Genome-wide expression profiling reveals distinct clusters of transcriptional regulation during bovine preimplantation development in vivo

Abstract

Bovine embryos can be generated by in vitro fertilization or somatic nuclear transfer; however, these differ from their in vivo counterparts in many aspects and exhibit a higher proportion of developmental abnormalities. Here, we determined for the first time the transcriptomes of bovine metaphase II oocytes and all stages of preimplantation embryos developing in vivo up to the blastocyst using the Affymetrix GeneChip Bovine Genome Array which examines approximately 23,000 transcripts. The data show that bovine oocytes and embryos transcribed a significantly higher number of genes than somatic cells. Several hundred genes were transcribed well before the 8-cell stage, at which the major activation of the bovine genome expression occurs. Importantly, stage-specific expression patterns in 2-cell, 4-cell, and 8-cell stages, and in morulae and blastocysts, were detected, indicating dynamic changes in the embryonic transcriptome and in groups of transiently active genes. Pathway analysis revealed >120 biochemical pathways that are operative in early preimplantation bovine development. Significant differences were observed between the mRNA expression profiles of in vivo and in vitro matured oocytes, highlighting the need to include in vivo derived oocytes/embryos in studies evaluating assisted reproductive techniques. This study provides the first comprehensive analysis of gene expression and transcriptome dynamics of in vivo developing bovine embryos and will serve as a basis for improving assisted reproductive technology.

Fig. 1.

Global analysis of transcriptome features in naturally developing bovine embryos. (A) shows Pearson correlation coefficients between replicates from all of the bovine embryonic stages as a “heat map” in which the color spectrum ranging from orange through yellow to green represents Pearson Correlation Coefficient values ranging from 0.519 to 1, and hierarchial clustering of various developmental stages. (B) Number of differentially regulated genes (>2-fold) of subsequent stages. (C) Stage-exclusive expressed genes (P call percentages).

Fig. 2.

Stage-specific onset of gene expression before major genome activation. Array data, white columns; Real Time RT-PCR with in vitro embryos, black columns; Real Time RTPCR with in vivo derived embryos, shaded columns.

Fig. 3.

Identification of new expression profiles by cluster analysis (grouping of genes by various algorithms (f.ex. K-means) according to their similarity in expression profiles across a set of experiments or samples). The expression graphs were generated by the K-Means clustering algorithm. (Top row) Three clusters showing maternal expression profiles and high, medium, or low expression at zygote stage. (Middle row) Three clusters with embryonic expression profiles and high, medium, or low expression at the blastocyst stage. (Bottom row) Two clusters with 8-cell-specific up and down-regulated transcript levels. Embryonic stages are depicted on the x axis.

Fig. 4.

Identification of oocyte enriched genes and of genes differentially expressed between in vivo and in vitro produced embryos. (A) Oocyte enriched genes. (B) Genes differentially expressed between in vivo and in vitro oocytes.