Investigating spindle assembly in mammalian zygotes using light sheet microscopy

Abstract

Mammalian life starts with the formation of a zygote through fertilization of an oocyte by the sperm; subsequently the first mitotic divisions take place. Errors in these earliest divisions can easily propagate throughout preimplantation development and can lead to pregnancy loss prior to its clinical recognition. Albeit one would expect that these divisions should be particularly safeguarded due to their fundamental importance for reproduction, the first embryonic mitoses are much more error-prone than mitosis in somatic cells. Interestingly, this is a paradox observed in all mammalian species analyzed so far, including human. The essential structure mediating chromosome segregation is the microtubule-based mitotic spindle, which has been extensively studied in cell-free systems, somatic cells and embryonic model systems, such as Drosophila melanogaster and Caenorhabditis elegans. By contrast, in mammals, we understand surprisingly little about the mechanisms and regulation of spindle assembly during preimplantation development. In this thesis I therefore studied the assembly of the first mitotic spindle in two mammalian model systems, mouse and cow, to obtain insights into the mechanisms that may underlie early embryonic division errors. In both systems, I found that the first mitotic division is mediated by a pair of spindles that handle the parental genomes separately. The data indicate that this dual spindle assembly is predominantly driven by chromosomal microtubule nucleation and subsequent self-organization of the microtubules around the two spatially separated genomes. First, I used the mouse embryo—a powerful and well-established research model—in which the dual spindle assembles in the presence of many acentriolar cytoplasmic microtubule organizing centers (MTOCs) that were assumed to functionally replace centrosomes. I developed several imaging-based assays to investigate spindle assembly mechanisms in live and fixed mouse embryos and could show that, surprisingly, initial spindle microtubules in the zygote are formed by chromosomal and kinetochore-based microtubule nucleation. By contrast, only some of the many cytoplasmic MTOCs participate in spindle assembly and seem to be more important for stabilization of the spindle than for its initial assembly. Second, I studied zygotic spindle assembly in the bovine embryo, which is in general physiologically more similar to humans, and in particular, the cow zygote also inherits a centrosome from the sperm. However, it is much less established for cell biological research. Using custom-developed light sheet microscopy, I could image live bovine embryos with high spatial and unprecedented temporal resolution and thereby established this system as a valuable model to study cellular mechanisms of mammalian preimplantation development. I could show that dual spindle assembly is conserved in bovine zygotes, despite the presence of two centrosomes. Interestingly, I discovered that the centrosomes are not very active in spindle assembly, and it is thus very likely that the spindle assembly mechanism of chromosomal nucleation and microtubule self-organization, which I observed in the mouse model, is conserved. Overall, my work shows that dual spindle assembly can be found in both acentrosomal and centrosomal zygotes. It is thus likely that this mechanism is conserved across mammalian species, where the parental genomes are present in separate pronuclei in early mitosis. This is also the case in human embryos, and thus my work has significant biomedical implications. In addition, the data presented in this thesis further strengthen the hypothesis that likely also in human zygotes the parental genomes do not mix during the first mitosis, which would have substantial legal and ethical implications for the definition of the beginning of human life.