Chromosomal mosaicism throughout human preimplantation development in vitro: incidence, type, and relevance to embryo outcome

Magdalena Bielanska1,3, Seang Lin Tan1,3 and Asangla Ao1,2,4

1 Departments of Obstetrics and Gynecology, 2 Human Genetics and 3 Experimental Medicine, Royal Victoria Hospital, McGill University, Montreal, QC H3A 1A1, Canada


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
BACKGROUND: A large percentage of in-vitro generated cleavage stage human embryos are chromosomally mosaic, consisting of both normal (diploid) and abnormal (non-diploid) cells. The present study characterized mosaicism at each stage of cleavage division and examined its effect on preimplantation development in vitro. METHODS: A total of 216 normally fertilized (two-pronucleate) embryos which were not selected for transfer to the patients were analysed for chromosomal abnormalities using multi-colour fluorescence in-situ hybridization DNA probes specific for three to five of nine different chromosomes (X, Y, 2, 7, 13, 16, 18, 21, 22). RESULTS: Overall, 48.1% of embryos were mosaic. The frequency of mosaic embryos increased from 15.2 to 49.4 to 58.1%, from the 2–4-cell to 5–8-cell to morula stages respectively, and the types of non-diploid cells detected were mostly aneuploid or chaotic. The incidence of mosaicism at the blastocyst stage was 90.9%; however, most of the mosaicism comprised diploid and polyploid cells. Arrested mosaic embryos had a higher incidence of chaotic abnormalities, and higher proportions of abnormal cells compared with the non-arrested group. CONCLUSIONS: Post-zygotic errors leading to mosaicism may occur, and persist throughout preimplantation development in vitro. Our results suggest that mosaicism involving multiple chromosomal imbalances and/or imbalances affecting a high proportion of cells in an embryo appear to impair development to the blastocyst stage.

Key words: chromosomes/embryo mosaicism/fluorescence in-situ hybridization/preimplantation development


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The widespread use of IVF for treatment of infertility has stimulated research into chromosomal abnormalities in human preimplantation embryos. Initially chromosome anomalies in in-vitro generated zygotes were ascertained by karyotype analysis (Angell et al., 1986Go; Plachot et al., 1987Go; Papadopoulos et al., 1989Go). However, due to low numbers of dividing cells, poor quality preparations, as well as artefactual loss of chromosomes, data from these investigations were limited. More recent fluorescence in-situ hybridization (FISH) of chromosome specific DNA probes to interphase cells has enabled a rapid and efficient assessment of numerical chromosomal imbalances in all or most cells from both cleaving and arrested embryos. Depending on the numbers and specificity of probes used, the quality of the examined embryos, as well as the patient population, the reported incidence of abnormalities detected by FISH has varied from 15% to >85% (Harper et al., 1994Go, 1995Go; Munné et al., 1994aGo,bGo, 1995aGo,bGo; Delhanty et al., 1997Go; Bahce et al., 1999Go; Iwarsson et al., 1999Go). FISH studies have demonstrated that the incidence of aneuploidy, haploidy and polyploidy in embryos is much higher than the incidence observed in clinically established pregnancies. Furthermore, these studies revealed that between 15% and >50% of monospermic cleavage stage embryos were mosaic, comprising cells with both diploid and non-diploid chromosome complements. The abnormal cell lines detected included monosomy, trisomy, polyploidy and haploidy. Another type of abnormality consisted of imbalances for multiple chromosomes which differed from cell to cell. Embryos with such anomalies have been identified as `chaotic' (Delhanty et al., 1997Go), `complex' (Munné et al., 1994aGo) or `extensive' (Munné et al., 1995bGo) mosaics. Mosaicism has been documented in embryos derived from various ovarian stimulation and culture protocols (Munné et al., 1997Go). Mosaicism has been found in embryos from women of young and advanced reproductive ages (Munné et al., 1995bGo). It has been described in arrested (Munné et al., 1994bGo, 1995bGo; Bahce et al., 1999Go), frozen–thawed (Munné et al., 1997Go; Iwarsson et al., 1999Go), and fragmented (Bongso et al., 1991Go; Munné et al., 1994aGo,bGo, 1995aGo,bGo) embryos, as well as in embryos of good quality (Munné et al., 1994aGo; Harper et al., 1995Go; Delhanty et al., 1997Go; Bahce et al., 1999Go; Bielanska et al., 2000Go).

Despite the prevalence of mosaicism in cleavage stage embryos, little is known about the effect of this condition on embryonic development and pregnancy outcome. Mosaicism has been reported in both fetal and placental tissues of first and second trimester pregnancies, but its incidence at these stages is very low. It is present in only ~5% of aneuploid spontaneous abortions at 6–20 weeks gestation (Hassold, 1982Go) and in 1–2% of viable pregnancies screened by chorionic villus sampling (CVS) (Ledbetter et al., 1992Go; Wang et al., 1993Go). Furthermore, unlike in embryos, the mosaicism detected in established pregnancies is mostly limited to diploid trisomy. The considerable reduction in mosaicism from early cleavage stages to clinically established pregnancies indicates that most mosaicism is eliminated prior to the first trimester. Cytogenetic analysis of blastomeres from embryos on day 2 to day 5 of development in vitro suggests a progressive loss of chromosomally abnormal embryos during preimplantation development (Almeida and Bolton, 1996Go). Several studies reported that the incidence of chromosomal anomalies at early cleavage divisions is higher among arrested compared with non-arrested embryos (Munné et al., 1994aGo, 1995aGo,bGo; Bahce et al., 1999Go). It is therefore possible that the selection against mosaic embryos may begin during preimplantation development. This early wastage of embryos may be a contributing factor to the low rates of blastocyst formation and high rates of implantation failure following embryo transfer on days 2–3 post-insemination.

