Abstract

BACKGROUND: Chromosomal mosaicism in human embryos may give rise to false positive or false negative results in preimplantation genetic diagnosis for aneuploidy screening (PGD‐AS). Therefore, we have investigated whether the results obtained from a 2‐cell biopsy of frozen–thawed embryos and fluorescence in situ hybridization (FISH) analysis are representative for the chromosome constitution of the remaining embryo on day 5. METHODS: Cryopreserved day 3 embryos were thawed and from surviving embryos two blastomeres were biopsied. FISH analysis was performed for chromosomes 1, 7, 13, 15, 16, 18, 21, 22, X and Y. After biopsy, the embryos were cultured until day 5 and further analysed using the same probe panels. RESULTS: In all, 17 embryos were available with a diagnosis based on two blastomeres on day 3 and confirmatory studies on day 5. In 10 of these 17 cases the initial diagnosis could be confirmed. However, in only six cases cytogenetic results were concordant. Besides the 10 cases with a ‘correct’ diagnosis, there were six false positive results and one false negative, all involving mosaicism. CONCLUSIONS: Investigating the chromosomal constitution of two blastomere nuclei offers a good opportunity to study the incidence of chromosomal mosaicism in early embryo development. The confirmation rate of the results obtained on day 3 depends on the interpretation and is higher when considered from a clinical than from a cytogenetic point of view.

Introduction

The advent of IVF as a treatment for infertility has created the opportunity to study the chromosomal constitution of surplus human preimplantation embryos. An increasing body of evidence suggests that the incidence of chromosomal abnormalities in embryos is extremely high (as reviewed by Macklon et al., 2002; Wilton, 2002). Accumulating evidence suggests that good embryo morphology does not necessarily exclude an abnormal chromosomal constitution (Magli et al., 2000; Voullaire et al., 2000; Wells and Delhanty, 2000; Sandalinas et al., 2001). Since aneuploidies are considered the main cause of embryonic wastage and loss, this phenomenon may be primarily responsible for the relatively poor pregnancy rates reported after IVF, as well as the poor fertility performance of humans in vivo (Delhanty, 2001). Multiple embryos are usually transferred in IVF in an attempt to overcome low implantation rates per embryo, resulting in considerable multiple pregnancy rates (Fauser, 1999; Hunault, 2002).

Preimplantation genetic diagnosis for aneuploidy screening (PGD‐AS) is proposed as an effective tool in selecting IVF embryos for transfer. Although used in over 24 IVF centres around the world, its beneficial effect remains difficult to assess. A positive effect on implantation and ongoing pregnancy rates in a group of patients with advanced maternal age was observed in retrospective studies (Munné et al., 1999, 2003). Other indications for which PGD‐AS has been proposed include recurrent implantation failure and recurrent miscarriage. However, clinical benefits have not yet been convincingly demonstrated (Gianaroli et al., 1999; ESHRE PGD Consortium Steering Committee, 2002; Pehlivan et al., 2003; Rubio et al., 2003).

An important factor affecting PGD‐AS outcome is the risk of misdiagnosis. Next to technical errors, chromosomal mosaicism within an embryo can be a source of misdiagnosis. Different studies show the percentage of mosaic embryos to be highly variable. The reported range in cleavage stage embryos is 2–90%, depending on maternal age, patient subgroup, method of analysis [karyotyping, fluorescence in situ hybridization (FISH), comparative genomic hybridization (CGH)], number of chromosomes analysed and how mosaicism is defined (Jamieson et al., 1994; Munné et al., 1995; Almeida and Bolton, 1996; Harper and Delhanty, 1996; Bahçe et al., 1999; Gianaroli et al., 1999; Iwarsson et al., 1999; Ruangvutilert et al., 2000; Voullaire et al., 2000; Wells and Delhanty, 2000; Bielanska et al., 2002; Voullaire et al., 2002). For the chromosomes investigated so far, and from CGH studies where all the chromosomes can be analysed, it appears that all chromosomes can be involved in mosaicism. Because of this phenomenon, analysis of a single cell may not represent the chromosomal content of the remaining embryo, leading to false positive or false negative results. It has therefore been proposed to analyse two rather than one blastomere from each embryo (Harper and Delhanty, 1996; Staessen et al., 1997; Iwarsson et al., 1999). Investigating two cells would provide more insight into the incidence of mosaicism of day 3 embryos. However, the impact of a 2‐cell biopsy on developmental potential of the embryo and the added value for achieving a higher rate of ongoing pregnancies is still controversial and needs to be further elucidated.

