SEARCH

SEARCH BY CITATION

Contents

  1. Top of page
  2. Contents
  3. Introduction
  4. The Follicular Phase
  5. The Transcription-Free Phase
  6. The Embryonic Genome Activation Phase
  7. Conclusion
  8. Acknowledgements
  9. Conflicts of interest
  10. References

The most important factor affecting the oocyte and early embryo transcriptome is the legacy from the follicular environment prior to meiotic resumption. Up to the 8-cell stage, the oocyte responds to maternal instructions stored before resumption of the meiotic division. Recent evidence suggests that properly prepared or programmed oocytes (in vivo) can achieve close to 100% blastocyst rates in standard in vitro conditions/media. Therefore, the optimal oocyte requires perfect follicular timing and differentiation, but the intra-oocyte mechanisms involved in such preparation are not completely understood. In addition, the influence of maternal mRNA storage and degradation, as well as the length of the poly A tail that influences the general pattern of the oocyte/early embryo transcriptome, is an important factor. Several hypotheses have been put forth to explain the depletion of the maternal store, including the potential role of microRNA (miRNA) in this process. The activation of the embryonic genome could be dependent on, or associated with, the process of maternal mRNA degradation, but obviously other functions are being activated at this critical time point. This review will focus on the period from full-size oocytes to the eight-cell stage and will summarize the impact of the important factors, that is, follicle, maternal RNA storage and embryonic genome activation, on the transcriptome.


Introduction

  1. Top of page
  2. Contents
  3. Introduction
  4. The Follicular Phase
  5. The Transcription-Free Phase
  6. The Embryonic Genome Activation Phase
  7. Conclusion
  8. Acknowledgements
  9. Conflicts of interest
  10. References

A number of parameters will likely influence the oocyte/embryo transcriptome; although we are only at the beginning of this gigantic exploration, some parameters are becoming important to consider. Using the most complete bovine microarray available, which includes nearly all 22 000 known bovine genes, 6000 splice variants, 9000 non-coding regions and 4000 3′UTR variants identified by a RNAseq experiment of more than a million sequences obtained from bovine oocytes and early embryos, we observed an average of 15 000 genes or variants being expressed (two standard deviations above background) in oocytes and a similar number in blastocysts (Robert et al. 2011).

Of these 15 000 genes, a large majority are not influenced by the environment or developmental competence (Mamo et al. 2011) but certainly by the embryonic stage, as embryos have a dynamic developmental process requiring numerous changes in gene expression as they differentiate (Vallee et al. 2009).

Another level of complexity is brought by the unique ability of oocytes to store RNA in a stable form which cannot be easily separated from the RNA being actively translated. Therefore, before the eight-cell stage in cattle, RNA levels are not directly informative unless confirmed by protein analysis.

To sort the factors affecting the transcriptome of oocytes and embryos, I will refer mainly to the bovine as the model of choice and I will discuss the following three perspectives:

  • 1
     The follicular phase.
  • 2
     The transcription-free phase.
  • 3
     The embryonic genome activation phase.

The Follicular Phase

  1. Top of page
  2. Contents
  3. Introduction
  4. The Follicular Phase
  5. The Transcription-Free Phase
  6. The Embryonic Genome Activation Phase
  7. Conclusion
  8. Acknowledgements
  9. Conflicts of interest
  10. References

The follicle–oocyte duality

In the female mammal, the oocyte and the follicle are intimately linked from the early phases of gametogenesis. They form a team in which the death of one component normally leads to the death of the other. Up until the oocyte reaches its full size, follicular cells provide all types of substrates, metabolites as well as complementary proteins necessary for oocyte function. The competence of the fully grown oocyte to resume meiosis is reached at a diameter of approximately 70 μm in the mouse, whereas the ability to be fertilized and undergo normal development is only achieved when the oocyte reaches maximum size (First et al. 1988). In contrast, in domestic animals, acquisition of developmental competence is not necessarily or even normally achieved in full-size oocytes. Other differentiation events related to follicular differentiation are clearly required for full developmental competence to be achieved. The reasons why fully grown oocytes are not already programmed to become viable embryos in large mammals remain elusive. A possible hypothesis is that fully totipotent oocytes may represent a threat to the ovary if not ovulated. In support of such hypothesis is the phenotype of cMos double KO mice, where the oocytes become parthenogenetically activated in the follicle and all females die of ovarian cancer before reaching their first birthday (Hashimoto et al. 1994). In this hypothesis, the oocyte represents a threat and the follicle equipped with LH receptors represents a way to send the oocyte to an environment less favourable to cancer growth, that is, the oviduct.