In order to help clarify the significance of co-existing diploid and non-diploid cells for preimplantation embryo wastage, implantation failure, and fetal and placental mosaicism, we (i) used multi-probe FISH to determine the incidence of mosaicism in 2-cell to blastocyst stage embryos, (ii) compared the composition (type and numbers of non-diploid cells) of mosaicism detected at each stage, and (iii) examined the relationship between the different patterns of mosaicism and embryonic development and viability.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Embryos
Spare embryos (those not selected for embryo transfer), were donated with written consent by 81 patients undergoing IVF or ICSI at the McGill Reproductive Center who chose not to have their supernumerary embryos frozen or whose embryos were not suitable for cryopreservation. All embryos included in the study had been normally fertilized (two-pronucleate: 2PN). The mean maternal age of patients was 36 years (range 22–45). Embryos were cultured in Medicult culture media (Medicult, Hopkington, MA, USA), or in extended culture G1 and G2 sequential media (IVF Science Scandinavia, Gothenburg, Sweden). Cleavage stage embryos were spread for chromosomal analysis on day 2 to day 5 of development in vitro. Embryos were considered to be arrested when no cleavage had occurred during the 24 h prior to embryo spreading. Alternatively, when the extent of cleavage previous to time of spreading was not known, embryos were classified as arrested when they consisted of: <5 cells on the evening of day 3 post fertilization, <8 cells on day 4, <12 cells on day 5. On the day of spreading, cleavage stage embryos were evaluated for the degree of fragmentation and blastomere size (Steer et al., 1992Go; Dean et al., 2000Go) with the following modifications: grade I: even-sized blastomeres with no fragmentation; grade II: even-sized blastomeres with <10% fragmentation; grade III: unequal-sized blastomeres and 10–30% fragmentation; grade IV uneven-sized blastomeres, degeneration, and >30% fragmentation. Embryos showing a blastocoel cavity were classified as blastocysts and were harvested on day 6 of development. Grade I blastocysts had a clear inner cell mass (ICM) and >=120 cells. Blastocysts with an ICM and 70–119 cells were scored as grade II. Grades III and IV were assigned to blastocysts with a less defined, or no apparent ICM and 51–69 cells, or <=50 cells respectively.

DNA probes
Directly labelled DNA probes were obtained from Vysis (Downers Grove, IL, USA). The probes varied according to their region of hybridization on the chromosome and fluorescent label as follows: CEP 2 (alpha satellite DNA, spectrum orange); CEP 7 (alpha satellite D7Z1, spectrum green); LSI 13 (region 13q14, spectrum red); CEP 16 (satellite II, D16Z3, spectrum aqua); CEP 18 (alpha satellite D18Z1, spectrum aqua, or spectrum blue); LSI 21 (region 21q22.13–21q22.2, spectrum green); LSI 22 (region 22q.11.2, spectrum gold); CEP X (alpha satellite DXZ1, spectrum green); CEP Y (alpha satellite DYZ3, spectrum orange). Each of the probes had been previously tested in our laboratory by hybridization to 200 normal 46,XY lymphocytes prepared by standard cytogenetic procedures (not reported here). The hybridization efficiency of each probe was determined as the number of interphase cells with the expected number of probe-specific signals, minus cells with less than or more than the expected number of signals divided by total number of cells scored. The efficiencies for the probes varied from 89 to 98%.

FISH protocol
Embryos were washed in phosphate-buffered saline (PBS) and transferred onto poly-L-lysine-coated slides. Cleavage stage and blastocyst stage embryos with intact zona pellucida were dissolved in 0.01 N HCl, 0.1% Tween 20 solution and nuclei spread and fixed on the slides (Coonen et al., 1994Go). The preparations were air-dried, washed in PBS for 5 min and dehydrated in 50, 70 and 100% ethanol. Fixed nuclei were viewed using a phase contrast microscope and their location marked with a diamond pen. Embryonic nuclei were pretreated with pepsin (0.1 mg/ml in 0.01 N HCl) for 10 min at 37°C, and again dehydrated. FISH probes for three to five different chromosomes were combined in hybridization buffer [50% formamide, 2xsaline sodium citrate (SSC); Oncor, Gaithersburg, MD, USA]. Probe mixtures used in the study included a three-colour probe mixture specific to chromosomes 2, 7, 18, a three-colour mixture specific to chromosomes X, Y, 18, and a five-colour mixture probe mixture specific for chromosomes 13, 16, 18, 21, 22. The probe mixtures were applied to the embryonic nuclei and sealed under a plastic coverslip. The target and probe DNA were co–denatured by heating the slide to 74–78°C, for 4–6 min. Hybridization was carried out in a moist chamber at 37°C, for 5–18 h. The unbound probe was washed off using 0.4xSSC/0.3% Tween 20, or 0.7xSSC/0.1% Tween 20 solution at 73°C, for 2–6 min, followed by rinsing in 2xSSC/0.1% Tween 20 at room temperature. Preparations were mounted in antifade solution (p-phenylenediamide dihydrochloride in PBS; Vector). Depending on fluorescence colour of probes used, antifade was applied alone, or with DAPI (Sigma) nuclear stain (0.25–0.5 ng/ml). Nuclei were viewed using a fluorescence Olympus DX 60 microscope equipped with appropriate filters. Images were captured using a CCD camera and Cytovision software (Applied Imaging, Santa Clara, CA, USA).

