Chromosomally abnormal cells are not selected for the extra-embryonic compartment of the human preimplantation embryo at the blastocyst stage

Josien G. Derhaag1,2,4, Edith Coonen1,2, Marijke Bras1,2, J.Marij Bergers Janssen1,2, Rosie Ignoul-Vanvuchelen1,2, Joep P.M. Geraedts1,3, Johannes L.H. Evers1,2 and John C.M. Dumoulin1,2

1 Research Institute Growth and Development (GROW), Maastricht University, 2 Department of Obstetrics and Gynaecology, University Hospital Maastricht and 3 Department of Molecular Cell Biology and Genetics, Maastricht University, Maastricht, The Netherlands

4 To whom correspondence should be addressed at: IVF Laboratory, Department of Obstetrics and Gynaecology, Academic Hospital Maastricht, P.O. Box 5800, 6202 AZ Maastricht, The Netherlands. e-mail: jderh{at}ms-azm-3.azm.nl


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
BACKGROUND: Although well defined for embryos at cleavage stages, the occurrence and frequency of chromosomal aberrations in human blastocysts is relatively unknown. It has been reported that only one in four blastocysts is comprised totally of chromosomally normal cells. One of the selection mechanisms for the embryo proper to become free of these chromosomally abnormal cells would be to sequester them to the extra-embryonic compartment during development. The study aim was to investigate whether such a mechanism of selection exists in human preimplantation embryos. METHODS: Inner cell mass (ICM)/trophectoderm (TE) differentiation was performed, followed by fluorescence in-situ hybridization (FISH), to study the chromosomal distribution in both populations of cells. RESULTS: Of the 94 successfully analysed blastocysts, 68.8 ± 1.5% of all analysable nuclei per blastocyst showed a disomic chromosomal content. Only 22.6% of blastocysts analysed were classified as normal. Of the embryos classified as abnormal at the blastocyst stage, 11.9% showed a simple mosaic pattern and 32.1% a complex mosaic pattern. An equally large group of blastocysts showed either a chaotic pattern (16.7%), or the chromosomal pattern could not be classified. The average degree of normal cells in the ICM (67.9%) was similar to the degree observed in the TE (69.5%). CONCLUSIONS: These findings indicate that chromosomally abnormal cells are not preferentially segregating to the extra-embryonic compartment of the human preimplantation embryo at the blastocyst stage. Hence, other mechanisms should be responsible for an absence of chromosomally abnormal cells in the embryo proper at later stages of development. One possible mechanism might be the elimination of the chromosomally abnormal cells by selective cell death activation.

Key words: chromosomal mosaicism/fluorescence in-situ hybridization/human blastocyst/inner cell mass/trophectoderm


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
In human preimplantation embryos—especially those with abnormal morphology—chromosomal mosaicism has been shown to be a frequent feature (Bongso et al., 1991Go; Pellestor et al., 1994aGo; b). Reported proportions of chromosomally abnormal preimplantation embryos ranged between 30 and 70% in embryos at day 2 or 3 of development (Munné et al., 1993Go; 1997; Delhanty et al., 1997Go; Laverge et al., 1997Go). Data appear to depend on the number of probes applied simultaneously, the type of probes used, the embryo morphology, embryo development, and the presence of multinucleated blastomeres. Numerical chromosome abnormalities often originate during male or female meiosis, but they may also arise at the time of fertilization or during further post-zygotic development to the blastocyst stage.

Reports of the presence of chromosomal abnormalities in embryos at later stages of development are relatively rare. At the cleavage stages, 29% of morphologically normal human embryos have been shown to be chromosomally abnormal (Munné et al., 1995Go), while in 10-day-old pig embryos (Long and Williams, 1982Go), in 13- to 14-day-old sheep blastocysts (Murray et al., 1986Go) and in bovine blastocysts obtained after IVF (Iwasaki et al., 1992Go) polyploidy and in particular mixoploidy (i.e. mosaicism of diploid and polyploid cells) has been reported. In an early study of human blastocysts (Benkhalifa et al., 1993Go), chromosomal mosaicism was reported to be present in 29% of cases, and it was noted later that the percentage of embryos showing chromosomal mosaicism increased to almost 100% at the blastocyst stage. The proportion of abnormal cells per embryo was 16% (Bielanska et al., 2000Go; 2002; Leonetti et al., 2000Go; Ruangvutilert et al., 2000Go). Data acquired by others (Sandalinas et al., 2001Go) when investigating 54 blastocysts showed that, during development up to the blastocyst stage, there was no definite selection against most of chromosomal abnormalities which occurred at cleavage stages. Another group (Clouston et al., 2002Go) examined human blastocysts using classical cytogenetic techniques and reported that, when compared with cleavage-stage embryos, there was a decrease in embryos which revealed a haploid or monosomic chromosomal content, as well as some trisomies. Studies applying the comparative genome hybridization technique reported that 25% (Voullaire et al., 2000Go) and in some cases even none of the embryos (Wells and Delhanty, 2000Go) comprised only normal cells. In the present group, a large population of blastocysts was studied and it was concluded that the proportion of embryos containing chromosomally abnormal cells has increased dramatically when compared with the chromosomal status of morphologically normal developing cleavage-stage embryos (Coonen et al., 2000Go). In summary, the aforementioned studies in general suggest that a chromosomally aberrant cell population is present in nearly all human spare embryos at each stage of cleavage division, and also at the blastocyst stage.