Follicular size and health effect

Follicle size is an important parameter that may influence the competence of the oocyte and indirectly its RNA content. Smaller bovine and porcine follicles (<3 mm) have reduced oocyte competence. Follicular effects on competence have been evaluated in a few studies, but most of these pools were made according to follicle size (Pavlok et al. 1992) (Lonergan et al. 1994). A few studies have described the fate of individual oocytes according to the follicular size, confirming an increased competence with follicle size (Blondin and Sirard 1995; Hagemann et al. 1999). Even cloning experiments support the concept of increased developmental competence of oocytes from large follicles (Barnes et al. 1993). Although size has a significant effect, some oocytes originating from large follicles still fail to produce embryos, whereas some oocytes from medium-sized follicles (5–19 mm) already have this capacity.

Follicular atresia starts with signs of follicular cell apoptosis from the granulosa cells that progressively move from outside to inside the cumulus and finally to the embedded oocyte (Zeuner et al. 2003). Dominant follicles demonstrate no or very slight atresia and their oocytes preserve good developmental competence (Vassena et al. 2003). Bob Moor was the first to report that oocyte viability was unrelated to the degree of atresia of sheep follicles cultured in vitro to mature oocytes for fertilization and development to term (Moor and Trounson 1977). Similar observations have been noted in rats (Tsafriri and Pomerantz 1984) and other species including humans (Barnes et al. 1993). In cattle, we were the first to demonstrate that higher developmental competence was achieved in bovine COC with slight signs of apoptosis in the outer layers of the cumulus (Blondin and Sirard 1995; Zeuner et al. 2003; Feng et al. 2007). Surprisingly, the group of oocytes with the highest competence possessed a less compact cumulus and originated from healthy or slightly atretic follicles, as measured by histology or flow cytometry (Blondin et al. 1996b). These early apoptotic events could be seen as a form of maturation-promoting/accelerating signals to the oocyte to improve its competence. A recent study (Salamone et al. 1999) revealed that during the growth of the first follicular wave, oocytes from the subordinate follicles remain viable through the static phase, but undergo degenerative changes in the regressing phase following the appearance of the dominant follicle. Another study evaluated the number of blastocysts obtained when aspiration was made on days 3, 4 or 5 of the first follicular wave; three times more viable embryos were produced at day 5 when dominance has occurred, compared with day 3 where all follicles are growing (Machatkova et al. 2000). A recent paper (Lodde et al. 2008) described the progressive chromatin condensation of the nucleus in bovine oocytes as they are harvested from follicles of different size and health status. My interpretation of their results is that transcription is progressively shut down in follicles approaching ovulation or in subordinate follicles beginning atresia and has the same consequence: better ability to develop if the transcription machinery is off.

The limitation with abattoir-derived material is the lack of knowledge of follicle status, which could originate from different days in the cycle or even from early pregnancy. To better assess the follicular status of medium and large follicles, it becomes necessary to use exogenous FSH at specific times during the sexual cycle. Following 2, 3 or 4 days of FSH (twice daily) treatment, the developmental competence of the oocytes obtained from large (>8 mm), growing, healthy follicles was inferior to the competence of control oocytes aspirated from non-stimulated animals (11 vs 32%, respectively) (Blondin et al. 1996a). These results suggest that healthy looking cumulus-oocyte complexes from actively growing follicles possess a reduced developmental competence when compared to oocytes obtained from non-treated animals. This would mean that the oocytes contained in actively growing follicles are still transcriptionally active and less ready to begin meiotic resumption.

