Interrogating the transcriptome of oocytes and preimplantation embryos

Abstract

During its growth phase, a mouse oocyte accumulates RNA that is the sole template for new protein synthesis in the transcriptionally silent interval between growth completion and transcriptional activation of the embryonic genome. Over this transcriptionally silent interval, almost half the quantity of RNA accumulated in the full-grown oocyte is degraded, while stable messages undergo major transcript-specific polyadenylation fluctuations associated with timely translation of new proteins. These processes, in the background of substantial RNA degradation, create unique pitfalls for transcriptome analysis. Three particular challenges are discussed herein. (1) Systematic errors of relative quantification occur if standard approaches are used, wherein samples are normalized to a constant quantity of RNA, or when computational analyses are normalized to an apparent "constant" endogenous to the sample. We show that use of a fixed quantity of exogenous RNA per oocyte or embryo alleviates this problem. (2) Comparison of large-scale expression analyses from widely disparate platforms highlights how the differing protocols produce correspondingly different lists of genes with significant changes in transcript abundance. Only with careful attention to the differences among experiments can such discrepancies be understood. (3) The complete assessment of changes in expression requires correspondingly comprehensive assessment of the role of isoform-specific changes.