P–207 Heterogoneic cell division proven to occur in bovine zygotes

Abstract Study question We hypothesize that the zygote can segregate parental genomes via a non-canonical pathway. We coined this heterogoneic cell division. Can we proof the existence of this new segregational pathway? Summary answer We confirmed the existence of this non-canonical segregation mechanism leading to mixoploidy and provide a catalogue of abnormal zygotic divisions. What is known already Embryos show a high degree of chromosomal instability leading to chromosomal mosaicism. Chromosomal aberrations affect the developmental potential. We developed haplarithmisis which determines the single-cell genome-wide haplotype and copy number and allows to deduce the parental and segregational origin. Analysis of cleavage-stage bovine embryos by haplarithmisis, discovered the presence of uniparental and biparental blastomere lineages in individual embryos. We hypothesized that whole genome segregations can occur via a non-canonical zygotic division, a process termed “heterogoneic” division. Abnormal zygotic division has been observed in bovine and human in vitro produced zygotes using time-lapse, but parental genome segregation has never been disclosed. Study design, size, duration We hypothesized that abnormal cytokinetics and spindle mechanics may underlie the segregation of parental genomes in a separate blastomere line. In vitro produced zygotes were monitored by time-lapse microscopy. Zygotes cleaving 3 or 4 cells were disaggregated, picked and analysed by haplarithmisis. Participants/materials, setting, methods Blastomeres from bovine in vitro produced zygotes cleaving directly into 3 or 4 blastomeres, identified by time-lapse monitoring, were tubed following zona removal and blastomere dissociation and whole-genome amplified. Samples were subsequently hybridized on Illumina Bovine HD BeadChip SNP arrays. Data was analyzed by haplarithmisis, using a the siCHILD-bovine algorithm, to infer the haplotypes and the copy number of the parental genomes. Blastomeres showing failed haplarithmisis plots were low-coverage whole-genome sequenced on a HiSeq4000 sequencer. Main results and the role of chance: We obtained 25 bovine embryos, comprising 82 blastomeres, derived from 12 families (12 cows and 2 bulls) that cleaved directly into 3 or 4 blastomeres. Sixteen, 7 and 2 out of 25 zygotes cleaved respectively in 3 cells, 4 cells and 3 cells and a fragment. In at least 20 embryos, more than one paternal haplotype was identified, showing that a polyspermic fertilization resulted in an abnormal division. All embryos contained a whole-genome abnormality in at least one blastomere, resulting in mixoploid (5), mixed diploid biparental and androgenetic (12), polyploid (4), mixed gynogenetic and androgenetic (2) and androgenetic (2) embryos. Twenty-one embryos had at least one blastomere containing a uniparental signature. Based on the blastomere haplotype profiles we classified the embryos in six segregation categories. In twelve blastomeres haplarithmisis failed. Massive parallel sequencing of the amplified DNA showed the presence of mitochondrial DNA, indicating the blastomere did not contain any genomic DNA. This observation confirms that heterogoneic cell division does occur via different non-canonical zygotic segregations, which result in a variety of chimeric and mixoploid embryos constitutions. Limitations, reasons for caution These findings apply to a small set of bovine in vitro produced abnormally cleaving embryos. Segregation patterns may be incomplete and their true in vitro and in vivo prevalence remains unknown. Based on the haplotypes the non-canonical divisions have been reconstructed. Those patterns can now be evaluated and tested. Wider implications of the findings: This study shows that heterogoneic cell division occurs. We hypothesize this non-canonical division occurs frequently in both in vitro and in vivo, not only in cattle but also in man and hypothesize that persistence of such cell lines might explain the development of androgenetic tumorous outgrowths and mosaic uniparental individuals. Trial registration number Not applicable