Follicular influence on oocyte transcription level

Clearly, the oocyte is almost constantly changing, as is the follicle from the primordial stage to ovulation. It seems that even when the oocyte reaches its final size, the transformation continues. The terminal change occurs when the oocyte progressively decreases its transcriptional activities (Fair et al. 1995). This transcriptional decline was demonstrated by both uridine incorporation experiments and a progressive change in the nuclear and nucleolar structures during the dominant pre-ovulatory period (Tan et al. 2009) (Lodde et al. 2008). For the mouse, rabbit, sheep, goat, cow, horse and human, the progressive configuration from diffuse to more condensed and from non-surrounded nucleolus (NSN) to surrounded nucleolus (SN) is associated with a progressive shutdown of transcriptional machinery (Escrich et al. 2009; Tan et al. 2009). The logical conclusion from such a transformation is that the oocyte finally has all the RNA required to proceed; without it, its developmental capacity may be impaired (Fig. 1).

image

Figure 1.  Schematic of chromatin condensation in the oocyte as a result of final follicular differentiation

Download figure to PowerPoint

Oocyte durability

Changes associated with final oocyte preparation could indicate that when the oocyte has begun the process of final differentiation, transcription slows down and results in a decrease of the production of new mRNA. One of the specific abilities of the oocyte is to go through several cell divisions without any transcription occurring. This rare phenotype implies that the RNA required for keeping the cell alive and dividing must be accumulated and stored. In somatic cells, newly synthesized RNA lasts 2–3 h and is then degraded. In cows, some oocyte RNAs may have to survive a minimum of 7 days to ensure enough material until the new embryonic genome can establish a new individual. This ability to live without transcription and to rely on very stable RNA is unique to the oocyte, with the exception of nerve cells where long-lasting RNA is also produced to travel the distance between the nucleus in the spine and the peripheral axon termination (Holt and Bullock 2009). One of the visible signs of transcription shutdown is compaction of the chromatin. Morphological evaluation of ovarian slices (histology) indicates that this condensed chromatin structure is present in follicles that either ovulate or disappear within 2–4 days (atresia). Very few studies have been carried out to examine the duration of the dominant follicle in cattle to see how long oocytes with compacted chromatin can survive. It is known that postponing ovulation by only 1 day using progesterone implants results in decreased fertility (Sirois and Fortune 1990). We have performed several experiments on coasting in cows by preventing ovulation using endogenous progesterone to show that 72 h after the last FSH dose, oocyte quality decreases rapidly (Sirard et al. 1999).

One factor is clear: like any somatic cell, the oocyte needs to renew its proteins to survive, which are being turned over at defined rates. This turnover requires new RNA for replacement and functional ribosomes to make them. As RNA transcription stops or seriously diminishes a few days before ovulation, different scenarios are possible: RNA required for translating housekeeping proteins is decreasing, and the survival of the oocyte is progressively compromised, or the stored RNA for the oocyte’s/embryo’s requirements after ovulation is not maintained so the oocyte can go through fertilization and early cleavage, or specific competence RNA gets degraded which results in a compromised embryo incapable of activating its own embryonic genome. Regardless, the result is reduced fertility.

Follicular influence on oocyte RNA species

As transcriptomics was only recently developed and oocytes are not an abundant source of material, very few studies have looked at the impact of follicular characteristics on oocyte RNA stores. In a paper from Mourot et al. (2006) using bovine ovaries, several target RNAs (genes) were investigated in relation to follicular size, as a proxy for competence, with the understanding that the relation is not linear between size and competence. Six of these genes are shown (Fig. 2); it is interesting to note that for four of them, there is a gradual increase in RNA in oocytes, whereas for the other two, the rise occurs only after the follicle has passed the 8-mm size, which is the size where LH receptors begin to appear (Ginther et al. 2003). This simple demonstration illustrates that the process of messenger RNA accumulation is not general but works through specific steps. This also supports the important role of final differentiation of the follicle where specific mRNAs are produced and probably accumulate after the oocyte has reached its final size (in a 1- to 2-mm follicle). The role of a stepwise competence accumulation is potentially related to the importance of limiting the number of embryos in farm animals, especially in mono-ovulating species. The ovary would then have a better control of oocyte quality vs number.