So far only one comprehensive report on results after PGD‐AS on two biopsied blastomeres has been published. This study considered only four embryos, analysing three chromosomes (Emiliani et al., 2000). One of these four embryos showed discordant results between the two analysed blastomeres and thus was found to be mosaic at day 3. This embryo was reanalysed at day 5 and appeared largely normal. Other studies only reanalysed those embryos that were considered not suitable for transfer due to chromosomal abnormalities diagnosed on a single blastomere (Veiga et al., 1999; Magli et al., 2000; Gianaroli et al., 2001; Sandalinas et al., 2001). Here, high rates of confirmation were described (91, 92, 65–100 and 50% respectively), but very few details were given as to what was considered confirmed. In conclusion, the effect of mosaicism at the 8‐cell stage on the reliability of the diagnosis and further development of the embryo is largely unknown.

We have biopsied frozen–thawed good quality embryos and performed FISH analysis for chromosomes 1, 7, 13, 15, 16, 18, 21, 22, X and Y on two blastomeres. After biopsy, the embryos were cultured until day 5 and subsequently reanalysed using the same probe panels. The current study was undertaken to assess how representative the results obtained from a 2‐cell embryo biopsy on day 3 are for the chromosome constitution of the remaining embryo on day 5, when the embryo is usually transferred in a clinical setting. Moreover, additional information can be obtained regarding the chromosomal content of good quality embryos after cryopreservation.

Materials and methods

Embryos

The spare cryopreserved preimplantation embryos used in this study were obtained from couples undergoing routine IVF procedures in the period between January 1993 and June 1999. The study was approved by the Dutch Central Committee on Research Involving Human Subjects (CCMO) and the local ethics review committee of the hospital. Written consent was obtained from the couples in order to confirm that the embryos could be used for research purposes.

Ovarian stimulation, ovulation induction, oocyte retrieval and IVF/ICSI were performed as described previously (Huisman et al., 2000). A maximum of two embryos was transferred on day 3 after oocyte retrieval according to our previously published protocol (Huisman et al., 2000). Supernumerary, good quality embryos were subsequently cryopreserved. The cryopreservation procedure was performed using a slow‐freezing 1.5 mol/l dimethylsulphoxide (DMSO) protocol in a programmed freezer (De Jong et al., 2002).

For this study, 121 frozen day 3 (4–8 cell) embryos from 25 patients were thawed and taken through consecutive washes of 1.2, 0.9, 0.6 and 0.3 mol/l DMSO for 5 min each. They were transferred to culture medium supplemented with 14.2% GPO (human plasma solution; CLB, The Netherlands) and cultured for ≥2 h. Before the biopsy procedure, the embryos were scored for quality and number of blastomeres. Embryo quality scores were assigned according to previously described criteria (Huisman et al., 2000; Hohmann et al., 2003). Only embryos with five or more blastomeres were biopsied, taking two cells only if the embryo consisted of seven or more blastomeres.

In total, 29 embryos from 12 patients survived the cryopreservation procedure, after storage for an average of 6.4 years (range: 10.2–3.8). These 12 patients (mean maternal age: 33.7 years) all had become pregnant after transfer of two embryos in the cycle where the embryos were frozen.

