Transcriptome analysis of bovine oocytes from distinct follicle sizes: Insights from correlation network analysis

Abstract

Follicle size is recognized as a predictor of the potential for the enclosed oocyte to yield an embryo following in vitro maturation and in vitro fertilization. Oocytes from larger follicles are more likely to reach the blastocyst stage than those from smaller follicles. A growing oocyte accumulates all the transcripts needed to ensure development until the maternal embryonic transition, and this accumulation must be completed before the period of transcriptional arrest. Accordingly, the transcriptomes of bovine germinal-vesicle-stage oocytes collected from follicles of increasing sizes (<3, 3-5, >5-8, and >8 mm) were evaluated, using the EmbryoGENE bovine transcriptomic platform (custom Agilent 4 × 44 K), to better understand transcriptional modulation in the oocyte as the follicle becomes larger. Microarray analyses revealed very few differences between oocytes from small follicles (<3 vs. 3-5 mm), whereas an important number of differences were detected at the mRNA level between oocytes from larger follicles. Weighted gene correlation network analysis allowed for the identification of several hub genes involved in crucial functions such as transcriptional regulation (TAF2), chromatin remodeling (PPP1CB), energy production (SLC25A31), as well as transport of key molecules within the cell (NAGPA, CYHR1, and SLC3A12). The results presented here thus reinforce the hypothesis that developmental competence acquisition cannot be seen as a simple one-step process, especially in regards to the modulation of mRNA. Mol. Reprod. Dev. 83: 558-569, 2016. © 2016 Wiley Periodicals, Inc.

Folliculogenesis and acquisition of oocyte competence in cows

“…As the follicle continues its growth, the chromatin starts to change and becomes more compact as GV-1, in clusters in GV-2 and as a very dense structure in GV3 oocytes (Dieci et al, 2016a). These changes are associated with the decrease in transcription but also involves several changes in the histones themselves (replacement with some H3.3) (Labrecque et al, 2016) and their post translational modifications (Lodde et al, 2017). This transformation is similar but somewhat different than the change from the non-surrounded nucleolus (NSN) pattern to the more compact form (surrounded nucleolus SN) as the mouse acquires development competence once it reaches its full size and the SN configuration while in large mammals the process is multi-step and terminates with the pre-ovulatory period where the oocytes have a GV-2 configuration and are ready to complete meiosis rapidly.…”

“…The progressive change in the chromatin configuration is not essential for meiotic resumption as GV-1 stage are fully capable of reaching the metaphase II in culture, but the developmental competence acquisition (Sirard, 2001) is not yet completed indicating that other changes must occurs in the chromatin, or in the same period as the chromatin changes. The transcriptome analysis of the oocyte according to the GV status has been done and revealed numerous histone variants changes (Labrecque et al, 2016) as well as hundreds of other modifications in the RNA content associated with genes that are stored or the ones that are translated during the maturation period (Gohin et al, 2014). The difficulty with RNA analysis in oocytes is that they store numerous transcripts through de-adenylation (leaving around 25 As) and the amplification systems that are used in molecular biology normally prime on the poly A tail and does not differentiate the short (stored) vs the long polyA tails (>100 As) that are rapidly translated creating a doubt if the transcript is used during the transition between GV stages or stored for maturation or post fertilization events (Gohin et al, 2014).…”

“…In somatic cells, the RNA as a half-life of a few minutes/hours while in oocyte an important fraction of RNAs are de-adenylated and stored (Tremblay et al, 2005;Gohin et al, 2014) Therefore, when a complete RNA analysis (transcriptome) is done (Robert et al, 2011) the results are completely masked by the ignorance of the timing of translation of RNAs into proteins (Gilbert et al, 2010) notwithstanding the fact that many proteins need a post translational modification to be active and such transformation needs hundreds of oocytes to be observed at a chemical level. In oocytes, as somatic cells, we easily observe 12000-15000 different transcripts coding for different proteins and without the capacity to distinguish their dynamics, the complexity of the analysis becomes overwhelming (Labrecque et al, 2013(Labrecque et al, , 2016Labrecque and Sirard, 2014). Studies to identify the translated RNAs has been done on very specific stages (GV and MII) using large amount of slaughterhouse oocytes (Gohin et al, 2014) but not on highly competent compared to less competent ones.…”

microRNA expression profile in porcine oocytes with different developmental competence derived from large or small follicles

“…This chromatin remodeling is associated with global transcriptional repression and ends up with fully grown oocytes being able to resume meiotic division (H. Liu & Aoki, ; Lodde et al, ; Tan et al, ). Oocytes from different‐sized follicles exhibited different transcript abundance with gradual modulation during follicular development (Labrecque, Fournier, & Sirard, ). In pigs, oocytes from 2 mm follicles exhibited a significant decrease in transcriptional activity, which is completely repressed in oocytes from follicles larger than 3 mm (Bjerregaard et al, ).…”

“…Oocytes from different-sized follicles exhibited different transcript abundance with gradual modulation during follicular development (Labrecque, Fournier, & Sirard, 2016). In pigs, oocytes from 2 mm follicles exhibited a significant decrease in transcriptional activity, which is completely repressed in oocytes from follicles larger than 3 mm (Bjerregaard et al, 2004).…”