image

Figure 2.  Specific oocyte RNA levels associated with different follicular sizes [adapted from (Mourot et al. 2006)]

Download figure to PowerPoint

The other intriguing observation from the results of Mourot et al. (2006) relates to the genes involved. Several of the RNAs identified as increasing in the latter part of follicle differentiation are cell cycle genes. Cyclins (like CCNB2) are required at every cell cycle as they are degraded between divisions. If an embryo has a limited supply of cyclins, this may impair its ability to reach the eight-cell stage or block cell division at earlier stages. The same argument could be applied to CSK1B which controls cyclin activities or PTTG1 required for chromosome segregation. Together, these results could support the simple hypothesis that the ability to cleave several times is improved as the follicle grows. It may sound simplistic, but would work well. We have shown that depleting CCNB1 RNA in bovine oocytes using RNAi would result in increased self-activation (return to interphase) and then block further cell division (Paradis et al. 2005).

The Transcription-Free Phase

  1. Top of page
  2. Contents
  3. Introduction
  4. The Follicular Phase
  5. The Transcription-Free Phase
  6. The Embryonic Genome Activation Phase
  7. Conclusion
  8. Acknowledgements
  9. Conflicts of interest
  10. References

It has previously been described that RNA accumulates in oocytes in a stable form. Xenopus is the most studied model to assess mRNA fate in animals. The dominant hypothesis is that maternal RNA is stored using a specific configuration where the mRNA is de-polyadenylated at the 3′ end and capped at the 5′end. The storage of mRNA is then associated with ribonucleoproteins, which repress translation. In Xenopus, more information is available about the various proteins involved in the repression of translation. One such protein is maskin, which associates with the cytoplasmic polyadenylation element binding protein (CPEB) located in the 3′ UTR region on mRNAs that contain a cytoplasmic polyadenylation element (CPE). This complex represses translation through the inhibitory action of maskin-elF4E located at the 5′ end of the RNA (Richter 2007). Translation of specific mRNAs then proceeds according to a combination of cytoplasmic codes acting on RNA-associated proteins interacting with the 3′ UTR sequence of the stored RNAs. Activation of translation seems to be associated with a longer polyA tail (80–150 and longer), whereas short tails (approximately 20 As) are repressed from translation (Richter 1999). A combination of regulatory motifs and proteins has recently been identified as regulators of re-polyadenylation of maternal RNAs (Pique et al. 2008). Progressively, the 3′UTR code is being understood, but the precise spatio-temporal regulation of mRNAs during the transition from maternal to embryonic control still awaits further insights.

Destruction of targeted maternal RNA

Not only does maternal mRNA get recruited at specific times, some mRNA gets degraded without translation. At least two degradation pathways are known to influence maternal RNA fate. The first one is somewhat similar to the recruitment for translation and is controlled by maternally encoded factors that target the 3′ UTR region at specific motifs (Stitzel and Seydoux 2007), whereas the second coincides with embryonic genome activation (EGA) and may also use 3′ UTR targets. Sequencing of these regions in Drosophila has revealed families of RNA-binding proteins associated with RNA fate like SMAUG and Pumilio (De Renzis et al. 2007). Clearly not all maternal RNA is suddenly destroyed at the time of EGA. Transcriptomic analysis of late zygotes indicated that approximately 60% of de novo transcripts are novel for the embryos, whereas the remaining 40% were already present in the maternal pool in mouse (Hamatani et al. 2004). In mice, unstable mRNAs (the ones that disappear) are also associated with the cell cycle, but the RNA associated with RNA processing seems to come from the new genome, rather than from the protected maternal pool (Hamatani et al. 2004). A second wave of new transcripts appears between the four- and eight-cell stages, and the new gene products are likely involved in the beginning of the compaction that precedes the differentiation of ICM into trophoblast cells.