Biopsy procedure and spreading of blastomeres and embryos

Prior to biopsy, embryos were washed five times in calcium/magnesium‐free medium (EB‐10 medium; Vitrolife, Sweden) and then incubated in EB‐10 medium for 5 min at 37°C. The biopsy was performed on the heated stage of a Nikon IX‐70 microscope, equipped with micromanipulation tools. An infrared diode laser system (OCTAX Laser Shot; OCTAX Microscience GmbH, Germany) with appropriate software (OCTAX EyeWare) was used for partial dissection of the zona pellucida prior to biopsy. After biopsy of one or two blastomeres, the embryos were removed and cultured for another 48 h under normal culture conditions. The remaining blastomeres were fixed according to the method described by Dozortsev and McGinnis (2001). In short, the blastomere was washed once in 1% sodium citrate and transferred onto a slide (Superfrost Plus; Menzel Gläser, Germany). Spreading solution (HCl/Tween 20; 0.01 N HCl/0.1% Tween 20) was added until the cytoplasm was largely removed. The solution was then allowed to dry and a drop of fresh fixative (methanol:acetic acid, 3:1) was added. After drying, the slide was incubated in 100% methanol for 1 min, air‐dried and stored at –20°C overnight. The biopsied embryos were analysed for cleavage stage and quality at day 5 and then spread as described above.

FISH procedure

The DNA probes used in the first round were centromere probes for chromosomes 1 (pUC 1.77; Cooke and Hindley, 1979), 7 (pα7t1; Waye et al., 1987), 15 (pTRA‐20; Choo et al., 1990), X (pBamX5; Willard et al., 1983) and a Y chromosome heterochromatin probe (RPN1305; Lau, 1985). They were labelled by nick translation with Alexa Fluor 594, Alexa Fluor 488, Alexa Fluor 350 or Pacific Blue using an ARES labelling kit (Molecular Probes Europe BV, Netherlands) according to the instructions of the manufacturers. The labelled probes were resuspended in hybridization buffer consisting of 60% formamide and 2×standard saline citrate (SSC; pH 7.0). The DNA probes used in the second round were centromere probes for chromosomes 16 and 18, labelled with Spectrum Blue and Spectrum Aqua, combined with LSI probes for chromosomes 13, 21 and 22 labelled with Spectrum Red, Green and Gold respectively (Multivysion PB kit; Vysis, USA).

Two rounds of five‐colour FISH were applied to each blastomere or embryo. In the first round, FISH was performed for chromosomes 1 (aqua), 7 (blue), 15 (red), X (yellow) and Y (green). A yellow signal was obtained by mixing equal volumes of green and red fluorescence probes. A hybridization mixture was made, containing 1.0 ng of each labelled probe in 50% formamide, 10% dextran sulphate, 1% Tween 20 and 2×SSC in a final volume of 10 µl. Per slide, 0.5 µl hybridization mixture was applied. Blastomere and probe DNA were denatured simultaneously at 75°C for 3 min and allowed to hybridize in a humid box (containing 50% formamide in 2×SSC, pH 7.2) at 37°C for 2 h. After hybridization, the coverslips were removed and slides were washed in 2×SSC/0.05% Tween 20 for 2 min at 42°C, then in 0.1×SSC for 6 min at 60°C and in 2×SSC/0.05% Tween 20 for 2 min at room temperature. The slides were allowed to air‐dry and mounted in Vectashield antifade medium (Vector Laboratories, USA), or in later experiments in Vectashield containing 1.25 ng/ml 4′,6‐diamidino‐2‐phenylindole (DAPI). Slides were examined with a Zeiss Axioplan 2 imaging epifluorescence‐equipped microscope, using appropriate filters. Images were captured with the Powergene™ Macprobe system (Applied Imaging International Ltd, UK), or in later experiments with the ISIS FISH Imaging System (MetaSystems, Germany). After the first hybridization round, antifade medium was removed with two washes in 2×SSC/0.05% Tween 20 for 5 min at room temperature. The slides were dehydrated and the probe mix for the second round was applied. The slides were denatured for 6 min at 69°C and allowed to hybridize overnight. Slides were washed in 2×SSC/0.05% Tween 20 for 2 min at 42°C, then in 0.6×SSC/0.05% Tween 20 for 6 min at 60°C and in 2×SSC/0.05% Tween 20 for 2 min at room temperature. The slides were allowed to air‐dry and mounted in Vectashield antifade medium. Signals from the second round were recorded and compared with those from the first round to ensure that they had not persisted. For both rounds, we used the scoring criteria previously published (Munné et al., 1998). The efficiency of the FISH probes was tested on uncultured peripheral lymphocyte spreads from two men and two women with normal karyotypes. Slides were prepared according to standard protocols and they were hybridized using the same protocol as for the embryonic cells. Signals were counted in 200 nuclei. In addition, the positions of 16 individual nuclei were recorded and images were obtained after each round to check for persisting signals.

