A Comparative Analysis of X-chromosome Inactivation in Early Mouse Embryonic Lineages

Abstract

X-chromosome inactivation is a dosage compensation mechanism that equalizes X-linked gene expression between XX female and XY male therian mammals. Two forms of X-inactivation take place in mice. Preimplantation female mouse embryos undergo imprinted X-inactivation of only the paternally-inherited X chromosome, which is subsequently maintained in the extraembryonic tissues of the developing embryo. The inactive paternal-X, however, is reactivated in cells of the mouse embryonic epiblast, which later undergo random inactivation of either the paternally- or maternally-inherited X chromosome. Random X-inactivation is then maintained through somatic cell divisions. X-inactivation silences almost all genes on one of the two X chromosomes in females. Some genes, however, escape X-inactivation and are expressed from both the active and inactive X-chromosomes. Thus, X-inactivation escapees are positioned to be expressed at higher levels in females vs. males, which may confer dose-dependent female-specific functions. The genomic features that contribute to escape from X-inactivation are lacking. My thesis work investigates genomic elements that underlie escape from X-chromosome inactivation in early mouse embryonic lineages. I delineated genes that are statistically likely to escape X-inactivation in the three primary lineages of the early mouse embryo: the trophectoderm (placental lineage), the primitive endoderm (yolk-sac), and the embryonic epiblast (fetal lineage). I identified three genes, Ddx3x, Eif2s3x, and Kdm5c, that escape X-inactivation in all lineages and are expressed more highly in females vs. males. I also identified four features that correlate with enrichment of X-inactivation escapees in some stem cell lines or embryonic tissues: 1) the presence of a Y-linked homolog; 2) proximity to the Xist locus; 3) proximity to the centromere; and 4) mapping to segments of the X chromosome whose homologous segments on the Y chromosome diverged evolutionarily more recently from the X chromosome. I also sought to improve the derivation efficiency of mouse epiblast stem cells (EpiSCs), which stably maintain random X-inactivation in culture. I developed a protocol to derive EpiSCs from embryonic day (E) 3.5 preimplantation mouse embryos that increased the derivation frequency by ~2-fold compared to previous methods. I utilized these EpiSCs to investigate the contribution of the X-inactivation escapee, Kdm5c, in the maintenance of random X-chromosome inactivation. KDM5C demethylates histone H3 lysine 4 di- and tri-methylation (H3K4me2/3) chromatin marks, which are associated with active transcription. I found that Kdm5c is necessary for maintaining gene silencing of a subset of X-linked genes in a dose-dependent manner in EpiSCs. I also found that Kdm5c is necessary to maintain silencing of a subset of inactivated X-linked genes in mouse cortical neurons. These data are the first indication that KDM5C is required to maintain X-inactivation. Future work will dissect the mechanism by which KDM5C loss results in the re-expression of silenced X-linked genes. Finally, I examined the role of the Polycomb repressive complex 2 (PRC2) component EED in imprinted X-inactivation. PRC2 catalyzes the chromatin mark histone H3 lysine 27 tri-methylation (H3K27me3), which is associated with transcriptional repression. I contributed to a published study that identified a role for the oocyte-derived maternally inherited EED protein in preventing silencing of the maternal X chromosome in imprinted X-inactivation.