Signal and data analysis
Only intact, undamaged nuclei were scored. Nuclei with a dim signal for more than one of the chromosomes tested were considered a hybridization failure and were not scored. Fluorescence spots of diameter noticeably smaller than other signals in the same nucleus were considered non-specific probe binding and were excluded from the results. Two small focal signals of the same colour, separated by a distance of less than one signal domain, were considered to represent a split signal from one chromosome. An embryo was diagnosed with an imbalance for a specific chromosome when the abnormal signal pattern was detected in a higher proportion of cells than in normal control lymphocytes. Polyploidy (more than two signals for each chromosome pair tested), haploidy (a single signal for each chromosome pair tested), or reciprocal gains and losses in sibling blastomeres were always considered as true imbalances. Only those embryos yielding results in >=75% of cells were included in the study.

The incidence of mosaicism and the percentage of normal to abnormal cells in mosaic developing and arrested embryos were compared at sequential stages of preimplantation development (2-cell to blastocyst stages). Mosaicism was compared among cleavage stage embryos of different morphological grades. The frequency of mosaicism in cleavage stage embryos was also compared among three different maternal age groups (<=34, 35–39, >=40 years of age). Statistical analysis was carried out using {chi}2-test and Fisher's exact (F) test.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
A total of 216 embryos, 59 at the 2–4-cell, 81 at the 5–8-cell, 43 at the morula, and 33 at the blastocyst stage, were analysed. The proportion of embryos screened by probes for chromosomes X, Y, 18, for 2, 7, 18 and for 13, 16, 18, 21, 22 at the different developmental stages were as follows: 23 (38.9%), 23 (38.9%) and 13 (22.0%) at 2–4 cell stage; 30 (37.0%), 31 (38.2%) and 20 (24.7%) at the 5–8-cell stage; 16 (37.2%), 17 (39.5%) and 10 (23.2%) at the morula stage. Blastocysts were analysed by two probe mixtures X, Y, 18 and 2, 7, 18 only, 16 (48.5%) by the first, and 17 (51.5%) by the second.

FISH results of 216 embryos are presented in Table IGo. Overall 29.6% of embryos were uniformly normal (diploid; 2N) for the chromosomes tested. Twenty-two per cent of embryos comprised only abnormal cells. Nine per cent of embryos were chaotic, with blastomeres showing various imbalances for multiple chromosomes. Seven per cent were aneuploid. One of these embryos showed double uniform aneuploidy, trisomy 13 and monosomy 21. One aneuploid embryo showed trisomy 2 in all 11 cells plus chaotic imbalances for chromosomes 7 and 18 in two of the cells. Another embryo showed trisomy 18 in 7/8 cells and chaotic complement in one cell. Haploidy and polyploidy accounted for 0.5 and 5.6% of embryos respectively. The largest cohort of embryos (48.1%) was mosaic, consisting of combinations of normal and abnormal cells. These embryos were classified into four groups based on the predominant type of non-diploid cells detected (Table IGo). Sixteen per cent of embryos contained both diploid and chaotic cells. This was followed by 2N-polyploid mosaicism, consisting predominantly of tetraploid (4N) cells (14.8%) and 2N-aneuploid mosaicism consisting of diploid cells and cells showing a gain and/or a loss for one of the chromosomes tested (14.4%). Only 2.8% of the embryos contained combinations of diploid and haploid (1N) cells. The proportion of haploid cells in the one 2N-haploid blastocyst was 55%.


View this table:
[in this window]
[in a new window]
 
Table I. FISH results from hybridization of 216 preimplantation embryos
 
A total of 115/216 embryos were arrested at the 2-cell to morula stages (Table IGo). Abnormalities affecting entire embryos were more frequent among the arrested group (F-test, P = 0.05). The overall incidence of mosaicism among the arrested and developing embryos varied with the type of non-diploid cells. Chaotic mosaicism was significantly more frequent among embryos which were blocked in development prior to blastocyst formation compared with the non-arrested group (F-test, P < 0.005). The incidence of polyploid mosaicism was significantly higher among the developing embryos (F-test, P < 0.0001). The majority of the polyploid mosaic embryos were blastocysts (23/26 cases; Table IGo). When blastocysts were excluded, the frequencies of 2N-polyploidy in the developing group and the arrested group were not statistically different (4.4 versus 5.2%).