Interpretation of FISH signals and definitions

Based on the analysis of two blastomeres per embryo, we classed day 3 embryos as normal (both nuclei showing the normal amount of signals for the chromosomes investigated), mosaic (one normal nucleus and one abnormal or each nucleus showing a different abnormality) or aneuploid (both nuclei carrying the same abnormality). After analysis of all the cells from each embryo on day 5, we chose the following definitions on the basis of the results obtained. We classed embryos as normal (≥80% normal nuclei and <10% of the nuclei with the same chromosome abnormality), aneuploid (≥90% of nuclei showing the same abnormality) or mosaic (between 10 and 90% of the cells showing the same chromosome abnormality). Embryos with ≥90% haploid, tetraploid or triploid nuclei were classed as such. However, we considered the occurrence of some tetraploid cells as a normal phenomenon of in vitro‐cultured embryos (Evsikov and Verlinsky, 1998; Bielanska et al., 2002) and treated them as normal cells.

A diagnosis of normal, aneuploid or mosaic on day 3 was considered to be confirmed on day 5 if normal, aneuploid or mosaic results were recovered on day 5 respectively. The diagnosis was considered cytogenetically confirmed, if the same chromosomal abnormalities seen on day 3 were observed on day 5. Furthermore, if aneuploidy was diagnosed on day 3 and a substantial proportion (∼80%) of the cells on day 5 showed the same chromosomal abnormality, the diagnosis was also considered cytogenetically confirmed as well.

Results

Control lymphocytes

The probes were tested for hybridization efficiency on lymphocyte spreads from two males and two females in 200 nuclei. The hybridization efficiencies for the individual probes were calculated as the percentage of nuclei showing the expected number of signals and ranged between 93 and 100%. Furthermore, 84% of the nuclei showed the normal amount of signals for all five probes in the first round and 90% in the second round. No persisting signals from the first round were observed in the second round.

Embryo biopsy and culture

After thawing, embryos were scored (average morphology score: 2.4) and 29 embryos were considered suitable for biopsy, based on the number of blastomeres. Only two embryos showed <10% fragmentation. From 21 embryos, two blastomeres could be biopsied and spread (Tables I and II). From five embryos, one of the biopsied blastomeres was lost during the spreading procedure and a further three embryos consisted of less than seven cells, so in these embryos only one cell was available for FISH analysis (Table III). After the procedure, biopsied embryos were returned to culture. On day 5, embryos were scored for cleavage stage and morphology (Tables I–III): 14 embryos had developed to the blastocyst stage (48%; average morphology score: 2.1), three embryos (10%) were still at the 6–10‐cell stage (arrested day 3), five embryos (14%) had arrested at compaction (arrested day 4) and seven embryos (24%) had degenerated to such an extent that no intact blastomeres were available for spreading and they were not analysed further.

FISH analysis of blastomeres and embryos

The FISH results on day 3 and day 5 and their interpretation are shown in Tables I–III. From the 21 cases where two blastomeres were available for analysis, nine embryos showed concordant FISH results on day 3 (Table I). From the five (24%) embryos diagnosed as normal, three of these could be confirmed after reanalysis on day 5. One embryo had degenerated, and the other embryo showed a false negative result: it was mosaic for monosomy 16. From the four embryos diagnosed as aneuploid or polyploid, the triploid embryo (case 9) had degenerated, but we could confirm the other cases. Case 6 (trisomy 1) developed into a blastocyst with the same abnormality in all but one cell (Table I; Figure 1). In case 7, monosomy 7 was found in almost 80% of the cells. Although the resulting embryo was classified as mosaic, we considered the abnormality to be cytogenetically confirmed. For case 8, monosomy 18 was recovered in all but one cell of an otherwise chaotic embryo. The monosomy 18 clearly had a meiotic origin, with the other chromosomal abnormalities resulting from mitotic events.