Since newborns have chromosomal abnormalities at a much reduced frequency (0.6%) (Plachot et al., 1987Go) than preimplantation embryos, a selection mechanism must be in operation. The low frequency of chromosomal abnormalities within the newborn might possibly be the result of segregation of the abnormal cells to the compartments or tissues that are less critical for implantation or post-implantation development. At the blastocyst stage, two distinct cell lineages can be recognized. The trophectoderm (TE) gives rise to most of the extra-embryonic parts such as the embryonic membranes and parts of the placenta, while the inner cell mass (ICM) provides cells for the formation of the embryo proper (McLaren, 1982Go). The fact that optimal development of the ICM has highest priority is in line with the finding of a homeostatic control of ICM cell numbers in preimplantation embryos up to moderate levels of fragmentation (Hardy et al., 2003Go). The existence of these two compartments is the result of incidental partitioning of cells between the two compartments. Individual early blastomeres are not yet committed to form either TE or ICM, but instead can form either rudiment (Mottla et al., 1995Go). At a certain time point, the differentiation process appears to become irreversible and both cell populations develop their own qualities—for example, the formation of a tight junction impermeable seal between TE cells (Wiley, 1987Go). As each blastomere of the human early cleavage-stage embryo can participate in both TE and ICM formation (Mottla et al., 1995Go), chromosomally abnormal blastomeres and normal blastomeres have the same theoretical change of contributing to the ICM as to the TE when no selection mechanism is present. The preferential allocation of chromosomally abnormal cells to one of the compartments, or a more efficient elimination of chromosomally abnormal cells from one of the compartments, results in an unequal proportion of chromosomally abnormal cells in the two cell compartments.

The aim of the present study was to investigate whether a mechanism of selection, which is reflected by the presence of more chromosomally aberrant cells in the TE than in the ICM, exists in human preimplantation embryos.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Human preimplantation embryos
Surplus IVF and ICSI embryos originating from normally fertilized (2 pronuclear; 2PN) zygotes, and which were considered unsuitable for freezing, were cultured until day 5–6 following fertilization in two different culture media. Embryos from each consecutive treatment cycle were alternately cultured in either IVF-50 (Scandinavian IVF-Science AB, Göteborg, Sweden) or in an ‘in-house’ prepared human tubal fluid (HTF) medium (Quinn et al., 1985Go) which was supplemented with 9% (v/v) plasma protein solution. On the second and third days after oocyte recovery, the developmental stages and morphological aspects of embryos originating from normally fertilized oocytes were assessed according to published criteria (Bolton et al., 1989Go). For each embryo, a score was calculated by multiplying the morphological grade by the number of blastomeres (Steer et al., 1992Go). If available, one or two embryos (depending on their developmental stage and morphological appearance as well as the age of the patient) were transferred. After transfer, all supernumerary embryos were cultured until day 3 after oocyte retrieval. On the third day, cryopreservation was performed if at least one embryo reached the 8-cell stage, combined with a good morphology, grades 3 and 4 according to published procedures (Bolton et al., 1989Go; Dumoulin et al., 2000Go). Surplus embryos were left in their original culture medium. Developmental stages were recorded once per day, during in-vitro development. On the morning of day 5 or 6 after oocyte retrieval, surplus embryos that had developed to the expanded blastocyst stage (defined as TE surrounding a cavity of extracellular fluid, accumulated within the embryo and the presence of a distinct ICM) were collected. Embryos that had already reached the blastocyst stage on day 5 were differentially labelled and spread to avoid hatching, and thus were not allowed to develop into day 6 blastocysts. Surplus embryos were used for research only if written informed consent had been obtained from participating patients. Investigation of the embryos was approved by the local Research Ethics Committee. IVF or ICSI cycles without any blastocyst formation were excluded from the present study.

Differential ICM/TE labelling
ICM and TE nuclei of the blastocysts were differentially labelled using a previously described method (Handyside and Hunter, 1984Go), albeit with modifications. Specific labelling of TE with propidium iodide (PI) (Sigma, St Louis, MO, USA) was achieved by rendering these cells permeable using incomplete antibody-mediated complement lysis. Both TE and ICM nuclei were labelled in situ with 4',6-diaminidino-2-phenylindole (DAPI). Zonae were removed by incubation for a few seconds in acid Tyrode’s solution (pH 1.5) or in pronase, 500 U/ml (Sigma). Zona-free blastocysts were washed in HEPES-buffered medium containing 4 mg/ml bovine serum albumin (BSA; Sigma) (HBSA). In order to label cell-surface proteins with trinitrophenol (TNP) groups, zona-free blastocysts were incubated in 10 mmol/l trinitrobenzenesulphonic (TNBS) acid (Sigma) in modified Tyrode’s medium (T6) containing 4 mg/ml polyvinylpyrrolidone (PVP; Sigma) on ice for 10 min, washed three times with HBSA and incubated with a rabbit antiserum against dinitrophenol (DNP) (0.1 mg/ml) + 3% normal human serum (NHS) for 20 min at 37°C, which cross-reacts with the TNP-labelled proteins. After further washing with HBSA, blastocysts were incubated in a 1:10 dilution of guinea pig serum as complement source in HBSA containing 0.01 mg/ml PI, at 37°C for 10–15 min. Differential labelling was checked using a fluorescence microscope. Blastocysts were briefly washed in phosphate-buffered saline (PBS) for 10 min.