MicroRNA

MicroRNAs are short non-coding RNAs of 19–24 nucleotides, known to regulate genes by targeting their 3′ untranslated regions (3′ UTRs) in a manner that is not dependent on total homology. The post-transcriptional regulation processes of miRNA make them interesting candidates to control maternal transcripts in the early embryo. The production of miRNA starts with the primary miRNA (pri-miRNA), transcribed in the nucleus and cleaved by the Microprocessor complex consisting of Drosha and DGCR8. The precursor miRNA (pre-miRNA) is obtained by cleavage of the double-stranded hairpin-shaped molecule, which is exported into the cytoplasm. The pre-miRNA is further cleaved by Dicer to produce the double-stranded miRNA duplex. The passenger strand is discarded, and the remaining strand forms the mature miRNA to be integrated into the RNA-induced silencing complex (miRISC). Coupled with the RISC complex, miRNAs induce translational repression and/or promote de-adenylation of mRNAs (Filipowicz et al. 2008). Obviously, miRNAs are interesting candidates for the control of maternal mRNA degradation in the embryonic transcriptional silence context because they regulate gene expression in a temporal- and spatial-specific manner. Recent studies showed the implication of miRNAs in the development of some species, especially in the degradation of the maternal transcripts. In C. elegans, miRNAs regulate and allow the transition between larval stages (Lee et al. 1993). In zebra fish, a single miRNA (miR430) drives the destruction of several hundreds of maternal RNAs (Giraldez et al. 2006). The Xenopus ortholog of this miRNA, MiR-427, leads to the de-adenylation and depletion of hundreds of maternal mRNAs in the frog embryo (Watanabe et al. 2005; Lund et al. 2009). A similar observation with Mir-21 has been recently made in the trout (Ramachandra et al. 2008). This miRNA is highly expressed just before EGA where an important regulatory mechanism must be applied, supporting the hypothesis that miR-21 is processed during EGA to act as a guide for maternal mRNA degradation (Ramachandra et al. 2008). Therefore, these upregulated miRNA candidates at EGA, like miR-21, suggest the functional contribution of some miRNAs in embryonic development.

In mammals, degradation of non-translated mRNAs was also reported to be a critical step for allowing EGA (Alizadeh et al. 2005). Although the mechanisms of miRNA action are not well defined, Schier was one of the first to hypothesize the potential role of miRNA as a control mechanism in the degradation of maternal mRNAs (Schier 2007). Dicer knocked out mouse oocytes and embryos are committed to die; they block at the first meiosis and embryos stop their development before the gastrulation stage (Bernstein et al. 2003; Murchison et al. 2007). Alternatively, a recent study showed that miRNAs seem to have a very minimal impact on the abundance of mouse maternal transcripts (Ma et al. 2010). Surprisingly, the lethal phenotype caused by Dicer KO could be explained by deregulation of the biogenesis of siRNAs (Bernstein et al. 2003; Ma et al. 2010).

In pigs, the role of miRNAs also appears to be important for the degradation of maternal material, as loss of function of EIF2C1 results in the stabilization of specific maternal RNAs and development arrest during the maternal embryonic transition (MET). EIF2C1 is part of the RISC complex required to process miRNAs (Whitworth et al. 2005).

