Thesis title: High-Resolution Multimodal Assessment of Mouse Embryonic Genome Activation
Embryonic Genome Activation (EGA) is the onset of productive transcription in the post-fertilization embryo, following a transcriptional quiescent state inherited from the gametes. In the mouse, EGA happens in two waves: a minor wave in the zygote and a major wave at the 2-cell stage. EGA occurs simultaneously with widespread epigenomic changes, and although regulators of EGA have been identified, the process is poorly understood. Overall, a detailed characterization of EGA is lacking, which is necessary to address the molecular regulation underlying its dynamic, stepwise nature. As a consequence of EGA, morphologically identical embryos harbor vastly different transcriptomes and cannot be directly compared. Here, I optimize in vitro fertilization (IVF) to enable precise staging of mouse embryos. This precision-IVF approach allows for the capture of embryos undergoing EGA. Single-embryo RNA-seq reveals that EGA occurs rapidly and in a stereotypical, robust manner, with gene sets activated and depleted at different time points during a 9-hour timecourse. During this time approximately 30% of all detectable genes change in expression levels, highlighting the widespread nature of EGA and the simultaneous depletion of maternally inherited transcripts. Using the high-resolution data, I identify histone demethylases among the first genes induced during early-EGA. Specifically, EGA happens concomitantly with a progressive removal of maternally inherited H3K4me3, which has been linked to transcriptional repression in oocytes. However, overexpression of Kdm5b, a H3K4me3 demethylase, in fully grown oocytes and embryos does not appear to affect global transcription levels or timing of EGA.
To address how EGA is coordinated with underlying chromatin changes, I developed Multiome of Embryonic Transcriptome and ATAC-seq (META-seq), a multiomics approach for assessing chromatin accessibility and the transcriptome, optimized for low-input embryonic samples. Here, I couple META-seq with precision-IVF. I find that chromatin accessibility is largely remodeled at the mid-to-late EGA stage and is associated with the closure of repetitive elements, which are open at early EGA. These TEs, including MERVL and SINE B1 elements, continue to open during EGA until they are closed at late-EGA. Motif analysis using the diffTF workflow identifies Zscan21 as a putative EGA regulator. Zscan21 protein accumulates in the nuclei during the 2-cell stage, and its presence correlates with EGA. Computational analysis identified Zscan21-predicted binding sites in SINE B1 elements, overlapping with the binding motif of the known EGA regulator Obox, suggesting a cooperative interaction between these transcription factors.