Spreading of the labelled blastocysts
Spreading of the labelled blastocyst was performed as described previously (Coonen et al., 1994Go), with modifications. In short, blastocysts were gently transferred with a minimal amount of PBS, using a small glass pipette to a microdrop (1–2 µl) of spreading solution containing 1 mmol/l HCl/1% Tween 20 (Janssen Chimica, Belgium) in double-distilled water on coated slides (Super Starfrost Plus®; Meanzel Glaezer, Braunschweig, Germany) using a microscope equipped with a Linkam heated stage (Paes Netherland b.v.). All of the cytoplasm was removed as quickly as possible. On occasion, 10 mmol/l HCl/1% Tween 20 solution was needed for removal of all remaining cytoplasm. The slides were air-dried, washed in PBS, and dehydrated through an ascending ethanol series. Slides were mounted in glycerol (100%) containing 1.25 ng/ml DAPI to stain all nuclei. After examination and recording of the results of the differential labelling (see Data analysis), the slides were rinsed in 4x standard saline citrate (SSC) containing 0.05% Tween 20 at room temperature (RT) until the coverslips detached spontaneously. The slides were then rinsed for a further 10 min in 4x SSC/0.05% Tween 20, and dehydrated.

Direct labelling of DNA probes
The DNA probes used to study the chromosomes of interest were: PBamX5: alphoid probe (insert size 2.0 kb), specific for the centromeric region of the human X chromosome (Willard et al., 1983Go); DYZ3, satellite probe (insert size 2.1 kb) specific for the long arm of the human Y chromosome (Cooke et al., 1982Go); and 1.84, satellite probe (insert size 0.68 kb), specific for the centromeric region of the human chromosome 18. Probes were directly labelled by nick translation with fluoroscein isothiocyanate (FITC)-12-dUTP (Boehringer-Mannheim, Germany; X and 18 probes) or rhodamine-4-dUTP (Amersham, UK; X and Y probes), dissolved in hybridization mixture (60% formamide/2x SSC, pH 7.0) and used at a final concentration of 1–2 ng/µl.

FISH
Slides were treated with pepsin (100 µg/ml) in 10 mmol/l HCl for 10 min at 37°C to remove any remnants of cytoplasm and to render the nuclei accessible for hybridization to the probes. The slides were rinsed in PBS and fixed for 5 min in 1% paraformaldehyde in PBS at 4°C. After fixing, the slides were rinsed in PBS and dehydrated through an ethanol series. The hybridization mixture was added to the slide under a coverslip, and the nuclear and probe DNA were heat-denatured simultaneously for 3 min at 70°C. The slides were then incubated in a moist chamber at 37°C for at least 2 h to allow hybridization of the DNA probes to the embryonic DNA. After hybridization, the slides were washed for 5 min at 42°C with 2x SSC/0.05% Tween 20, followed by 5 min at 60°C with 0.01x SSC and 5 min at RT with 4x SSC/0.05% Tween 20. The slides were dehydrated through an ethanol series and mounted in glycerol (100%) containing antifade (DABCO; Sigma) and 1.25 ng/ml DAPI to stain the nuclei. Nuclei were examined using a Leica DMRBE fluorescence microscope equipped with separate filters specific for FITC and rhodamine to detect FISH signals, and with a triple FITC/TRITC/DAPI excitation filter to exclude overlap of signals. Published scoring criteria (Munné et al., 1996Go) were used to differentiate FISH errors from mosaicism. The efficiency of the FISH was tested in each experiment on interphase nuclei obtained from ethanol-fixed, single-cell suspensions of healthy male leukocytes as described previously (Harper et al., 1994Go). Data were collected at a time when only two-colour FISH was available to the authors, but they are well aware of the consequences of this technical inadequacy with respect to the differentiation of chromosome abnormalities.

Data analysis
The position of the nuclei on the slide was recorded using a schematic drawing and a digitized image (using a CCD camera and software from Perceptive Scientific Instruments), and the ICM/TE ratio was calculated. After FISH, nuclei were reallocated using the schematic picture, and the chromosomal constitution was combined with the earlier found ICM/TE specification. Cells that were lost or were without interpretable FISH signals were recorded as not analysable (NA). Only blastocysts consisting of at least 25 cells were taken into account, and >=75% of their nuclei had to be analysable after FISH.

Blastocyst classification
In order to clarify the nature of the chromosomal patterns displayed at the blastocyst stage, preimplantation embryos were classified according to their most prominent chromosomal feature(s). Blastocysts containing a relative majority of disomic nuclei were regarded as chromosomally normal in origin (directly after fertilization) for the chromosomes under study. All others were classified as abnormal in origin.

Embryos were regarded as chromosomally normal at the blastocyst stage if not more than one cell revealed a non-disomic pattern, or in case of multiple non-disomic cells, if this pattern seemed to have arisen from chromosome doubling (i.e. four copies for all chromosomes investigated). All other embryos were regarded as chromosomally mosaic at the blastocyst stage. Chromosomal mosaicism was defined as the coexistence of two or more chromosomally distinctive cell populations (and thus two or more chromosomally distinctive karyotypes) within a single embryo. For any chromosomal karyotype to be biologically meaningful, it had to be present in at least two cells (ISCN, 1978Go).

Mosaic chromosome patterns usually seem to originate from errors occurring at the second or later cleavage division. Two major pathways can be distinguished that may contribute to the chromosomal variation seen in mosaic chromosome patterns, namely mitotic non-disjunction and anaphase lagging. With this in mind, blastocysts were classified into the following categories: <-, a simple mosaic chromosome pattern (different cell lines resulting from one chromosomal error); {uparrow}, a complex mosaic chromosome pattern (different cell lines resulting from more than one chromosomal error); and ->, a chaotic chromosome distribution pattern (four or more chromosomally unrelated cell lines). The patterns that needed an extraordinary number of chromosomal errors to be explained, were defined as unexplained mosaics.

As only two pairs of chromosomes were tested, care must be taken with the classification of cells and embryos in the above-mentioned categories.

