Deciphering differences in DNA methylation and transcriptome profiles of oocytes from pigs with high and low developmental competence

Abstract

In vitro maturation (IVM) is a critical step in animal in vitro embryo production, yet oocytes matured in vitro often exhibit lower developmental competence than their in vivo counterparts. However, the molecular mechanisms behind this observation remain unclear. This study investigated the gene expression and DNA methylation profiles in porcine oocytes with different developmental competencies. To study these differences, we used as a model oocytes from prepubertal gilts (IVM) and sows (in vivo matured) and assessed their developmental competence up to the blastocyst stage. We also examined their gene expression and DNA methylation profiles at single-cell resolution using RNA sequencing and bisulfite sequencing. Oocytes were obtained by aspiration of either ovarian follicles between 3 and 6 mm diameter, and the subsequent IVM, or ovarian follicles from 8 to 10 mm diameter, with no need for maturation (in vivo matured oocytes). Cleavage rates (58.2 ± 3.0 and 45.7 ± 4.4) and blastocyst rates (31.4 ± 3.7 and 47.5 ± 6.6) for IVM and in vivo groups differed significantly. Using the in vivo group as a reference, IVM oocytes had 1297 downregulated and 476 upregulated differentially expressed genes (DEGs), with upregulated DEGs associated with organelle organization and cell cycle processes, and downregulated genes involved in protein synthesis and metabolomic processes. While global DNA methylation levels were similar between groups, a few differentially methylated regions were found in CpG islands, promoters, and coding regions. Our integrative analysis identified key methylated regions and genes that distinguish each group, suggesting that both donor age and maturation conditions significantly influence gene expression regulation in oocytes with different developmental competencies.

Keywords: cellular division; epigenomic; multi-omic analysis; oocyte maturation; porcine.

Figure 1.

(a) Visualization of highly variable genes from the porcine oocyte pool obtained from single-cell RNA sequencing data using Seurat. Each point represents a single gene, with those in red representing the 2000 genes with the highest variability. (b) UMAP plot shows distribution of RNA-sequencing samples, where colours indicate the in vitro and in vivo treatments. (c) Heatmap showing the gene expression profile of porcine oocytes. The red colour indicates lower expression, while the green colour indicates a higher level of expression. (d) Volcano plot of DEGs in the samples obtained by in vitro and in vivo culture. Each point represents a single gene; in blue are shown those genes whose expression is decreased and, in red, those whose expression is increased when comparing in vitro vs. in vivo matured oocytes. Genes with an adjusted P-value < .05 and an FC > or <±1.5 were considered statistically significant.

Figure 2.

Functional enrichment analysis of (a) genes with higher expression when comparing in vitro vs. in vivo groups and (b) genes with lower expression when comparing in vitro vs. in vivo group was performed using ClueGO software within the Cytoscape environment. The analysis was performed with the ClueGO 2.1.3 plugin of the Cytoscape 3.9.1. The size of the nodules represents their significance; different pathways are represented with different colours. (c) The top 10 of upregulated and (d) downregulated TFs after comparing the gene expression profile of porcine oocytes cultured in vitro vs. in vivo.

Figure 3.

Direct targets of TDFP1 up-expressed in the IVM porcine oocytes compared to the in vivo group. All genes are significantly upregulated by TDFP1, and are predicted as TDFP1 targets by motif discovery in iRegulon.

Figure 4.

Violin plots of the distribution of average methylation values at both IVM and in vivo groups for the analysis performed on (a) CGIs, (b) promoters, (c) coding regions, and (d) transposons.

Figure 5.

UMAP plots to visualize the intrinsic structure and relationships between samples in the dataset. The cells are coloured based on the factor they belong to (a, b) and the score of each factor in each cell (c). Each point in the plot represents a sample, and its position is determined by nonlinear dimensionality reduction.

Figure 6.

Top 100 of the regulatory relationships between gene expression (RNA-seq) and methylation (promoter regions, VMR and CGIs) in the IVM (a) and in the in vivo (b) matured oocytes. Methylation is represented by red labels and genes by black labels. Positive ratios are shown in red and negative ratios in blue. The thickness of the line indicates the strength of the correlation.

Figure 7.

Functional enrichment analysis of (a) genes identified in the in vitro cluster (factor 1) and (b) genes identified in the in vivo cluster (factor 2) was performed using ClueGO software within the Cytoscape environment. The analysis was performed with the ClueGO 2.1.3 plugin of the Cytoscape 3.9.1. The size of the nodules represents their significance; different pathways are represented with different colours.