The incidence of mosaicism was compared at each developmental stage. The overall frequency of mosaicism increased significantly at sequential stages of preimplantation development from 15.2% at the 2–4-cell, to 49.4% at the 5–8-cell, to 58.1% at the morula and to 90.9% at the blastocyst stage ({chi}2-test, P < 0.0001; Table II AGo). The increase in frequency of overall mosaicism was found in both developing and arrested embryos (Table II B and CGo). The majority of the non-diploid blastomeres observed in the 2-cell to morula stage mosaic embryos were chaotic or aneuploid. Chaotic mosaicism decreased significantly from the arrested morula stage to the blastocyst stage embryos (F-test, P < 0.04). Diploid-aneuploid mosaicism also decreased by the blastocyst stage; however, because blastocysts were analysed for only three chromosomes, the rate of mosaic aneuploidy at the blastocyst stage may be slightly higher than the rate detected. The predominant form of mosaicism found at the blastocyst stage was diploid-polyploid (mostly tetraploid) mosaicism. The increase in 2N-polyploid mosaicism from both developing and arrested morulae to the blastocyst stage was statistically significant (F-tests, P < 0.0001).


View this table:
[in this window]
[in a new window]
 
Table II. Distribution of different forms of mosaicism according to stage of preimplantation development
 
The average percentages of non-diploid nuclei in the cohort of developing and arrested mosaic embryos from sequential stages of development are shown in Table IIIGo. At each stage, mosaic embryos which were arrested in development by the time of spreading showed higher proportions of abnormal cells compared with the non-arrested group. The majority of cells in blastocysts (78.2%) were diploid for chromosomes tested. In contrast to cleavage stage embryos, aneuploid cells among blastocysts were relatively few, ranging from only 6 to 8 cells per blastocyst. The average proportion of polyploid cells among mosaic blastocysts was 16.4%. However, the specific occurrence of polyploid cells per blastocyst varied from 1.4% (1/72 cells) to 47.1% (33/70 cells). The degree of polyploidy was not correlated with the total number of cells in the blastocysts. Due to the small number of other types of abnormalities, we could not comment on relationship between other imbalances and blastocyst quality.


View this table:
[in this window]
[in a new window]
 
Table III. Average percentage of diploid (2N) and non-diploid (Non-2N) blastomeres in mosaic 2-cell to blastocyst stage embryos
 
Of the 183 cleavage stage embryos in the study, 55, 89 and 39 were assigned grade II, III and IV respectively. Grade I embryos, which we considered those with symmetrically sized blastomeres and no fragmentation were few, and were not included in this study. A diploid chromosome complement was observed more frequently among grade II compared with grade III or IV embryos (47.3 versus 29.2%, versus 30.8%). The difference between grade II and III embryos was statistically significant (F-test, P = 0.02). However, the incidence of mosaicism did not change with respect to embryo grade. The specific frequencies of different forms of mosaicism among 2-cell to morula stage embryos of different morphologies are presented in Table IVGo.


View this table:
[in this window]
[in a new window]
 
Table IV. Distribution of mosaicism in 2-cell to morula stage embryos according to embryo morphology
 
We compared the chromosomal status of cleavage stage embryos obtained from three maternal age groups: <=34, 35–39 and >=40 years. Blastocyst stage embryos were not included due to high prevalence of 2N-polyploid mosaicism and uneven distribution of blastocysts among the three age categories. The distribution of arrested and non-arrested embryos among the three age categories was similar: <=34 years: 34 (59.6%) versus 23 (40.3%); 35–39 years: 58 (65.9%) versus 30 (34.1%); >=40 years: 23 (65.7%) versus 12 (34.3%). The frequency of aneuploid embryos was significantly higher among the oldest (17.1%) compared with the intermediate (3.8%) age group (F-tests, P = 0.01). Due to the small number of aneuploid embryos, the relationship between specific aneuploidies and maternal age could not be analysed. Unlike uniform aneuploidy, the incidence of mosaicism did not increase with advanced maternal age.


    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Despite the high incidence of mosaicism observed in human IVF-generated embryos, studies of chromosomal mosaicism at the preimplanation stages are few, with the majority of studies being focused on the incidence of maternal age-related aneuploidy. To our knowledge, this is the first detailed analysis of chromosomal mosaicism throughout all stages of preimplantation development.

Our FISH results revealed four major types of diploid mosaicism. One of the two most frequent types afflicting 2-cell to morula stage embryos comprised diploid cells and cells showing `chaotic' imbalances, often involving all of the chromosomes tested. These imbalances may have originated in a normal zygote or may have been superimposed on chromosome-specific meiotic or mitotic gains or losses and may be a consequence of chromosome misalignment on a disorganized spindle, in combination with a non-functional metaphase/anaphase checkpoint control (LeMarie-Adkins et al., 1997Go; Harrison et al., 2000Go). The abnormalities of the mitotic spindle could be related to suboptimal in-vitro culture environment (Pickering et al., 1990Go; A'arabi et al., 1997Go).