Statistical analysis
Data were analysed using the {chi}2-test and paired or unpaired Student’s t-test as appropriate. To detect correlation, linear regression analysis was used, and Pearson’s correlation coefficient calculated.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
A total of 433 surplus embryos originating from 2PN zygotes, remaining after transfer and unsuitable for freezing, were collected at day 5 or 6. Of the surplus embryos from ART cycles in which at least one spare embryo developed to the blastocyst stage, 164 (37.8%) developed to the blastocyst stage. The overall blastocyst formation rate in the regular IVF programme during the period in which data underlying this report were collected was 26.4% (Dumoulin et al., 2000Go). ICM/TE differentiation procedures were carried out successfully in 94 cases (57.3%). Unsuccessful differential labelling resulted from unaccountable causes. The total mean (± SEM) cell number for the whole embryo series was 78.3 ± 4.3, and was 61.6 ± 5.8 (n = 27) and 83.7 ± 5.1 (n = 67) for day 5 and day 6 blastocysts respectively (Table I). The numbers of ICM and TE cells in the blastocysts on days 5–6 are presented in Table I. Comparison of the expanded blastocysts on day 5 with the expanded blastocysts on day 6 showed that the mean number of ICM and TE cells each increased during further development. The contribution of the ICM to the total embryo was also slightly increased between days 5 and 6.


View this table:
[in this window]
[in a new window]
 
Table I. Numbers of inner cell mass (ICM) and trophectoderm (TE) cells in blastocysts originating from 2PN zygotes grown in vitro and developed to blastocyst stage at day 5 or 6
 
After ICM/TE differentiation and fixation of the nuclei of the blastocyst, two types of nuclei were observed. One population had small, condensed round nuclei, which in most cases were of TE origin, while a second population with larger round nuclei were of ICM origin. This condensation of the TE nuclei appeared to be the result of the differential labelling procedure (complement activation) as it was never observed after sole spreading and FISH analysis of the blastocysts.

ICM/TE differentiation was followed by FISH analysis. Differential labelling was only successful on morphologically good blastocysts (Bolton et al., 1989Go) (Figure 1A). Prolongation of the complement reaction resulted either in nuclear loss or in condensation of the DNA. The latter resulted in non-analysable nuclei after FISH. Spreading of the differentially labelled blastocysts resulted in blue ICM nuclei (DAPI) and pink TE nuclei (DAPI + PI) (Figure 1B). The degree of chromosomally normal/abnormal nuclei in the ICM and TE was determined by analysing the copy numbers for the chromosomes X, Y and 18. Among a population of 94 successful differential labelling procedures, 84 (89.4%) produced clear FISH signals in at least 75% of their nuclei. When chromosomal features were determined in these blastocysts, all were of normal chromosomal origin (two copies) for the chromosomes studied. Only 22.6% of the blastocysts analysed were chromosomally normal for the chromosomes tested at the blastocyst stage. Of the embryos classified as abnormal at the blastocyst stage, 11.9% showed a simple mosaic pattern and 32.1% a complex mosaic pattern. An equally large group of blastocysts showed either a chaotic pattern (16.7%) or could not be classified on the basis of chromosomal errors (Table II). In unravelling the nature of the chromosomal patterns, the incidence of the various karyotypes was calculated (Table III). One in two mosaic blastocysts contained cells monosomic for all chromosomes tested, with a mean incidence of 5% per blastocyst. The monosomic karyotype was most frequently observed in complex mosaic embryos, although the percentage of monosomic cells did not vary among the different chromosome patterns. All mosaic embryos contained disomic cells. Trisomic nuclei were observed in one-quarter of all mosaic embryos, the incidence being almost equal to the proportion of monosomic nuclei. Tetrasomic nuclei, indicative of chromosome doubling, were seen in one in three mosaic blastocysts, with a mean proportion of 5.6%. The ‘SSA’ karyotype was the one most frequently noticed in mosaic blastocysts. Three-quarters of all blastocysts were comprised of cells which had lost one of their chromosomes 18. About half of the mosaic blastocysts contained cells that displayed only one sex chromosome, with a preference for the complex/chaotic pattern. The ‘SSAAA’ and ‘SSSAA’ karyotypes were less prominent features of human mosaic blastocysts. The ‘SSSA’ and ‘SAAA’ karyotypes were hardly observed in mosaic blastocysts.



View larger version (24K):
[in this window]
[in a new window]
 
Figure 1. (A) Ethanol-fixed blastocyst after differential labelling of inner cell mass (ICM) and trophectoderm (TE). TE nuclei appear pink; ICM nuclei appear blue. The DNA was counterstained using Hoechst 33256. (B, C and E) Nuclei after differential labelling followed by the spreading procedure; TE nuclei appear pink, ICM nuclei appear blue. DNA was counterstained using DAPI. (D) Same nuclei as (C) after hybridization with a probe cocktail consisting of a probe specific for chromosome X (labelled in green and red, yellow signal), a probe specific for chromosome Y (red signal), and a probe specific for chromosome 18 (green signal). Both nuclei are normal diploid. (F) Same nuclei as (E) after fluorescence in-situ hybridization, i = aberrant nuclei X,X,18,18,18; ii = aberrant nuclei X,X,18.

 

View this table:
[in this window]
[in a new window]
 
Table II. Chromosomal classification of human preimplantation embryos grown in vitro until the blastocyst stage
 

View this table:
[in this window]
[in a new window]
 
Table III. Incidence of chromosomal karyotypes in mosaic human blastocysts
 
In both the ICM and TE cell populations, normal (disomic for chromosome X, Y and 18; Figure 1C and D) and abnormal (monosomic, trisomic; Figure 1E and F) nuclei were detected. Some 74.1% of all analysable nuclei per blastocyst showed a normal chromosomal content with respect to the sex chromosomes and chromosome 18. No significant difference was found between the mean percentages of normal cells in the ICM and the TE (paired Student’s t-test, P = 0.08). The individual chromosomes also showed no significant differences in the distribution of aberrations between the two compartments. The blastocysts contained a mean of 77.1 ± 1.4% cells that were disomic for the sex chromosomes, and 77.6 ± 1.1% for chromosome 18 (Table IV). The distribution of aberrations between the two compartments resulted in a mean ICM/TE ratio of disomic cells of 0.97 for sex chromosomes and 0.96 for chromosome 18.


