Diss. (sammanfattning) Stockholm : Karolinska institutet, 2006. / Härtill 6 uppsatser.

The mammalian zygote and resulting embryo is the starting point of life, and thus must overcome continuous insult from DNA stress and damage while maintaining genome stability and integrity. This thesis examines genome stability in the context of chromosome changes, both in the context of ploidy and whole genome duplications as well as double-strand DNA breakage and chromosome loss. Regarding the ploidy portion of this work, while possible to derive and maintain, mammalian haploid stem cells undergo spontaneous, irreversible diploidization. Here, we investigated the mechanisms driving diploidization using human and mouse embryos, and human embryonic stem cells experimental systems. We demonstrate that diploidization occurs early in development and is often unproductive, with diploidized cells failing to contribute to the developing embryo. Diploidization involves delayed mitotic progression, incomplete alignment of chromosomes, and occurs through mitotic slippage or failed cytokinesis after exit from mitosis without formation of a midbody. Diploidization is associated with DNA damage and aneuploidies, with an upstream component being a decreased nuclear to cytoplasmic ratio. Increasing this ratio in haploid mouse embryos improves developmental outcomes and decreasing this ratio in diploids results in poor outcomes. A sensor of the nuclear to cytoplasmic ratio, CHK1, is required for haploid maintenance as inhibition increases binucleation and diploidization in haploid human embryonic stem cells. Thus, we demonstrate the earliest upstream driver of diploidization as being the nuclear-cytoplasmic ratio in haploid mammalian cells, rather than the actual haploid state. Regarding the double-strand DNA breakage portion of this work, the preferred mechanism by which human embryos repair double-strand breaks was investigated. Utilizing allele-specific CRISPR-Cas9 cleavage, we show that human embryos repair double-strand breaks primarily through non-homologous end joining. In embryos left unrepaired or misrepaired, partial or whole chromosome loss occurs, which can be easily overlooked and misinterpreted with common on-target analyses such as PCR. Off-target Cas-9 activity recapitulated findings on an entirely separate chromosome, confirming the preference of the human embryo for non-homologous end joining and microhomology-mediated end joining, as well as chromosome loss where repair was unsuccessful.