The other 12 embryos (57%) showed different results for the two blastomeres and were considered mosaic (Table II). Most mosaic embryos (cases 10–17) showed one normal and one abnormal blastomere. The other mosaic embryos (cases 18–21) showed two different abnormal blastomeres. One biopsied blastomere was binucleated, with both nuclei showing different abnormalities (case 18). On day 5, four of these 12 cases could be confirmed as mosaic. However, in two cases, totally different chromosome abnormalities were involved (cases 12 and 19). In case 11, monosomy X was confirmed, but the embryo also showed mosaicism for monosomy 16, which was not detected on day 3. In case 17, mosaic monosomy 1 was confirmed on day 5, but monosomy 15 was not recovered. Two embryos had degenerated (cases 18 and 21), and the other six developed into chromosomally normal embryos.

In all, 17 embryos were available with a diagnosis based on two blastomeres on day 3 and confirmatory studies on day 5. In 10 of these 17 cases the initial diagnosis of normal (n = 3), aneuploid (n = 3) or mosaic (n = 4) chromosome constitution could be confirmed on day 5. However, in only six of these 10 cases (three normal, one monosomy 7, one monosomy 18 and one trisomy 1) were cytogenetic results considered concordant. Besides these 10 cases with a ‘correct’ diagnosis, there were six false positive results and one false negative, all involving mosaicism.

From the embryos where only one blastomere was available for FISH analysis, two were diagnosed as normal and six as abnormal on day 3 (Table III). Both normal embryos were mosaic on day 5. Case 22 showed non‐disjunction of chromosome 18 in about half of the cells, which was not detected on day 3. Case 23 showed non‐disjunction of chromosome 16 in a low percentage of cells. Three abnormal embryos had degenerated on day 5, and after reanalysis of the other three the initial diagnosis could be confirmed only in case 24, which showed substantial mosaicism for trisomy 18. So, from the five cases where confirmatory studies were available on day 5, only one was confirmed.

Discussion

In the current study, blastomere biopsy was performed on 29 cryopreserved embryos donated for research followed by reanalysis of the remaining embryo on day 5. This offers the unique opportunity to study their chromosomal constitution on day 3 and day 5 and to investigate the developmental capacity of embryos after biopsy of two blastomeres. A good proportion of embryos (14 out of 26, 54%) developed to the blastocyst stage after biopsy of two blastomeres. This rate is similar to the blastocyst rate on day 5 of non‐biopsied good quality embryos in our laboratory, as reported previously (54%; Huisman et al., 2000). Although larger numbers are needed to confirm these observations, biopsy of two blastomeres does not seem detrimental to the development of the embryo.

Our data confirm previous reports on the lack of correlation between the presence of blastomeres with a chromosome abnormality on day 3 and the likelihood of development to the blastocyst stage (Magli et al., 2001; Sandalinas et al., 2001; Rubio et al., 2003). Extended culture is no reliable means to select chromosomally normal embryos. This is well illustrated with case 6. This embryo, although morphologically scored as excellent, appeared to be uniformly trisomic for chromosome 1 (Figure 1). PGD‐AS may therefore be a useful tool in selecting embryos with the best chances for implantation and further development. The efficiency, however, depends largely on the predictive value of the results obtained on day 3.

Reanalysis of blastocysts can be a useful tool to investigate how representative the day 3 results are for the remaining embryo (Gianaroli et al., 1999; Veiga et al., 1999; Emiliani et al., 2000; Magli et al., 2000). However, if a confirmation rate should be established, it is useful to define what can be considered confirmed. From a cytogenetic point of view, confirmation would entail that the chromosome constitution of the two investigated blastomeres is exactly reflected in the cell line(s) found at the blastocyst stage. In the group of embryos with two analysed blastomeres, we found this to be the case in six out of 17 reanalysed embryos. After analysis of one blastomere, only one out of five gave the same results. Hence, cytogenetic confirmation was obtained in only 32% (7/22).