View this table:
[in this window]
[in a new window]
 
Table IV. Chromosomal features of the inner cell mass (ICM) and trophectoderm (TE) (n = 84)
 
Populations of embryos classified as normal and abnormal (simple mosaic, complex mosaic, chaotic and unexplained) at the blastocyst stage were studied separately for both the sex chromosomes and chromosome 18. ICM/TE cell number distributions and total cell number of these subpopulations are presented in Table V. The ICM/TE cell number ratio of the normal cell population was not significantly different from that of the blastocyst population classified abnormal (unpaired Student’s t-test, P = 0.29). The mean percentages of nuclei disomic for the chromosomes tested in the ICM and TE were not significantly different (paired Student’s t-test, P < 0.05) within all the subpopulations tested (Table IV). More detailed data relating to the results of the separate chromosomes are shown in Table IV.


View this table:
[in this window]
[in a new window]
 
Table V. Cell number of inner cell mass (ICM) and trophectoderm (TE) respectively, and of the whole blastocyst analysable for FISH (n = 84)
 
The blastocysts were further analysed by studying the correlation between the overall incidence of abnormal cells per blastocyst and the ICM/TE ratio for percentage disomy, analysed for both the sex chromosomes and chromosome 18. No correlation was found between the two parameters (Pearson’s correlation, r = 0.15, P > 0.05).

Since in previous non-human studies an allocation of tetraploid cells to the TE was noticed, the presence of 4N cells was studied separately. Tetrasomic cells (for both the sex chromosomes and chromosome 18) constituted 2.3% of the total number of cells in the entire blastocyst population. For those embryos in which >10% of the cells were tetrasomic, differences were seen in the distribution of the polyploid cells between ICM and TE (Table VI). In these embryos (n = 5), tetrasomic cells appeared to be more frequently present in the ICM than in the TE, but the number of embryos was too small to be valid for statistical analysis.


View this table:
[in this window]
[in a new window]
 
Table VI. Tetraploidy distribution determined in a blastocyst population with >10% tetrasomic nuclei (n = 5)
 
Embryos were collected during either IVF (n = 39) or ICSI (n = 45) ART cycles in which at least one spare embryo developed to the blastocyst stage. No differences were noted in the percentages of disomic cells in IVF and ICSI blastocysts (unpaired Student’s t-test, P = 0.164). Blastocysts classified as chromosomally normal and abnormal at the blastocyst stage were equally distributed among the IVF and ICSI populations ({chi}2-test, P < 0.05). The sex chromosome distribution among the blastocysts was XX = 36/XY = 48.


    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
To the present authors’ knowledge, this is the first large-scale study of the distribution of chromosomally abnormal cells over ICM and TE in human blastocysts. In accordance with a previous study of the nature and origin of chromosomal patterns in a large population of human embryos at the blastocyst stage (Coonen et al., 2000Go), and also in the embryo population underlying the present report, almost all blastocysts studied contained abnormal cells 74.8% of the cells were normal. Only one embryo turned out to be completely normal in all cells for both the sex chromosomes and chromosome 18. The observations of others (Benkhalifa et al., 1993Go; Evsikov and Verlinsky, 1998Go; Bielanska et al., 2000Go; Leonetti et al., 2000Go) were in agreement with the present results, and suggest that the presence of chromosomal abnormalities is a common feature during the development of spare embryos obtained after hyperstimulation in an IVF programme. The embryos of best quality (according to the laboratory standards employed; see Materials and methods) were either used for transfer or frozen as part of the IVF/ICSI procedure. Although these embryos should be studied in order to obtain complete information regarding the chromosomal constitution of preimplantation embryos at the blastocyst stage, this has insurmountable ethical limitations within an infertile population. The fact that the incidence of chromosomal aberrations appears rather high in preimplantation embryos requires further investigation, and this should result in further insight into the development of preimplantation embryos to the blastocyst stage. This is especially important when selecting the embryo with the greatest chance of implantation, and this becomes increasingly important during assisted reproduction schemes when only one single blastocyst is transferred electively.

The selection against aneuploid embryos is thought to occur at the time of morula/blastocyst transition (Edwards, 1986Go). It can be hypothesized that there exists a certain ‘cut-off point’, meaning that when the proportion of abnormal cells exceeds a specific percentage a selection mechanism results in fragmentation of the embryonic cells and degradation of the embryo. This may explain the fact that during development of day 3 spare embryos to the blastocyst stage, only one-third of the blastocysts developed to the expanded blastocyst stage. Data relating to the chromosomal status of embryos that did not reach the blastocyst stage—that is, the majority of the embryos—were unfortunately not collected in the present study. Hence, any comparison of the chromosomal status of early cleavage-stage embryos with data obtained from blastocysts was biased by this phenomenon.