Very few studies have investigated the role of miRNAs during the bovine maternal-to-zygotic transition, and no complete oocyte/embryo-specific libraries are available at this time. Quantification of miR-10 and miR-424 in early bovine development showed a steady expression level from the GV oocyte until the 16-cell embryo (Tripurani et al. 2010). Different profiles were presented for miR-125, miR-127 and miR-145, which showed an increased expression level by the four- and eight-cell stages (Tesfaye et al. 2009). More recently, miR-196a was shown to increase steadily from the two- to the eight-cell stage (Tripurani et al. 2011). Additionally, they showed potential evidence of a direct negative regulation of a maternal transcript, the NOBOX element, by miR-196a. In our laboratory, we explored the roles of miRNAs in the course of bovine early development by identification of differently expressed miRNAs (by microarray) and then quantified selected candidates in oocytes and pre-implantation stage embryos. Our results showed that two miRNAs, miR-21 and miR-130a, accumulated in a linear fashion between the zygote and the eight-cell stage. To confirm this finding, the pre-forms (miRNA precursors) were also measured across stages to reveal a pattern similar to the mature forms. Finally, to prove that these new miRNAs were a result of de novo transcription, we cultured the post-fertilization embryos in the presence of alpha-amanitin which ablated the rise in the miRNAs at the two-cell stage (Mondou et al., 2012). These unexpected patterns could indicate an activity of miRNAs on the cellular processes associated with the mRNAs carrying a seed sequence in their 3′UTR during the first segmentations of the embryo. For miR-21, the results highlight a clear increase in the precursor form between the MII stage and the two-cell embryo, which could be associated with the hypothetical minor transcription observed at that stage in bovine embryos (Memili et al. 1998). This increase is particularly interesting because there is no other known transcription at the bovine two-cell embryo stage (Kanka et al. 2009).

The Embryonic Genome Activation Phase

  1. Top of page
  2. Contents
  3. Introduction
  4. The Follicular Phase
  5. The Transcription-Free Phase
  6. The Embryonic Genome Activation Phase
  7. Conclusion
  8. Acknowledgements
  9. Conflicts of interest
  10. References

As it becomes difficult to assess future competence of embryos at the time they are harvested for RNA analysis, scientists must proceed with group analogy where the competence is estimated from a control group left in culture or in vivo conditions. Obviously, in vivo-generated oocytes or embryos represent the ultimate control, but these are very difficult to obtain (Vallee et al. 2009). The time to the first embryonic division has been used to discriminate oocytes with different capacities to develop in vitro in cattle (Wrenzycki et al. 2007) The separation of gametes according to their quality has been demonstrated in females of several animal species in which it has been observed in vitro that zygotes that cleave earlier produce much higher blastocyst rates than the slow ones (Lonergan et al. 1999; Gutierrez-Adan et al. 2004; Dode et al. 2006; Mourot et al. 2006; Patel et al. 2007). This last study reported that bovine fast-cleaving two-cell embryos possess distinct levels of expression of genes such as PTTG1, CKS1, YEAF, IDH, CycB1 and Follistatin, involved in crucial biological functions, in comparison with slow-dividing two-cell embryos. It must be re-stated that RNA levels before the eight-cell stage are not directly a demonstration of function as stored RNA has no impact on function but more RNA means a potential for more function.

In cattle, the transcriptional machinery is activated between the eight- and 16-cell stages (Barnes and First 1991). More recently, the possibility of a minor embryonic activation was suggested (Memili and First 1999), based on the appearance of novel proteins during the two- to four-cell stages (Memili and First 2000). The possibility that maternal RNA is re-polyadenylated could be considered as one explanation for the sudden appearance of some undetected RNAs (with a very short polyA tail). Because maternal RNA is stored as de-polyadenylated, the length of the polyA tail is an issue. When the tail is short, it requires extraction protocols that do not use polyT columns or filters. If the primers used for reverse transcription include a polyT sequence, it will exclude an undefined portion of the stored RNA, and finally, if a T7 polymerase is used for RNA amplification in microarray experiments, the presence of a polyA tail will likely make a difference (Gilbert et al. 2009).

Embryonic genome activation

In the mouse, maternal-to-embryonic transition is difficult to analyse owing to the rapid activation of the embryonic genome at the end of the first cell cycle. It is clear from that species that a number of specific factors or transcripts eventually translated into proteins are required for the proper genomic function of the embryo. Regarding maternal degradation, in addition to the factors mentioned above like Dicer, Ago2 (eukaryotic elongation factor 2c) and Atg5 (autophagy related 5) are required (Tsukamoto et al. 2008). Genes like Hr6a (Ubiquitin conjugating enzyme E2A), Nucleoplasmin 2, Tif1a (tripartite motif containing 24) and Smarca4 are believed to be important for chromatin remodelling. Transcription factors such as Oct-4 and Sox 2 influence early differentiation of blastomeres (Zuccotti et al. 2009). The re-establishment of methylation with Dnmt3a and Dnmt3c and their impact on the ability of DNA to be transcribed are associated with specific gene transcription. To this group of better known actors, some maternal-specific genes are slowly being added as they are being characterized: Zar-1, Mater, Padi6 and Filia as upstream factors in oocyte differentiation.

