In this report, we describe the formation of ICMs of mosaicism/aneuploid mosaicism in vitro cultured stem cells and their elimination of chimerism by apoptosis or reducing the division of abnormal cells, enabling the self-repair of embryonic chromosomes. We observed that 24 mosaicism/aneuploid mosaicism blastocysts were separated from the ICM, and six ES cells were obtained. Finally, the chimerism of five blastocysts was repaired, and one blastocyst was partially repaired. During this process, we collected exfoliated cells for next-generation sequencing and found that the exfoliated cells of three stem-cell ICMs contained chimeric cells.
This study explored whether chimeric embryos can self-repair their chromosomes during development. Unlike in previous studies, the embryos were only in vitro cultured up to 12-days post-fertilization in this investigation[12]. We established ES cell lines from chromosomally abnormal embryos to explore the genetic characteristics of human embryonic development after the blastocyst stage. To the best of our knowledge, this is the first study to use chromosomal abnormalities and chimeric human embryos using established ES cell lines.
The development of new diagnostic techniques for PGT-A, i.e., rapid low-pass whole-genome next-generation sequencing methods that allow the detection of embryonic aneuploidy in single cells, has led to an increase in reports of embryonic mosaicism[13]. However, how chimeric embryos should be handled is still being debated. Hundreds of chromosomally healthy births have been reported following the transfer of embryos given false-positive chromosomally abnormal diagnoses by PGT-A (“mosaic” and “aneuploid”), confirming that a considerable number of embryos with normal pregnancy potential are unnecessarily discarded, which was recently also pointed out by Paulson[14]. Although the information on neonatal outcomes is limited, more than 100 live births without abnormal phenotypes following chimeric embryo transfer have been documented[15]. Numerous studies have shown that transferring mosaic embryos produces healthy babies because the chimera repairs itself[16]. The mechanism by which this occurs is not well understood, but studies suggest three hypotheses: the embryonic mortality model[17], the clonal depletion model[18], and the preferential assignment of aneuploid cells to the TE model[19].
The incidence of chimerism is known to decrease during development and occurs less frequently in blastocysts than in cleavage stage embryos[20]. This can be explained by the fact that the proliferation of euploid cells competes with that of aneuploid cells, resulting in the death of the aneuploid cells.
Preimplantation embryos from chimeric mouse models were found to have a progressive decrease in aneuploid cells. Chimeras of euploid cells and aneuploid cells showed selective apoptosis of aneuploid cells in the ICM and proliferative defects in aneuploid cells of the TE, leading to a progressive depletion of aneuploid cells from the blastocyst stage onward[21]. This may correspond to the proposed self-correcting mechanism, as aneuploid cells may proliferate slower or undergo apoptosis, whereas euploid cells may compensate by increasing their proliferation rate. Research by Bolton et al. provided direct evidence that apoptosis within the ICM can occur as a mechanism to eliminate chromosomally abnormal cells[22]. The ability of aneuploid embryos to self-correct downstream of the blastocyst stage was first reported in mice and more recently demonstrated in human embryonic cell lineages and the human blastocyst stage. In the current study of extended embryo culture, we validated the self-correcting mechanisms of human embryos and their ability to eliminate/clear abnormal cellular debris[23].
In 2019, Popovich et al. applied next-generation sequencing to study the progression of blastocysts diagnosed as aneuploid, euploid, or mosaic using an expanded in vitro culture model. Although the study did not provide direct evidence of self-correction, a large proportion of mosaic embryos showed euploid growth by day 12 of development, providing evidence for the depletion of abnormal cells throughout the early post-implantation period[11]. In our study, five blastocyst-derived stem cells underwent chromosomal chimeric repair, one blastocyst-derived stem cell underwent partial chromosomal chimeric repair, three of which shed chimeric cells. However, some blastocysts had chromosomal abnormalities that were not found in the debris and stem cells.
The chromosomal abnormality detection methods are significant improvements over previous ones and that NGS technology in PGT-A has significantly improved the identification of diploid/aneuploid mosaicism in multicellular biopsy samples. The accuracy of phenomenon detection is challenged by several biological biases, including sampling and the possibility of mutual errors[24]. Although up to 86.2% of the studied embryos showed consistent PGS conclusions between TE and ICM[25]. In a small study involving array CGH screening of 10 embryos, the concordance between TE and ICM was reported to be as high as 100%[26]. Using the same approach, Johnson et al. showed the agreement between TE and ICM was 96%[27].
However, chromosomal abnormality detection methods continue to result in a clinically significant prevalence of blastocyst mosaicism, given that PGT-A considers the chromosomal composition of an average of 5-6 cells on biopsy to represent the entire embryo, and this brings controversy. Gleicher et al. used a mathematical model to propose that chimeric cells are evenly distributed in the trophectoderm, and at least 27 trophectoderm cells are required in a biopsy to represent the karyotype of the entire blastocyst[28]. Studies have shown there was a 13.8% incidence of inconsistency in PGS conclusions between TE and ICM due to sampling error in embryos with restricted mosaicism, and a 3.4% probability of reciprocal error[29]. In a recent experimental study, Victor et al. reported whole-chromosomal aneuploidy in TE biopsies of 100 human blastocysts with a 96.8% concordance between TE and ICM. However, in the presence of chimeras, the concordance of TE with ICM decreases significantly[30]. In addition, although NGS technology has a high sensitivity and detection rate for chimeric embryos, the length of NGS read data is shorter than that of traditional sequencing data, and the error rate in the sequence assembly process is 0.1% to 15%. The analysis of large amounts of data poses several challenges[31]. This could explain why the final results for three of the six stem cell lines we established did not match the biopsy results perfectly. Again, the sample size of our study was small, and larger studies are needed to confirm or refute our findings.