There are several explanations for this high rate of discordance. Misdiagnosis can occur because of technical problems related to the FISH procedure, especially when using more probes simultaneously in successive rounds of FISH. The most extensive FISH screen reported so far, included probes for nine different chromosomes in two rounds of hybridization (Munné at al., 1998). We used probes for 10 different chromosomes in two rounds of hybridization and found an overall rate of 84% of lymphocytes to display the normal amount of signals. Although this shows that the probe panels we used can be considered efficient for simultaneous enumeration of the selected chromosomes, it has to be kept in mind that results from lymphocyte nuclei cannot be directly compared with those from blastomere nuclei. Lymphocyte nuclei are subjected to extensive fixation in methanol/acetic acid and are smaller and more compact in appearance. This can increase the likelihood of loss of signal or signal overlap. Furthermore, since most of the probes used are alpha‐satellite probes, signal size can vary due to individual variability in size of the heterochromatic region. To detect this, lymphocyte nuclei from both parents should be used as controls. In these experiments, however, that was not possible. So, although FISH artefact may offer an explanation for the cytogenetic discordance, especially since most of the abnormalities encountered involved loss of signal, there may be other explanations.

First, the abnormal cell(s) could have been removed by the biopsy procedure, thus leaving only normal cells or cells with a different chromosome abnormality. Second, from the 29 biopsied embryos, 14 embryos showed >10% fragmentation and 13 embryos >20% at the time of biopsy. Though the origin and the mechanisms of this fragmentation process are still unclear, a strong association was established between percentage of fragmentation and chromosome abnormalities (Magli et al., 2001). It has also been shown that fragments can contain chromatin (Jurisicova et al., 2003), and, although highly speculative, this process may provide an explanation for the high incidence of chromosomal mosaicism involving chromosome loss we encountered at the 8‐cell stage.

It has been shown that significant evidence of apoptosis is not observed until the morula stage in human embryos (Jurisicova et al., 1996; Hardy, 1997, 1999; Hardy et al., 2001). This process may be responsible for the elimination of cells carrying a chromosome abnormality making the cell less viable, such as monosomies. So it would be likely for a large part of the aberrant cells to have disappeared at the blastocyst stage, thus confounding the confirmation rate. A previous study performing reanalysis of blastocysts after PGD, indeed reported that correspondence to the day 3 diagnosis was lowest for monosomies and haploidies (Magli et al., 2000). Furthermore, other studies showed that a lower percentage of embryos diagnosed with a monosomy developed to the blastocyst stage compared with normal embryos, or embryos with a trisomy (Sandalinas et al., 2001; Rubio et al., 2003).

A high incidence of mosaicism on day 3 (57%) and day 5 (50%) was observed in the current group of cryopreserved embryos. Others described a similar high incidence of mosaicism in cryopreserved embryos after analysis of entire day 3 embryos (Iwarsson et al., 1999). The possibility that chromosome abnormalities may actually be induced by the cryopreservation procedure cannot be excluded. This may explain the poor added value of cryopreservation in IVF programmes, as published recently (De Jong et al., 2002). However, it should be considered that the embryos in the present study were selected in two ways. First, these embryos have survived the cryopreservation procedure, where 76% of the embryos degenerated. Second, although of good morphology, all embryos were surplus in an IVF cycle where the woman became pregnant from the fresh transfer of embryos. It is feasible that those embryos with a normal chromosomal constitution and thus the best chances for implantation were by chance selected in this cycle, leaving the embryos with chromosome abnormalities. Therefore, a direct comparison between fresh and cryopreserved embryos is needed to see if these abnormalities result from the cryopreservation procedure or are increased due to selection of a subgroup of embryos.

