Background
Embryos derived from in vitro fertilization have a high level of aneuploidy [
1,
2], which is believed to be a major factor affecting success of human assisted reproduction treatment. Therefore, preimplantation genetic screening (PGS), i.e., screening for chromosomal anomalies in preimplantation embryos, is advised before embryo transfer in treatment. Previously, PGS was performed using fluorescence in situ hybridization (FISH) on 5–12 chromosomes. Subsequent randomized controlled trials failed to show a benefit of PGS by FISH on the outcome of assisted reproduction, due to embryo mosaicism and technical limitation of FISH in analysing only a limited number of chromosomes [
3,
4]. In recent years, array-based comparative chromosomal hybridization (aCGH) was developed to solve the latter problem. With this method, all 24 chromosomes in a single cell can be analysed. Two recent randomized trials show an improvement in success rate after PGS by aCGH [
5] and quantitative polymerase chain reaction (PCR) based aneuploidy screening [
6].
Although mosaicism and aneuploidy in cleavage stage embryos based on 1–2 biopsied blastomeres have been well reported, the true aneuploidy rate and the extent of mosaicism are not as clear as it requires analysis of all 24 chromosomes in every blastomeres in an embryo. There are only a few studies reporting such information on a limited number of cleavage stage embryos. Two studies used abnormal embryos derived from preimplantation genetic diagnosis (PGD) or PGS programs [
7,
8]. Several reports used cryopreserved embryos with quality suitable for transfer [
9‐
11]. However, due to local regulatory requirements, the investigators of these reports allowed the embryos to succumb at room temperature for 24 h before determination of the chromosomal content [
10,
11]. The effect of the treatment on chromosome separation is not known. Notably, a significant proportion of the blastomeres in these studies failed to produce a conclusive result. Malmgren et al. [
12] studied 22 embryos unsuitable for transfer after PGD for structural aberration and 6 embryos donated from IVF patients with no known structural aberration. These embryos had a total cell number from 1–10 on day 4 of culture when their chromosomal content was determined by CGH. Wells and Delhanty [
13] studied 12 embryos on day 3 of culture with some having cleavage arrest. Johnson and co-workers [
14] studied 26 day 3 cryopreserved embryos from a group of advanced aged women with a mean age of 38.8. There is no data on mosaicism of high quality embryos from young women without chromosomal abnormality.
Both meiotic errors and post-zygotic mitotic errors contribute to aneuploidy in preimplantation embryos. The reported mechanisms causing these errors include non-disjunction, anaphase lagging and selective endoreduplication. Non-disjunction produces two daughter cells with one gain and one loss of the same chromosome. It results from failure to properly separate the sister chromatids during mitosis. Anaphase lagging is due to failure in attachment of a chromatid to the spindle apparatus and subsequent exclusion from the reforming nucleus. Selective endoreduplication refers to the replication of a chromosome without cell division, resulting in one normal cell and one with trisomy of the reduplicated chromosome. It has been suggested that anaphase lag leading to chromosome loss is the most common mechanism causing mosaicism [
15].
Most data on aneuploidy in preimplantation embryos were derived from studies using FISH for determination of the chromosomal content. These studies had three shortcomings. First, most of the studied embryos were diagnosed to be abnormal after PGS. Second, only a limited number of chromosomes were studied. Third, not all blastomeres of the embryos were investigated.
In this report, we aim to determine the extent of mosaicism in good quality frozen embryos from young IVF patients with no known indication for PGD. The chromosome content of each individual blastomere was analysed by aCGH, and microsatellite marker analysis were performed on the aneuploid chromosomes.
Discussion
This is the first report on the extent of mosaicism in high-quality embryos donated from young women, and more than half of the embryos were from successful assisted reproduction treatment cycles. Previous studies on embryo mosaicism typically used genetically abnormal embryos acquired from the PGD/PGS program [
7,
8,
12] or cryopreserved embryos with lysis of some blastomeres after thawing [
9]. The average number of cells per day 3 embryo analysed in these studies ranged from 5.2 to 6.9 [
8‐
10,
13]. There is only one study on two chromosomally abnormal embryos from PGD with an average of 8 cells per embryo [
7]. In our study, the embryos were donated from patients with a mean age of 30. All of the studied embryos had no blastomere lysis after thawing and developed past the 6-cell stage with good morphology after 24 h of culture. The average number of blastomeres per embryo on day 3 was 7.8. The strength of this study is that we had conclusive result on over 90% of the studied blastomeres. Thus, our data reflect a more complete picture of embryo mosaicism in high quality embryos.