As observed previously (Munné et al., 1994aGo,bGo, 1995bGo; Magli et al., 2000Go), chaotic chromosome complements in this study were more frequent among arrested embryos. Detection of chaotic mosaicism up to the 12-cell stage demonstrates that blastomeres with extensive chromosome imbalances may not always compromise the first four cleavage divisions. This may be explained by the relatively late activation of the human embryonic genome at the 4–8-cell stage (Braude et al., 1988Go). However, the majority (8/10) of morulae with chaotic cells were arrested. The two which were not arrested had only a low proportion of abnormal cells. In agreement with previous findings, the incidence of blastocyst chaotic mosaicism was significantly reduced (Evsikov and Verlinsky, 1998Go; Veiga et al., 1999Go; Magli et al., 2000Go; Ruangvutilert et al., 2000Go). Together, these data suggest that most extensive imbalances are incompatible with blastocyst formation. Early wastage of chaotic embryos is consistent with the rarity of multiple chromosome anomalies, and the very low rates of double trisomies (0.21–2.8% for all chromosomes combined) among spontaneous abortions (Reddy, 1997Go). Replacing embryos reaching the blastocyst stage has yielded higher rates of implantation per embryo transferred compared with day 2 or 3 embryo transfer (Edwards and Beard, 1999Go; Coskun et al., 2000Go). Our study suggests that the greater implantation success following replacement of embryos cultured to the blastocyst stage in vitro may be associated with lower rates of the chaotic non-viable abnormalities among these embryos.

The second most frequent form of mosaicism detected in cleavage stage embryos in this study consisted of 2N cells in combination with cells showing imbalances for one of the chromosomes tested (aneuploid mosaicism). Diploid and trisomic and/or monosomic cell lines may originate from chromosome-specific mitotic non-disjunction, chromosome duplication or chromosome loss in selected cells from a diploid or an aneuploid zygote (Ford et al., 1988Go; Robinson et al., 1995Go; Los et al., 1998Go). The predominance of the diploid cell line among the aneuploid mosaic embryos, as well as the lack of a higher frequency of mosaicism in embryo donors aged >=40 years, suggests that most of these cases originated from a mitotic error in a diploid zygote.

Recently, it was demonstrated that aneuploid cell lines may persist to the blastocyst stage from day 3 post insemination (Magli et al., 2000Go). However, relative to cleavage stage embryos, the number of aneuploid cells in the five 2N-aneuploid blastocysts examined in this study were low (6–8 cells). Relatively small numbers of aneuploid cells have also been detected in other blastocysts (Evsikov and Verlinsky, 1998Go; Ruangvutilert et al., 2000Go). The minor aneuploid cell lines indicate that the aneuploid cells did not persist from early cleavage, but were formed at, or shortly before, blastulation, and suggest that most cleavage stage embryos with a high degree of mosaic aneuploidy do not complete preimplantation development.

Unlike that which has been observed in tetraploid-diploid mouse aggregation chimeras (James and West, 1994Go), recent studies of human blastocysts did not show evidence of preferential allocation of aneuploid cells into the trophectoderm cell lineage (Evsikov and Verlinsky, 1998Go; Magli et al., 2000Go). Therefore the small numbers of aneuploid cells detected in our cohort of blastocysts may have originated from the ICM or from trophectoderm, and could lead to generalized or to confined placental mosaicism (CPM) (Kalousek and Dill, 1983Go).

A diploid-haploid (2N/1N) chromosome complement comprised the least frequent class of mosaicism detected in this study. The origins of haploid cells in diploid embryos have been discussed previously (Bongso et al., 1991Go; Delhanty et al., 1997Go). In most cases in this, and other, studies, haploidy was limited to one or two blastomeres per embryo suggesting that 1N cells have a proliferative disadvantage among 2N cells (Munné et al., 1994aGo, 1994bGo; Harper et al., 1995Go; Bahce et al., 1999Go). However, our finding of one blastocyst with a large fraction (55%) of haploid cells demonstrates that haploid cells may occasionally proliferate throughout the preimplantation period. As 2N/1N mosaicism has not been documented in fetal tissues, such embryos must become eliminated at, or shortly after, implantation.

A small percentage of 2–8-cell embryos, ~2% of morulae, and ~89% of the blastocysts stage embryos in this study showed 2N-polyploid mosaicism. This form of mosaicism was also common in human morula and blastocyst stage embryos examined by others (Angell et al., 1987Go; Benkhalifa et al., 1993Go; Clouston et al., 1997Go; Evsikov and Verlinsky, 1998Go; Ruangvutilert et al., 2000Go). Polyploid cells have also been observed in a normal first trimester human placenta (Sarto et al., 1982Go). The dramatic increase in incidence of polyploid cells from day 4 to day 6 of development in this study suggests that most polyploid cells arose during blastocyst formation, and may, as in other mammalian species, be a hallmark of trophoblast differentiation (Barlow and Sherman, 1972Go; Hare et al., 1980Go; Long and Williams, 1982Go; Murray et al., 1986Go). However, the proportion of polyploid cells in five of the blastocysts was very high (33–47%). If tetraploidy in these blastocysts represents impaired cytokinesis caused by suboptimal culture conditions, the blastocysts may be arrested and unable to implant. Alternatively, the 4N cells may continue proliferation and contribute to the 2% of cases of spontanous abortions (Eiben et al., 1990Go).