Embryonic genome activation in pigs also occurs later than in the mouse, at the four-cell stage. A recent review (Prather et al. 2009) indicates the usefulness of the genomic tools (microarrays and subtractive libraries) to identify the players involved in this process. Using an array dedicated to this species, even though the genome was incomplete at the time, they profiled the transcriptome of the MET period in pig embryos (Whitworth et al. 2005).

In cows, several studies investigated the period of MET using molecular subtraction from various states (Kanka et al. 2009; for a review, see Sirard 2010). The subtraction libraries obtained between the eight-cell plus or minus alpha-amanitin are enriched with RNA processing elements (Vigneault et al. 2009). Although the list of ‘early’ embryonic genes is short (<200 genes), several were found in common with the same type of approach used at the 4-cell stage in rabbit (Leandri et al. 2009): H2Afz (histone H2A), SAP18 (Sin-3-associated proteins), SLC family (solute carrier family), Mthfd1 (Methylenetetrahydrofolate deshydrogenase), ATP 5b (ATP synthase), EIF5 (eukaryotic translation initiation factor 5. Although these subtracted libraries are necessarily incomplete, the number of similarities between species is interesting from a comparative biology perspective. As expected, several maternal genes that do decrease during the MET are also common between the rabbit and the bovine studies (Leandri et al. 2009) (Vigneault et al. 2009).

The group of George Smith at Michigan State University (Bettegowda et al. 2008) has made a list of important genes that are inherited from the maternal pool for continued development or proper embryonic activation. Some of these genes like mater (Pennetier et al. 2006), Zar-1 (Pennetier et al. 2004) and NPM2 (Vallee et al. 2008), which lead to embryonic arrest phenotypes in the mouse, are also present in cattle, but others like stella, HsF1 (heat shock factor 1), RAD 6 and basonuclin have not been identified yet in cattle (Bettegowda et al. 2008). The same group has recently established the importance of a new gene; importin-alpha 8 or KPNA7, as a nuclear importer of specific cargo like DNMT1 and HDAC4, and which could be involved in embryonic initiation of transcription (Tejomurtula et al. 2009).

Conclusion

  1. Top of page
  2. Contents
  3. Introduction
  4. The Follicular Phase
  5. The Transcription-Free Phase
  6. The Embryonic Genome Activation Phase
  7. Conclusion
  8. Acknowledgements
  9. Conflicts of interest
  10. References

The quest for a complete understanding of all the processes involved in the transition from a maternal cell (oocyte) to a new embryo is just beginning. The scientific community still faces many challenges. First of all, the availability of the biological material is limited by the size of oocytes and embryos in mammals. Also, the ovulation rate or the number of good oocytes per animal at a given time in the sexual cycle is also limiting. The target that we are studying is not stable in time, but has a very dynamic gene expression pattern compared to somatic tissues. To further compound the issue, the transcriptome has little physiological value unless validated by proteins or KO studies, as the RNA measured is either stored and not translated or being translated and degraded. Finally, because several genes are specific to the oocyte or early embryo, their functions have not been identified as such characterization often requires KO studies or chemical investigation/purification of the corresponding proteins. Nevertheless, this field is attracting intense interest as the early phase of life seems to dictate the future of each individual.

References

  1. Top of page
  2. Contents
  3. Introduction
  4. The Follicular Phase
  5. The Transcription-Free Phase
  6. The Embryonic Genome Activation Phase
  7. Conclusion
  8. Acknowledgements
  9. Conflicts of interest
  10. References