A recent study reported mosaicism in 14 embryos from 9 young couples (mean maternal age: 31.3) with live birth from the same assisted reproduction treatment cycle [
10]. In the study, conclusive results could not be obtained in 33.3% (35/105) of the blastomeres due to loss of some blastomeres during thawing or disaggregation and failure of analysis, leading to no result. The low rate of obtaining conclusive results may be partly because the embryos were allowed to succumb overnight, leading to degradation of DNA and subsequent difficulties in analyses. Similarly, only 52% of blastomeres can be analysed in a cohort of embryos diagnosed to be abnormal in the PGD program [
8]. In the present study, inconclusive results were found in only 7 blastomeres, and the chromosomal content in 92.8% of the blastomeres was successfully determined.
Among the 12 studied embryos, 16.7% were diploid and 58.3% were diploid-aneuploid mosaic. These percentages are similar to previous studies on surplus embryos. Both Wells and co-workers [
13] and Voullaire and co-workers [
9] reported a diploid rate of 25% and a diploid-aneuploid mosaic rate of 41.7%. Similar rates were found in good quality embryos from advanced aged women [
14] (diploid: 23%, diploid-aneuploid: 46.2%) and young women [
10] (diploid: 28.6%, diploid-aneuploid: 57.1%).
If assuming that only diploid embryos could implant, the percentage of diploid embryos in our study is 16.7%, which is much lower than the implantation rate in the frozen-thawed embryo cycle for young women, which is 30.9% in our program. The observation suggests that some of the embryos with a minor proportion of abnormal blastomeres may implant. For instance, only 1 out of 8 blastomeres was abnormal in Embryo 1. It proposed that the implantation potential of mosaic embryos depended on the number of chromosomally abnormal blastomeres in the embryos [
18]. Frozen-thawed embryo transfer data show that 8-celled embryos that have three blastomeres (37.5%, 3/8) lysed after thawing are able to develop normally to term [
17]. Therefore, it is reasonable to assume that the embryos with less than 38% chromosomally abnormal blastomeres could implant. Among the 7 diploid-aneuploid mosaic embryos, 3 had <38% of abnormal blastomeres. Thus the percentage of studied embryos with implantation potential, i.e., diploid and diploid-aneuploidy mosaic with <38% abnormal blastomeres, is 41.7% (5/12), which is consistent with the implantation rate of frozen-thawed embryos.
The fate of chromosomally abnormal blastomeres is not fully understood. Studies show that the proportion of these blastomeres drops as the embryos develop to the blastocyst stage [
19,
20]. Several possibilities may explain the phenomenon. First, a high rate of mosaicism in early cleavage embryos may be due to degradation of the maternal transcripts leading to inadequacy of in cell cycle control and incomplete activation of embryonic genome [
21]. The development of cell cycling genes after embryonic genome activation at the 8-celled stage reduces the proportion of abnormal cells formed in later developmental stages. Second, there could be preferential growth of the euploid cells, loss of the aneuploid cells due to apoptosis or reduced division of the abnormal blastomeres [
22]. Third, it is possible that some of the abnormal blastomeres undergo “self-correction”. Several mechanisms of self-correction have been proposed including anaphase-lagging or non-disjunction [
23‐
25]. Mertzanidou and co-workers [
11] suggested that self-correction mechanisms start after day 4 of preimplantation development. It was once suggested that preferential allocation of abnormal cells to the trophectoderm could be one of the mechanisms, but this was contradicted by later observations [
26].
In this study, 48.9% of the blastomeres were euploid with no segmental aberration. The percentage of normal blastomeres is comparable to that in previous reports on day 3 embryos [
9,
13] that range from 41% to 56%. We found similar percentages of blastomeres with single monosomy (11.4%) and single trisomy (10.2%). These values vary among other studies. While Voullaire and co-workers [
9] reported 27% single monosomy and 3% single trisomy, the corresponding values by Wells and Delhanty [
13] are 8% and 17%, respectively. Factors affecting the proportion of monosomy and trisomy are not fully known. A recent retrospective analysis of over 15,000 trophectoderm biopsies showed equal prevalence of trisomies and monosomies [
27]. The percentage of blastomeres with more than one aneuploidy in the present study is 15.9%, which is similar to other reports [
9,
13].