A number of studies have shown that dysmorphic embryos exhibit a higher incidence of chromosomal abnormalities (Munné et al., 1994bGo, 1995bGo; Almeida and Bolton, 1996Go; Delhanty et al., 1997Go). Degree of fragmentation and relative size of blastomeres did not predict the incidence of mosaicism among embryos in this study (Table IVGo). The 38.2% mosaicism rate among grade II cleavage stage embryos in this study suggests that inclusion of embryos with even-sized blastomeres and a small degree (<10%) of fragmentation among cohort being transferred to the patient may result in replacement of embryos with aneuploid or chaotic cells.

In conclusion, this study demonstrates that post-meiotic errors leading to mosaicism may occur, and persist throughout preimplantation development in vitro. Our results suggest that the developmental potential of mosaic embryos is dependent on the type and proportion of non-diploid cells. Mosaicism involving multiple chromosomal anomalies, or a non-diploid karyotype in >50% of blastomeres per embryo, appears to reduce embryonic capacity to develop into a blastocyst. This form of mosaicism may contribute to the high rates of implantation failures following cleavage stage embryo transfer. In contrast, small numbers of aneuploid cells in early embryos may not be detrimental to successive cleavage divisions. If compatible with implantation, embryos with mild aneuploidy may be a candidate for fetal, or for confined placental mosaicism. Most mosaic polyploidy in the blastocysts appears distinct from the mosaicism observed at the early stages of cleavage division, and may be a normal part of mammalian trophoblast differentiation. The extent to which the different classes of mosaicism observed in the present study exist among embryos selected for transfer to patients and in preimplantation embryos in vivo, and their contribution to the low rate of fecundity in the human, remains to be investigated.


    Acknowledgements
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The authors are very grateful to the embryologists and clinicians at the McGill Reproductive Center for their assistance and cooperation. We would like to thank Martin Bernier and Erin Adams for their contribution to this work. This study was approved by the Royal Victoria Ethics Review Board. M.B. was in receipt of a doctoral Lloyd-Carr Harris–McGill Major fellowship.


    Notes
 
4 To whom correspondence should be addressed at: Royal Victoria Hospital, Dept Obs/Gyn, Women's Pavillion, F-3–16, 687 Pine Avenue West, Montreal, QC H3A 1A1, Canada. E-mail: asangla.ao{at}muhc.mcgill.ca Back

Submitted on November 17, 2000; resubmitted on June 11, 2001


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Almeida, P.A. and Bolton, V.N. (1996) The relationship between chromosomal abnormality in the human preimplantation embryo and development in vitro. Reprod. Fertil. Dev., 8, 235–241.[ISI][Medline]

Angell, R.R., Templeton, A.A. and Aitken, R.J. (1986) Chromosome studies in human in vitro fertilization. Hum. Genet., 72, 333–339.[ISI][Medline]

Angell, R.R., Sumner, A.T., West, J.D., Thatcher, S.S., Glasier, A.F. and Baird, D.T. (1987) Post-fertilization polyploidy in human preimplantation embryos fertilized in vitro. Hum. Reprod., 2, 721–727.[Abstract]

A'arabi, S.Y., Roussel, J.D. and Chandler, J.E. (1997) Chromosomal analysis of mammalian oocytes matured in vitro with various culture systems. Theriogenology, 48, 1173–1183.[ISI]

Bahce, M., Cohen, J. and Munné, S. (1999) Preimplantation genetic diagnosis of aneuploidy: were we looking at the wrong chromosomes? J. Assist. Reprod. Genet., 16, 176–181.[ISI][Medline]

Barlow, P.W. and Sherman, M.I. (1972) The biochemistry of differentiation of mouse trophoblast: studies on polyploidy. J. Embryol. Exp. Morphol., 27, 447–465.[ISI][Medline]

Benkhalifa, M., Janny, L., Vye, P., Malet, P., Boucher, D. and Ménézo, Y. (1993) Assessment of polyploidy in human morulae and blastocysts using co-culture and fluorescent in-situ hybridization. Hum. Reprod., 8, 895–902.[Abstract]

Bielanska, M., Tan, S.L. and Ao, A. (2000) Fluorescence in-situ hybridization of sex chromosomes in spermatozoa and spare preimplantation embryos of a Klinefelter 46,XY/47,XXY male. Hum. Reprod., 15, 440–444.[Abstract/Free Full Text]

Bongso, A., Ng, S.C., Lim, J., Fong, C.Y. and Ratnam, S. (1991) Preimplantation genetics: chromosomes of fragmented human embryos. Fertil. Steril., 56, 66–70.[ISI][Medline]

Braude, P., Bolton, V. and Moore, S. (1988) Human gene expression first occurs between the four-and eight-cell stages of preimplantation development. Nature, 332, 459–461.[ISI][Medline]

Clouston, H.J., Fenwick, J., Webb, A.L., Herbert, M., Murdoch, A. and Wolstenholme, J. (1997) Detection of mosaic and non-mosaic chromosome abnormalities in 6- to 8-day old human blastocysts. Hum. Genet., 101, 30–36.[ISI][Medline]

Coonen, E., Harper, J.C., Ramaekers, F.C.S., Delhanty, J.D.A., Hopman, A.H.N., Geraedts, J.P.M. and Handyside, A.H. (1994) Presence of chromosomal mosaicism in abnormal human preimplantation embryos detected by fluorescent in situ hybridization (FISH). Hum. Genet., 54, 609–615.