The existence of blastocysts with a relatively high percentage of abnormal cells, as found in the present study, suggests that the presence of an abnormal subpopulation of cells is not detrimental to blastocyst formation and does not lead to severe impairment of preimplantation development. It has been shown previously that different chromosomal abnormalities and loads of abnormal cells in mosaics die at different stages of development. Monosomies, with the exception of monosomy X and 21, haploidies and aneuploidies, combined with extensive mosaicism do not preclude full differentiation in vitro (Sandalinas et al., 2001Go). These authors reported that chaotic preimplantation embryos do not reach more than 60 cells. However, the present chaotic blastocyst population showed a mean of 91.9 ± 11.3 cells, which was the largest number of cells in the abnormal blastocyst population. This discrepancy might be caused by differences in experimental design and variation in classification of the different chromosome patterns. Furthermore, it should be borne in mind that in the underlying study only a limited number of chromosomes was investigated.

It might be assumed that a relatively high percentage of abnormal cells is also present among the preimplantation embryos transferred, but newborns have much fewer (0.6%) chromosomal abnormalities than blastocysts (Plachot et al., 1987Go). Therefore the hypothesis was investigated that, in the embryo at the blastocyst stage, chromosomally abnormal cells would be preferentially segregated to the TE, which does not take part in formation of the embryo proper.

Differential labelling of ICM/TE was successful in only 57.3% of cases. Unsuccessful differential labelling procedures resulted in 100% PI/DAPI-positive nuclei, which suggested that both the ICM and TE cell populations were labelled with antibodies. This may have been caused by leakage of TNBS through the tight junction barrier, resulting in labelling of the cell surface proteins of both ICM and TE cells. This phenomenon does not seem to be related to morphology as scored in standard IVF procedures. Non-random selection of a specific blastocyst population by this phenomenon can be excluded, because the chromosomal features (percentage normal cells and relevant abnormal genotypes found) of the investigated blastocyst population described herein do not differ from the data reported previously (Coonen et al., 2000Go) which described a large-scale chromosomal inventory of whole human blastocysts.

In the population studied, 37.9% of the 2PN surplus zygotes reached the expanded blastocyst stage. This value was high in comparison with that of 21.6% reported elsewhere in large populations of preimplantation embryos obtained after IVF and ICSI (Shoukir et al., 1998Go; Coonen et al., 2000Go; Dumoulin et al., 2000Go; Sandalinas et al., 2001Go). In the present study, this was the result of the exclusion of IVF or ICSI cycles without any blastocyst formation. A comparison of the proportions of blastocyst ICM that reached this stage on day 5 or day 6 showed that the total cell numbers and proportion of ICM cells increased during development. Comparable data have been reported by others (Hardy et al., 1989Go), although these authors indicated that numbers of TE and ICM cells both doubled between days 5 and 6 in human blastocysts—a larger increase than seen in the present blastocyst population.

The mosaic chromosome patterns observed in the blastocysts usually seem to originate from errors occurring at the second or later cleavage division, giving rise to a diploid-derived chromosomal pattern. The ‘SSA’ karyotype was the one most frequently noted. Approximately 75% of all mosaic blastocysts contained some cells that had lost one of their chromosomes 18, and non-disjunctional processes appear to be the most logical cause with respect to chromosome 18 aneuploidy (Hassold and Hunt, 2001Go). Embryos with a simple chromosome pattern were mostly comprised of cells having a ‘SSA’ karyotype, caused by anaphase lagging of chromosome 18. The complex chromosome pattern contained mostly ‘SA’, ‘SSA’ and ‘SAA’ karyotypes, almost exclusively caused by anaphase lagging of one of the sex chromosomes accompanied by anaphase lagging of the chromosome 18.

Others (Evsikov and Verlinsky, 1998Go) have shown that the chromosomal abnormalities observed in human blastocysts must be equally present in both compartments of the blastocyst. Another group (Magli et al., 2000Go) studied 18 ICM compartments of aneuploid preimplantation embryos that were grown in vitro until the blastocyst stage, but their data did not support the hypothesis of a preferential allocation of euploid cells to the ICM. However, as both studies were based on few blastocysts, more information about the ICM as well as the TE cell population is required. In the present study, determination of the degree of mosaicism in the blastocyst was not based solely upon analysis of the ICM, as was carried out by others (Evsikov and Verlinsky, 1998Go; Magli et al., 2000Go). The present study provided the opportunity to investigate both cell populations and single cells in more detail. As each blastomere of the early cleavage-stage human embryo can (in theory) participate in either TE and/or ICM formation (Mottla et al., 1995Go), it can be assumed that all chromosomally abnormal blastomeres have the same chance of contributing to the ICM as normal blastomeres. This ignores the fact that the contribution of the ICM to the blastocyst increases from day 5 to day 6 during development, resulting in a higher theoretical chance for cells with chromosomal aberrations to be present in the ICM of a blastocyst. Experiments in mice suggest that, prior to implantation, blastocysts have already lost some chromosomally abnormal cells from their embryonic and extra-embryonic lineages due to a combination of preferential allocation of the abnormal cells to the TE and selection against chromosomally abnormal cells in the embryo (Everett and West, 1998Go). The present data show that the chromosomal aberrations studied (for the chromosomes X, Y and 18) were present at the same rate in both ICM and TE populations. Classification of the blastocysts into the two subpopulations which were chromosomally normal and abnormal at the blastocyst stage showed that, in both blastocyst populations, the percentages of disomic cells were comparable in the TE and the ICM compartment. In embryos classified as chromosomally normal at the blastocyst stage, the highest percentage of disomic cells was always found in the ICM compartment, whereas in embryos classified as chromosomally abnormal, the highest percentage of disomic cells was always found in the TE compartment. This could imply that, in normal blastocysts with an excess of chromosomally abnormal cells, a selection mechanism seems to be lost beyond a certain ‘cut-off point’. However, this suggestion is not in line with there being no correlation between the percentage disomy of the whole blastocyst with the ‘% disomy ICM/% disomy TE ratio’.

