic Ca2+ in all species studied so far. However, in the two most used genetic models in developmental biology (Drosophila and C. elegans), the events that are associated with sperm–egg fusion and egg activation are relatively poorly understood because post-embryonic lethality issues complicated previous mutational screens. Combining a proteomic approach and a genetic screen, the group of Mariana Wolfner (Cornell University, Ithaca, NY) investigated the relationship between known egg activation genes and changes in the phosphorylation patterns that occur upon egg activation in Drosophila melanogaster. Using transcriptomic and proteomic approaches, they identified 4171 transcripts that become polyadenylated and 311 proteins that change their phosphorylation state during egg activation. Their work suggests that Sarah (calcipressin) and the Ca2+-activated phosphatase calcineurin are required for the APC/C-dependent degradation of Cortex (a meiosis-specific CDC20) upon egg activation (Krauchunas et al., 2013). Calcineurin was known to mediate release from metaphase II arrest in Xenopus (but not in mouse) eggs, but as Mariana Wolfner showed here, calcineurin probably mediates the release from metaphase I in Drosophila through its effects on the APC/C, suggesting a conserved role for calcineurin during egg activation. Another argument in favor of a conserved role for calcineurin during egg activation came from the data presented by Mark Levasseur (Newcastle University, Newcastle, UK) who showed that calcineurin also mediates metaphase I arrest by CytoStatic Factor (CSF) in urochordates (ascidian) by regulating the APC/C. The proteomic approach performed by Mariana Wolfner in such a genetically tractable model offers the possibility to identify the numerous actors of egg activation and validate them by knockdown of each of the proteins identified. If the pathways that involve Ca2+ are largely conserved during Drosophila egg activation (Pesin and Orr-Weaver, 2007; Sackton et al., 2007; Yamamoto et al., 2008; Dumollard et al., 2011), the proteomic approach performed by Mariana Wolfner and other Drosophila researchers will help to shed new light on the processes that support egg activation.
One session of the meeting was devoted to sperm, the partners of eggs in fertilization. Harvey Florman (University of Massachusetts Medical School, Worcester, USA) outlined how sperm learn to be fertile. In mammals, sperm are infertile when released by the males. During their journey through the oviduct, they are exposed to factors that originate from the eggs and the female reproductive tract, which stimulate sperm motility and become progressively fertile. Harvey Florman presented data from his group obtained from transcriptional profiling in the oviduct aimed at identifying oviduct regulators that control sperm fertility. They found 93 genes whose expression increased in the oviduct by more than three-fold on the day of estrus. Among them, twelve have been classified as cell surface or secreted proteins that could represent good candidates to influence sperm fertility.
Andrew Singson (Rutgers University, Picaraway, USA) presented data he obtained using the nematode C. elegans that are aimed at understanding the mechanisms of gamete activation and fertilization. C. elegans is an excellent model for the physiology of reproduction and has the advantage that mutant that only affect sperms or eggs can be isolated and maintained. Genetic analysis has identified several genes that are essential for fertility in sperm or oocytes (Singson et al., 2008). Several genes that function in sperm were identified and classified as the Spe group (spermatogenesis defective). Andrew Singson talked about the spe-38 mutant they obtained that cannot enter the oocyte despite direct contact. Based on analysis of the mutant, he identified SPE-38 as a transmembrane protein that interacts with the Ca2+-permeable channel TRP-3/SPE41 and regulates its localization (Singaravelu et al., 2012). This work will enable to better understand sperm-oocyte interactions and could complement fertility research in other species.
Sperm entry triggers a Ca2+ wave in the eggs of almost every species, which is thought to be induced either by binding of the sperm to a sperm receptor on the egg surface or by the release of a soluble sperm factor after its fusion with the egg. Evidence for this so-called ‘sperm factor’ theory come from mammals, as presented in the talk by Karl Swann (Cardiff University, Cardiff, United Kingdom). He showed that phospholipase Cζ (PLCζ) is the mammalian sperm factor and appears to generate the Ca2+-releasing messenger InsP3 from phosphatidylinositol (4,5)-bisphosphate [PtdIns(4,5)P2] on intracellular membranes, rather than from the plasma membrane PtdIns(4,5)P2, as is the case for other mammalian PLC isozymes (Yu et al., 2012). However, it is unclear at present whether this mechanism is at play in other phyla. Ryusaku Deguchi (Miyagi University of Education, Sendai, Japan) showed live images of sperm entry in jellyfish eggs, which clearly showed that fusion of the sperm with the egg was immediately followed by the Ca2+ wave. This observation suggests that in jellyfish, like in mammals, the sperm might release a soluble factor into the egg cytosol, which then induces the release of Ca2+. He then described the mechanisms that prevent polyspermy in the jellyfish; these consist of the suppression of sperm–egg fusion and the reversible inhibition of sperm attraction (Kondoh et al., 2006). The sea urchin egg has been a favorite model for studying intracellular Ca2+ signaling for more than 30 years. Here, intracellular Ca2+ release is mediated by InsP3, cADPR and NAADP, and is used by eggs to increase cytosolic Ca2+ levels. The Ca2+ channels that are targeted by InsP3 (i.e. InsP3 receptors) and cADPR (i.e. the ryanodine receptors) have been known for some time, but the target of NAADP remained elusive until recently (Hooper and Patel, 2012). Sandip Patel (University College London, UK) detailed the history of NAADP-dependent Ca2+ signaling in the sea urchin egg. He described the discovery of a new class of Ca2+-permeable channels, the two-pore channels (TPCs) as probable NAADP targets. It was also discussed how the principles of NAADP signaling in the egg can be applied to a range of other cells. Isabela Ramos (Brown University, Providence, USA) presented her work on the closely related starfish oocytes, which, in contrast to sea urchin eggs, are easily amenable to in vitro maturation, thereby allowing the knockdown of proteins upon injection of morpholino antisense oligonucleotides. By knocking down TPCs in fully grown oocytes, she demonstrated their role during the early development. Therefore, the starfish oocyte might prove to be a very useful model for the molecular characterization of the Ca2+ signaling machinery in echinoderm oocytes.
Just as the study of developmental mechanisms during embryogenesis (‘EvoDevo’) has helped to elucidate how developmental mechanisms have evolved in metazoans, a similar study of the mechanisms of reproduction (‘EvoRepro’) will help us to understand how they have been adapted or retained during animal evolution. Historically, the main models used to study oocyte maturation and fertilization were the animals that are amenable to in vitro fertilization and in vitro maturation. Genetic models did not provide much functional data because homozygous mutants for essential maternal factors contained in oocytes are difficult to obtain and maintain. This meeting revealed that the situation has now changed with the knowledge of genomes of non-genetic models from cnidarians, echinoderms, ascidians or bovine. Genomic and transcriptomic data from these different types of animals now allow for the proteomic identification of egg factors and make possible their specific knockdown in eggs that are amenable to in vitro maturation using approaches, such as RNAi and mopholino antisense oligonucleotides. Examples are Xenopus, mouse, bovine, ascidian, starfish and jellyfish eggs. EvoRepro studies will benefit greatly from these new molecular resources as they allow researchers to compare a greater number of species that encompass a more representative number of phyla in order to unravel the evolution of the mechanisms of sexual reproduction.
The authors thank the organizers and the participants of the meeting.