Coskun, S., Hollanders, J., Al Hassan, S., Al Sufyan, H., Al Mayman, H. and Jaroudi, K. (2000) Day 5 versus day 3 embryo transfer: a controlled randomized trial. Hum. Reprod., 15, 1947–1952.[Abstract/Free Full Text]

Dean, N.L., Philips, J.S., Buckett, W.M., Biljan, M.M. and Tan, S.L. (2000) Impact of reducing the number of embryos transferred from three to two in women under the age of 35 who produced three or more high-quality embryos. Fertil. Steril., 74, 820–823.[ISI][Medline]

Delhanty, J.D., Harper, J.C., Ao, A., Handyside, A.H. and Winston, R.M. (1997) Multicolour FISH detects frequent chromosomal mosaicism and chaotic division in normal preimplantation embryos from fertile patients. Hum. Genet., 99, 755–760.[ISI][Medline]

Edwards, R.G. and Beard, H.K. (1999) Blastocyst stage transfer: pitfalls and benefits. Hum. Reprod., 14, 1–6.[Free Full Text]

Eiben, B., Bartels, I., Bahr-Porsch, S., Borgmann, S., Gatz, G., Gellert, G., Goebel, R., Hammans, W., Hentemann, M. and Osmers, R. (1990) Cytogenetic analysis of 750 spontaneous abortions with the direct-preparation method of chorionic villi and its implications for studying genetic causes of pregnancy wastage. Am. J. Hum. Genet., 47, 656–663.[ISI][Medline]

Evsikov, S. and Verlinsky, Y. (1998) Mosaicism in the inner cell mass of human blastocysts. Hum. Reprod., 13, 3151–3155.[Abstract]

Ford, J.H., Schultz, C.J. and Correll, A.T. (1988) Chromosome elimination in micronuclei: a common cause of hypoploidy. Am. J. Hum. Genet., 43, 733–740.[ISI][Medline]

Hare, W.C., Singh, E.L., Betteridge, K.J., Eaglesome, M.D., Randall, G.C., Mitchell, D., Bilton, R.J. and Trounson, A.O. (1980) Chromosomal analysis of 159 bovine embryos collected 12 to 18 days after estrus. Can. J. Genet. Cytol., 22, 615–626.[ISI][Medline]

Harper, J.C., Robinson, F., Duffy, S., Griffin, D.K., Handyside, A.H., Delhanty, J.D. and Winston, R.M. (1994) Detection of fertilization in embryos with accelerated cleavage by fluorescent in-situ hybridization (FISH). Hum. Reprod., 9, 1733–1737.[Abstract]

Harper, J.C., Coonen, E., Handyside, A.H., Winston, R.M., Hopman, A.H. and Delhanty, J.D. (1995) Mosaicism of autosomes and sex chromosomes in morphologically normal, monospermic preimplantation human embryos. Prenat. Diagn., 15, 41–49.[ISI][Medline]

Harrison, R.H., Kuo, H. C., Scriven, P.N., Handyside, A.H. and Ogilvie, C.M. (2000) Lack of cell cycle checkpoints in human cleavage stage embryos revealed by a clonal pattern of chromosomal mosaicism analysed by sequential multicolour FISH. Zygote, 8, 217–224.[ISI][Medline]

Hassold, T. (1982) Mosaic trisomies in human spontaneous abortions. Hum. Genet., 61, 31–35.[ISI][Medline]

Iwarsson, E., Lundqvist, M., Inzunza, J., Ahrlund-Richter, L., Sjoblom, P., Lundkvist, O., Simberg, N., Nordenskjold, M. and Blennow, E. (1999) A high degree of aneuploidy in frozen–thawed human preimplantation embryos. Hum. Genet., 104, 376–382.[ISI][Medline]

James, R.M. and West, J.D. (1994) A chimaeric animal model for confined placental mosaicism. Hum. Genet., 93, 603–604.[ISI][Medline]

Kalousek, D.K. and Dill, F.J. (1983) Chromosomal mosaicism confined to the placenta in human conceptions. Science, 221, 65–667.[ISI][Medline]

Ledbetter, D.H., Zachary, J.M., Simpson, J.L., Golbus, M.S., Pergament, E., Jackson, L., Mahoney, M.J., Desnick, R.J., Schulman, J. and Copeland, K.L. (1992) Cytogenetic results from the U.S. Collaborative Study on CVS. Prenat. Diagn., 12, 317–345.[ISI][Medline]

LeMarie-Adkins, R., Radke, K. and Hunt, P.A. (1997) Lack of checkpoint control at the metaphase/anaphase transition: a mechanism of meiotic nondisjunction in mammalian females. J. Cell Biol., 139, 1611–1619.[Abstract/Free Full Text]

