The human embryonic genome is karyotypically complex, with chromosomally abnormal cells preferentially located away from the developing fetus.

study questionAre chromosome abnormalities detected at Day 3 post-fertilization predominantly retained in structures of the blastocyst other than the inner cell mass (ICM), where chromosomally normal cells are preferentially retained? SUMMARY ANSWER: In human embryos, aneuploid cells are sequestered away from the ICM, partly to the trophectoderm (TE) but more significantly to the blastocoel fluid within the blastocoel cavity (Bc) and to peripheral cells (PCs) surrounding the blastocyst during Day 3 to Day 5 progression. WHAT IS KNOWN ALREADY: A commonly held dogma in all diploid eukaryotes is that two gametes, each with 'n' chromosomes (23 in humans), fuse to form a '2n' zygote (46 in humans); a state that remains in perpetuity for all somatic cell divisions. Human embryos, however, display high levels of chromosomal aneuploidy in early stages that reportedly declines from Day 3 (cleavage stage) to Day 5 (blastocyst) post-fertilization. While this observation may be partly because of aneuploid embryonic arrest before blastulation, it could also be due to embryo 'normalization' to a euploid state during blastulation. If and how this normalization occurs requires further investigation. STUDY DESIGN, SIZE, DURATION: A total of 964 cleavage-stage (Day 3) embryos underwent single-cell biopsy and diagnosis for chromosome constitution. All were maintained in culture, assessing blastulation rate, both for those assessed euploid and aneuploid. Pregnancy rate was assessed for those determined euploid, blastulated and subsequently transferred. For those determined aneuploid and blastulated (174 embryos), ICM (all 174 embryos), TE (all 174), Bc (47 embryos) and PC (38 embryos) were analyzed for chromosome constitution. Specifically, concordance with the original Day 3 diagnosis and determination if any 'normalized' to euploid karyotypes within all four structures was assessed. PARTICIPANTS/MATERIALS, SETTING,

methodsAll patients (144 couples) were undergoing routine preimplantation genetic testing for aneuploidy in three IVF clinical settings. Cleavage-stage biopsy preceded chromosome analysis by next-generation sequencing. All patients provided informed consent. Additional molecular testing was carried out on blastocyst embryos and was analyzed for up to four embryonic structures (ICM, TE, Bc and PC). MAIN RESULTS AND THE ROLE OF CHANCE: Of 463/964 embryos (48%) diagnosed as euploid at Day 3, 70% blastulated (leading to a 59% pregnancy rate) and 30% degenerated. Conversely, of the 501 (52%) diagnosed as aneuploid, 65% degenerated and 35% (174) blastulated, a highly significant difference (P < 0.0001). Of the 174 that blastulated, the ratio of '(semi)concordant-aneuploid' versus 'normalized-euploid' versus 'other-aneuploid' embryos was, respectively, 39%/57%/3% in the ICM; 49%/48%/3% in the TE; 78%/21%/0% in the PC; and 83%/10%/5% in the Bc. The TE karyotype therefore has a positive predictive value of 86.7% in determining that of the ICM, albeit with marginally higher aneuploid rates of abnormalities (P = .071). Levels of abnormality in Bc/PC were significantly higher (P < 0.0001) versus the ploidy of the ICM and TE and nearly all chromosome abnormalities were (at least partially) concordant with Day 3 diagnoses. LIMITATIONS, REASONS FOR CAUTION: The results only pertain to human IVF embryos so extrapolation to the in vivo situation and to other species is not certain. We acknowledge (rather than lineage-specific survival, as we suggest here) the possibility of other mechanisms, such as lineage-specific movement of cells, during blastulation. Ethical considerations, however, make investigating this mechanism difficult on human embryos. WIDER IMPLICATIONS OF THE

findingsMosaic human cleavage-stage embryos can differentiate into a euploid ICM where euploid cell populations predominate. Sequestering of aneuploid cells/nuclei to structures no longer involved in fetal development has important implications for preimplantation and prenatal genetic testing. These results also challenge previous fundamental understandings of mitotic fidelity in early human development and indicate a complex and fluid nature of the human embryonic genome. STUDY FUNDING/COMPETING INTEREST(S): This research was funded by Organon Pharmaceuticals and Merck Serono by grants to W.G.K. W.G.K. is also an employee of AdvaGenix, who could, potentially, indirectly benefit financially from publication of this manuscript. R.C.M. is supported by the National Institute of General Medical Sciences of the National Institutes of Health under award number R35GM133747. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. D.K.G. provides paid consultancy services for Care Fertility. TRIAL REGISTRATION NUMBER: : N/A.