The presence of tetraploid cells is a phenomenon which has been suggested to be a part of normal development (Murray et al., 1986Go; West, 1990Go), although these 4n cells seem to be allocated preferentially to the TE. The allocation of polyploid (mainly 4n) cells was observed in the trophoblast in pig blastocysts (Long and Williams, 1982Go), while others reported that in bovine blastocysts resulting from IVF, the polyploid cells were located in the TE rather than in the ICM (Iwasaki et al., 1992Go). This implies that, in these species, the embryo proper may be capable of sequestering these chromosomally polyploid cells during preimplantation to the trophectodermal (non-embryonic) compartment. In human embryos, a high proportion of polyploid cells may be detrimental (Sandalinas et al., 2001Go). In contrast to data relating to non-human preimplantation embryos, the present study (despite comprising very few blastocysts) demonstrated an allocation of tetrasomic cells to the ICM rather than to the TE in human blastocysts. This predisposition of 4n cells to the ICM in human preimplantation embryos might be explained by the relatively high proliferation index of the ICM cells (in comparison with TE cells) during days 5 and 6 of embryonic development. However, this does not explain the difference between embryos from humans (presence of chromosomally abnormal cells in the ICM) and other species (presence in the TE). Further analysis of more chromosomes (e.g. chromosome 16, for which more aberrant cells may be expected) will most likely provide more information about the distribution and redistribution of chromosomally aberrant cells among both cell populations.

In conclusion, it can be stated that during development of the human embryo, comparable numbers of chromosomally aberrant cells are present in the embryo proper as well as in the extra-embryonic compartment. These findings confirm those of others which were presented earlier (Evsikov and Verlinsky, 1998Go; Magli et al., 2000Go). Despite using an intensive examination approach, the results obtained in animal experiments could not be confirmed by the present investigation, and alternative mechanisms must be responsible for the absence of aneuploidy at the later stage of post-implantation embryo, and also in the fetus during further development. This situation may be brought about by a decrease of proliferation or an increase in cell death activation of the chromosomally abnormal cells that are found in almost all preimplantation embryos during development up to the blastocyst stage.


    Acknowledgements
 
The authors wish to thank all members of the IVF team, University Hospital Maastricht, The Netherlands, for their technical and clinical assistance.


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Benkhalifa, M., Janny, L., Vye, P., Malet, P., Boucher, D. et al. (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, O. (2000) Significance of mosaicism in the human preimplantation embryo. Hum. Reprod., 15, Abstract book 1, 12.

Bielanska, M., Tan, S.L. and Ao, O. (2002) Chromosomal mosaicism throughout human pre-implantation development in vitro: incidence, type, and relevance to embryo outcome. Hum. Reprod., 17, 413–419.[Abstract/Free Full Text]

Bolton, V., Hawes, S., Taylor, C. and Parsons, J. (1989) Development of spare human pre-implantation embryos in vitro: an analysis of the correlations among gross morphology, cleavage rates, and development to the blastocyst. J. In Vitro Fertil. Embryo Transfer, 6, 30–35.[ISI][Medline]

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]

Clouston, H.J., Herbert, M., Fenwick, J., Murdoch, A.P. and Wolstenholme, J. (2002) Cytogenetic analysis of human blastocysts. Prenat. Diagn., 22, 1143–1152.[CrossRef][ISI][Medline]

Cooke, H.J., Schmidtke, J. and Gosden, J.R. (1982) Characterisation of a human Y chromosome repeated sequence and related sequences in higher primates. Chromosoma, 87, 491–502.[ISI][Medline]

Coonen, E., Dumoulin, J.C.M., Ramaekers, F.C.S. and Hopman, A.H.N. (1994) Optimal preparation of preimplantation embryo interphase nuclei for analysis by fluorescent in situ hybridisation. Hum. Reprod., 3, 533–537.

Coonen, E., Dumoulin, J.C.M., Derhaag, J.G., Van Wissen, L.C.P., Bras, M., Bergers-Jansen, J.M. et al. (2000) Genetic make-up of human embryos grown in vitro until the blastocyst stage. Hum. Reprod., 15, Abstract book 1, 180,

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

Dumoulin, J.C.M., Coonen, E., Bras, M., Wissen van, L.C.P., Ignoul van Vuchelen, R. et al. (2000) Comparison of in vitro development of embryos originating from either conventional IVF or ICSI. Hum. Reprod., 15, 402–409.[Abstract/Free Full Text]

Edwards, R.G. (1986) Causes of early embryonic loss in human pregnancy. Hum. Reprod., 3, 185–196.

Everett, C.A. and West, J.D. (1998) Evidence for selection against tetraploid cell in tetraploid<->diploid mouse chimaeras before the late blastocyst stage. Gene Res. Camb., 72, 225–228.

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

Handyside, A.H. and Hunter, S. (1984) A rapid procedure for visualising the inner cell mass and trophectoderm nuclei of mouse blastocysts in situ using polynucleotide specific fluorochromes. J. Exp. Zool., 231, 429–434.[ISI][Medline]

Hardy, K., Handyside, A.H. and Winston, R.M.L. (1989) The human blastocyst: cell number, death and allocation during late preimplantation development. In Vitro Dev., 107, 597–604.