Long, S.E. and Williams, C.V. (1982) A comparison of the chromosome complement of inner cell mass and trophoblast cells in day-10 pig embryos. J. Reprod. Fertil., 66, 645–648.[Medline]

Los, F. J., Van Opstal, D., van den Berg, C., Braat, A.P., Verhoef, S., Wesby-van Swaay, E., van den Ouweland, A.M. and Halley, D.J. (1998) Uniparental disomy with and without confined placental mosaicism: a model for trisomic zygote rescue. Prenat. Diagn., 18, 659–668.[ISI][Medline]

Magli, M.C., Jones, G.M., Gras, L., Gianaroli, L., Korman, I. and Trounson, A.O. (2000) Chromosome mosaicism in day 3 aneuploid embryos that develop to morphologically normal blastocysts in vitro. Hum. Reprod., 15, 1781–1786.[Abstract/Free Full Text]

Munné, S., Weier, H.U., Grifo, J. and Cohen, J. (1994a) Chromosome mosaicism in human embryos. Biol. Reprod., 5, 373–379.

Munné, S., Grifo, J., Cohen, J. and Weier, H.U. (1994b) Chromosome abnormalities in human arrested preimplantation embryos: a multiple-probe FISH study. Am. J. Hum. Genet., 55, 150–159.[ISI][Medline]

Munné, S., Sultan, K.M., Weier, H.U., Grifo, J.A., Cohen, J. and Rosenwaks, Z. (1995a) Assessment of numeric abnormalities of X, Y, 18, and 16 chromosomes in preimplantation human embryos before transfer. Am. J. Obstet. Gynecol., 172, 1191–1199.[ISI][Medline]

Munné, S., Alikani, M., Tomkin, G., Grifo, J. and Cohen, J. (1995b) Embryo morphology, developmental rates, and maternal age are correlated with chromosome abnormalities. Fertil. Steril., 64, 382–391.[ISI][Medline]

Munné, S., Magli, C. Adler, A., Wright, G., de Boer, K., Mortimer, D., Tucker, M., Cohen, J. and Gianaroli, L. (1997) Treatment-related chromosome abnormalities in human embryos. Hum. Reprod., 12, 780–784.[Abstract]

Murray, J.D., Moran, C., Boland, M.P., Nancarrow, C.D., Sutton, R., Hoskinson, R.M. and Scaramuzzi, R.J. (1986) Polyploid cells in blastocysts and early fetuses from Australian Merino sheep. J. Reprod. Fertil., 78, 439–446.[Abstract]

Papadopoulos, G., Templeton, A.A., Fisk, N. and Randall, J. (1989) The frequency of chromosome anomalies in human preimplantation embryos after in-vitro fertilization. Hum. Reprod., 4, 91–98.[Abstract]

Pickering, S.J., Braude, P.R., Johnson, M.H., Cant, A. and Currie, J. (1990) Transient cooling to room temperature can cause irreversible disruption of the meiotic spindle in the human oocyte. Fertil. Steril., 54, 102–108.[ISI][Medline]

Plachot, M., Junca, A.M., Mandelbaum, J., de Grouchy, J., Salat Baroux, J. and Cohen, J. (1987) Chromosome investigations in early life. II. Human preimplantation embryos. Hum. Reprod., 2, 29–35.[Abstract]

Reddy, K.S. (1997) Double trisomy in spontaneous abortions. Hum. Genet., 101, 339–345.[ISI][Medline]

Robinson, W.P., Binkert, F., Bernasconi, F., Lorda-Sanchez, I., Werder, E.A. and Schinzel, A.A. (1995) Molecular studies of chromosomal mosaicism: relative frequency of chromosome gain or loss and possible role of cell selection. Am. J. Hum. Genet., 56, 444–451.[ISI][Medline]

Ruangvutilert, P., Delhanty, J.D., Serhal, P., Simopoulou, M., Rodeck, C.H. and Harper, J.C. (2000) FISH analysis on day 5 post-insemination of human arrested and blastocyst stage embryos. Prenat. Diagn., 20, 552–560.[ISI][Medline]

Sarto, G.E., Stubblefield, P.A. and Therman, E. (1982) Endomitosis in human trophoblast. Hum. Genet., 62, 228–232.[ISI][Medline]

Steer, C.V. Mills, C.L., Tan, S.L., Campbell, S. and Edwards, R.G. (1992) The cumulative embryo score: a predictive embryo scoring technique to select the optimal number of embryos to transfer in an in-vitro fertilization and embryo transfer programme. Hum. Reprod., 7, 117–119.[Abstract]

Veiga, A., Gil, Y., Boada, M., Carrera, M., Vidal, F., Boiso, I., Ménézo, Y. and Barri, P.N. (1999) Confirmation of diagnosis in preimplantation genetic diagnosis (PGD) through blastocyst culture: preliminary experience. Prenat. Diagn., 19, 1242–1247.[ISI][Medline]

Wang, B.B.T., Rubin, C.H. and Williams, J. (1993) Mosaicism in chorionic villus sampling: an analysis of incidence and chromosomes involved in 2612 consecutive cases. Prenat. Diagn., 13, 179–190.[ISI][Medline]

accepted on October 8, 2001.