An Analysis of Polycomb Repressive Complex 2 Function Through Imprinted Mouse X-chromosome Inactivation.

Abstract

Abstract Polycomb proteins comprise two major classes of evolutionarily conserved epigenetic transcription repressors, Polycomb repressive complex 2 and 1 (PRC2 and PRC1). PRCs are thought to catalyze epigenetic silencing via histone modifications and/or physical compaction of the surrounding chromatin. The inactive X-chromosome is a common target of PRCs. X-chromosome inactivation is a paradigmatic epigenetic phenomenon resulting in the equal expression of genes from the X-chromosome between XY male and XX female mammals. The initial form of X-inactivation during murine embryogenesis is imprinted X-inactivation, during which the paternally inherited X-chromosome is preferentially silenced. In my thesis work, I tested the hypothesis that PRC2 proteins orchestrate gene silencing on the paternal X-chromosome during imprinted X-inactivation. To test if PRC2 subunits are required to propagate the X-inactive state, I derived and investigated X-linked gene silencing in mouse trophoblast stem cells (TSCs), an ex vivo model of imprinted X-inactivation. In TSCs lacking the core PRC2 proteins EZH2 and its homologue EZH1, which catalyze trimethylation of histone H3 at lysine 27 (H3-K27me3), I found that X-inactivation was unperturbed. In TSCs lacking EED, which is required for the assembly of PRC2, I found that imprinted X-inactivation was defective. In Eed-/- TSCs, enrichment of H3-K27me3 and the Xist long non-coding RNA, which is required for stable X-inactivation, are lost from the inactive-X. Despite the absence of H3-K27me3 and Xist RNA, only a subset of the genes on the inactive X-chromosome is reactivated in Eed-/- TSCs. Lack of a silencing defect for a majority of X-linked genes in Eed-/- TSCs suggests that factors other than EED, H3-K27me3, and Xist RNA are essential for propagating X-chromosome inactivation. To assess if my findings from TSCs applied in vivo, I generated embryos lacking maternal and zygotic EZH2, or EZH2 and 1, or EED. I discovered that EED, but not EZH2/1, is necessary to trigger imprinted X-inactivation in the embryo. This comparative analysis of PRC2 components suggests a PRC2 independent role for EED in imprinted X-inactivation. Moreover, these results are the initial demonstration that maternal factors control the silencing of the X-chromosome in the embryo, an example of a transgenerational epigenetic regulation.