On fragmentation: origin and consequences of abnormal cell division in human embryos in vitro

Cytoplasmic fragmentation is the most common cause of embryonic loss after in vitro fertilization and culture of human eggs. The occurrence of fragmentation often coincides with the presence of other abnormalities, including low cell number, uneven blastomere size, disrupted cell-cell contact,blastomere multi-nucleation, and chaotic chromosome mosaicism. As commonly as fragmentationis observed during clinical practice of in vitro fertilization and embryo transfer, the phenomenon itself rarely has been studied. The occasional investigative attempts have narrowly focused on linking fragmentation to apoptosis. Undoubtedly, one reason for this limited inquiry is the practical difficulties and obstacles associated with the use of human eggs and embryos for experimentation. The lack of a suitable animal model and the relative unimportance of fragmentation in experimental mouse embryology have also contributed to its dismissal as a worthy research subject. The present studies aim to reach beyond these imposed and inherent limitations and uncover the causes and mechanisms of cytoplasmic fragmentation, and delineate its developmental consequences. Chapter 3 describes a fragmentation classification system which was developed through non-invasive morphological evaluation of large numbers of embryos and was based on fragmentation patterns: the size and distribution of fragments relative to the size and position of nucleated cells. These patterns along with (but independent of) the degree of fragmentation were shown to be of predictive value for clinical outcome of in vitro fertilization and embryo transfer and therefore useful for selection of embryos for transfer. The loss of more than one third of the cytoplasmic volume to fragmentation or formation of large fragments was found to lead to a significant decrease in implantation and pregnancy rates following intrauterine transfer on day-3 of development. In Chapter 4, embryos with fragmentation and other abnormalities were followed through day-S of development in culture; it was discovered that the processes of compaction, cavitation, and blastulation were often abnormal in such embryos. In Chapters 5 and 6, attempts were made to discover whether (and how) fragments interfered with development. It was shown in a mouse model that continued development of blastomeres was generally unaffected by the presence of fragments. In the human, isolation of intact blastomeres from several fragmented embryos and their aggregation within a host zona pellucida led regularly to formation of chimaeric blastocysts. This work demonstrated that viable cells with apparently normal regulatory capacity can be found within non-viable embryos. In Chapter 7, the distribution of a vital cell adhesion protein, E-cadherin, was investigated in abnormal embryos by immunocytochemistry and confocal fluorescence microscopy. The results suggested that the characteristic distribution pattern of Ecadherin is perturbed and erratic in such embryos, providing one explanation for their failure to compact, cavitate or blastulate normally. Finally, the experiments described in Chapter 8 revealed the key to the nature of fragmentation: its resemblance to cytokinesis in its requirement for activation, its timing in the cell cycle, and its mediation by reorganization of the cytoskeleton. This was investigated in a mouse model, using enucleation as a way to make eggs and embryos "fragmentation- prone': In meiotic cells, the timing of fragmentation coincided with second polar body extrusion, and in mitotic cells, it coincided with mitosis and cell division. Therefore, far from being random, fragmentation occurred only during theM phase of the cell cycle. By taking into consideration the observation that the non-activated mature eggs neither divide nor fragment and that such eggs are arrested in metaphase, it was also possible to point to cytokinesis as the phase during which fragmentation occurs. The close association of fragmentation with the M phase of the cell division cycle, and in particular with the failing coordination of microtubule function would account for the numerous chromosomal and nuclear abnormalities that accompany fragmentation. Furthermore, by firmly establishing a link between fragmentation and cell division, both nuclear and cytoplasmic, the present work has definitively opened an important area for future research, most critically, into strategies to minim.ize or prevent fragmentation and its associated abnormalities.