As over 90% of the blastomeres had a definitive analysis, we reconstructed the cell lineages of each studied embryo and deduced the genotype of their zygotes based on chromosomal content and microsatellite marker analysis of blastomeres on day 3. Among the 12 studied embryos, only 3 chromosomal errors were found at the zygote stage. These chromosome errors could be due to meiotic errors of paternal and/or maternal origins. Although spermatozoa from men with severe oligoasthenoteratozoospermia have increased aneuploidy rates [
28], the estimated aneuploidy rate is less than 5% [
29,
30], which is much lower than the reported aneuploidy rate of the oocytes (22–57.1%) [
31]. Thus the observed errors are likely due to meiotic errors that occurred during oogenesis. Maternal meiotic error is well known to be positively correlated with advanced maternal age [
32]. Our studied embryos were donated from young patients. Therefore, a low incidence of meiotic error was expected. Aneuploidy rates of 3–17.9% based on CGH of polar bodies have been reported for young women [
31,
33].
In contrast to meiotic error, the rate of mitotic errors does not increase with maternal age [
32,
34]. We found 16 mitotic errors in this study resulting in whole chromosome gain or loss. There were 5 endoreduplication events, accounting for 31.3% of the mitotic errors observed (Table
2). Trophoblast cells derived from the trophectoderm of blastocysts undergo physiological endoreduplication to become the polyploid syncytiotrophoblast [
35]. Although endoreduplication usually involves the whole chromosome set, selective endoreduplication of isolated chromosomes has been reported in a human tripronucleated zygote [
36] and in cleavage stage embryos [
11].
Most of the previous studies on the mechanism of aneuploidy in preimplantation embryos were performed by FISH based on a limited number of chromosomes [
15]. Ioannou and co-workers [
37] studied all 24 chromosomes in blastocysts by 4 rounds of FISH, and concluded that anaphase lagging was the most common mechanism causing post-zygotic abnormalities. However, as only one probe per chromosome and blastocysts diagnosed to be abnormal after PGS were used in the study, the mechanisms of aneuploidy in unselected good quality embryos are not known. We found 4 non-disjunction and 4 anaphase lagging events out of 85 divisions in the studied embryos. In a similar reconstruction analysis on 13 day 4 embryos, 5 non-disjunction and 7 anaphase lagging events were postulated on day 3, but the incidence of non-disjunction increased dramatically as the embryos developed to day 4 [
11]. Whether the mechanisms of aneuploidy change with the development of the embryos awaits further investigation. It is noteworthy that there are other possible mechanisms of mitotic error such as premature cell division, chromosome demolition, cell fusion and errors in cytokinesis.
Chromosomal structural aberrations are common in preimplantation embryos, though their true frequency and biological significance are not fully known [
8]. Evidence suggests that these aberrations are independent of maternal age [
32]. It has been suggested that the prevalence of segmental aberration is higher in frozen-thawed embryos than in fresh embryos [
11]. Review of the literature shows a segmental aberration rate per blastomere from 6.3% [
13] to 8% [
11] in fresh embryos and 7.1% [
10] to 12% [
11,
14] in frozen-thawed embryos. In the present study, segmental change of chromosome was noted in 17% of the blastomeres after thawing. If there is a difference in the rate of segmental aberration, the difference is likely to be small and is of doubtful significance. It should be noted that the frequency of structural aberrations depends on the resolution of the microarray. Thus, Vanneste and co-workers [
8] reported a much higher frequency (70%) with the use of a SNP array having a resolution hundreds-fold higher than that used in the present study.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made.
The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
JFCC carried out the collection of blastomeres, aCGH studies and microsatellite marker analysis and drafted the manuscripts. WSBY participated in the design of the study and performed the embryo biopsy and statistical analysis. EYLL, VCYL, PCH and EHYN conceived of the study, participated in its design and coordination and helped to draft the manuscript. All authors read and approved the final manuscript.