Hardy, K., Stark, J. and Winston, R.M.L. (2003) Maintenance of the inner cell mass in human blastocysts from fragmented embryos. Biol. Reprod., 68, 1165–1169.[Abstract/Free Full Text]

Harper, J.C., Coonen, E., Ramaekers, F.C.S., Delhanty, J.D.A., Handyside, A.H., Winston, R.M.L. and Hopman, A.H.N. (1994) Identification of the sex of human pre-implantation embryos in two hours using an improved spreading method and fluorescent in-hybridization (FISH) using directly labelled probes. Hum. Reprod., 9, 721–724.[Abstract]

Hassold, T. and Hunt, P. (2001) To err (meiotically) is human: the genesis of human aneuploidy. Nat. Rev. Genet., 2, 280–291.[CrossRef][ISI][Medline]

ISCN (1978) Report of the Standing Committee on Human Cytogenetic Nomenclature. An international system for human cytogenetic nomenclature. Cytogenet. Cell Genet., 21, 309–409.[Medline]

Iwasaki, S., Hamano, S., Kuwayama, M., Yamashita, M., Ushijima, H. et al. (1992) Developmental changes in the incidence of chromosome anomalies of bovine embryos fertilized in vitro. J. Exp. Zool., 261, 79–85.[ISI][Medline]

Laverge, H., De Sutter, P., Verschaegen-Spae, M.R., DePaepe, A. and Dhont, M. (1997) Triple colour fluorescent-in situ hybridisation for chromosome X, Y, and 1 on spare human embryos. Hum. Reprod., 12, 809–814.[Abstract]

Leonetti, T., Fischetto, R., Schonauer, L.M. and Causio, F. (2000) Mosaicism evaluation by FISH in human preimplantation embryos. Hum. Reprod., 15, Abstract book 1, 185.

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.[CrossRef][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]

McLaren, A. (1982) Reproduction in mammals: 2. Embryonic and fetal development. In Austin, C.R. and Short, R.V. (eds), The Embryo. Cambridge University Press, Cambridge, pp. 1–26.

Mottla, G.L., Adelman, M.R., Hall, J.L., Gindoff, P.R., Stillman, R.J. et al. (1995) Lineage tracing demonstrates the blastomeres of early cleavage-stage human pre-embryos contribute to both trophectoderm and inner cell mass. Hum. Reprod., 10, 384–391.[Abstract]

Munné, S., Lee, A., Rosenwaks, Z., Gifro, J. and Cohen, J. (1993) Diagnosis of major aneuploidies in human preimplantation embryos. Hum. Reprod., 8, 2185–2191.[Abstract]

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

Munné, S., Marquez, C., Magli, C., Morton, P. and Morrison, L. (1996) Scoring criteria for preimplantation genetic diagnosis of numerical abnormalities for chromosomes X, Y, 13,16, 18 and 21. Mol. Hum. Reprod., 4, 863–870.

Munné, S., Magli, C., Adler, A., Wright, G., Boer de, K. et al. (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. et al. (1986) Polyploid cells in blastocysts and early fetuses from Australian Merino sheep. J. Reprod. Fertil., 78, 439–446.[Abstract]

Pellestor, F., Dufour, M.C., Arnal, F. and Humeau, C. (1994a) Direct assessment of the rate of chromosomal abnormalities in grade IV human embryos produced by in-vitro fertilization procedure. Hum. Reprod., 9, 293–302.[Abstract]

Pellestor, F., Girardet, A., Andreo, B., Arnal, F. and Humeau, C. (1994b) Relationship between morphology and chromosomal constitution in human preimplantation embryo. Mol. Reprod. Dev., 39, 141–146.[ISI][Medline]

Plachot, M., Grouchy de, J., Junca, A.-M., Mandelbaum, J., Turleau, C. et al. (1987) From oocyte to embryo: a model, deduced from in vitro fertilization, for natural selection against chromosome abnormalities. Ann. Genet., 30, 22–32.[ISI][Medline]

Quinn, P., Kerin, J.F. and Warnes, G.M. (1985) Improved pregnancy rate in human in vitro fertilization with the use of a medium based on the composition of human tubal fluid. Fertil. Steril., 44, 493–489.[ISI][Medline]

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

Sandalinas, M., Sadowy, S., Alikani, M., Calderon, G., Cohen, J. and Munné, S. (2001) Developmental ability of chromosomally abnormal human embryos to develop to the blastocyst stage. Hum. Reprod., 16, 1954–1958.[Abstract/Free Full Text]

Shoukir, Y., Chardonnens, D., Canpana, A. and Sakkas, D. (1998) Blastocyst development from supernumerary embryos after intracytoplasmic sperm injection: a paternal influence? Hum. Reprod., 13, 1632–1637.[Abstract]

Steer, C.V., Mills, C.L., Tan, S.L. et al. (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]

Voullaire, L., Slater, H., Williamson, R. and Wilton, L. (2000) Chromosome analysis of blastomeres from human embryos by using comparative genomic hybridization. Hum. Genet., 106, 210–217.[CrossRef][ISI][Medline]

Wells, D. and Delhanty, J.D.A. (2000) Comprehensive chromosomal analysis of human preimplantation embryos using whole genome amplification and single cell comparative genome hybridization. Mol. Hum. Reprod., 6, 1055–1062.[Abstract/Free Full Text]

West, J.D. (1990) Sexing the human conceptus by in situ hybridisation. In Harris, N. and Wilkinson, D.G. (eds), In-situ Hybridisation: Application to Developmental Biology and Medicine. Cambridge University Press, Cambridge, p. 205.

Wiley, L.M. (1987) Development of the blastocyst: role of cell polarity in cavitation and cell differentiation. In Bavister, B. (ed.), The Mammalian Preimplantation Embryo: Regulation of Growth and Differentiation In Vitro. Plenum Publishing Company, New York, pp. 65–93.

Willard, H.F., Smith, K.D. and Sutherland, J. (1983) Isolation and characterization of a major tandem repeat family from the human X chromosome. Nucleic Acids Res., 11, 2017–2033.[Abstract]

Submitted on September 3, 2002; resubmitted on May 16, 2003; accepted on August 26, 2003.