In the interpretation of the results, we considered the presence of some tetraploid cells to be a normal feature in preimplantation embryos as also seen in other mammalian species (Benkhalifa et al., 1993). A marked increase in incidence of polyploid cells from day 4 to day 6 of development was described by others, suggesting that most polyploid cells arise during blastocyst formation and that they are a hallmark of trophoblast differentiation (Bielanska et al., 2002). Indeed, in most of our cases with tetraploid cells, only a few cells were involved. Tetraploid cells can even undergo correction to normal diploid cells (Staessen et al., 1998). However, it cannot be excluded that a high percentage of tetraploid cells, as in case 16 (Table II) which arose during one of the first four post‐zygotic cell divisions, would be detrimental. Although it has been found that tetraploid cells are almost exclusively allocated to the trophoblast compartment (Angel et al., 1987; Benkhalifa et al., 1993), resulting in confined placental mosaicism as detected by prenatal diagnosis in chorionic villi (Noomen et al., 2001), distribution to the inner cell mass may result in a non‐viable (mosaic) tetraploid fetus.

If we consider the results obtained in the present study from a clinical viewpoint, confirmation would be that an embryo diagnosed on day 3 as normal or abnormal is found to be the same after analysis on day 5. After analysis of only one blastomere, this was the case for only one embryo out of five. After analysis of two blastomeres, the initial diagnosis was found to be most reliable if the results from both blastomeres were concordant (six out of seven). In the case of discordant results, confirming the diagnosis was more complicated. Although initially we considered mosaic 8‐cell embryos as abnormal, these data show that three out of 12 mosaic embryos became normal and morphologically good quality embryos. A recent study also showed embryos diagnosed on day 3 as mosaic to have the same developmental potential as normal embryos, but no reanalysis was performed (Rubio et al., 2003). More data are needed on reanalysis of mosaic embryos, since it may help in understanding the fate of specific chromosome abnormalities during embryo development. This may eventually help to determine if some mosaic embryos can be considered safe for transfer.

In conclusion, investigation of two blastomeres allows for a better differentiation between uniformly diploid, uniformly aneuploid and mosaic embryos. We show here that the chromosome constitution of mosaic embryos is subject to changes during further development of the embryo. Thus, the analysis of two blastomeres on day 3 yields a better prediction of the chromosome constitution on day 5 and the developmental potential of the embryo.

Acknowledgements

We would like to thank the patients, laboratory and clinical staff of the IVF department at the Erasmus MC for participating and/or assisting in this study. This research was financially supported by Erasmus University (AIO) and ‘Stichting Voortplantingsgeneeskunde Rotterdam’.

Figure 1. (A) Blastomere nucleus after the first round of FISH, showing the normal amount of signals for chromosome 1 (aqua), 7 (blue), 15 (red), X (yellow) and Y (green). DNA is counterstained with DAPI. (B) Same nucleus after the second round of FISH, showing the normal amount of signals for chromosome 13 (red), 16 (aqua), 18 (blue), 21 (green) and 22 (gold). (C) Light micrograph showing good quality day 3 embryo before biopsy. (D and E) Nuclei of two blastomeres from this embryo after biopsy and the first round of FISH, showing three signals for chromosome 1 (aqua, small arrowheads). (F) Light micrograph showing the same embryo on day 5. (G) Two representative nuclei of this embryo on day 5 after the first round of FISH. All nuclei showed an extra signal for chromosome 1 (aqua). DNA is counterstained with DAPI.

Figure 1. (A) Blastomere nucleus after the first round of FISH, showing the normal amount of signals for chromosome 1 (aqua), 7 (blue), 15 (red), X (yellow) and Y (green). DNA is counterstained with DAPI. (B) Same nucleus after the second round of FISH, showing the normal amount of signals for chromosome 13 (red), 16 (aqua), 18 (blue), 21 (green) and 22 (gold). (C) Light micrograph showing good quality day 3 embryo before biopsy. (D and E) Nuclei of two blastomeres from this embryo after biopsy and the first round of FISH, showing three signals for chromosome 1 (aqua, small arrowheads). (F) Light micrograph showing the same embryo on day 5. (G) Two representative nuclei of this embryo on day 5 after the first round of FISH. All nuclei showed an extra signal for chromosome 1 (aqua). DNA is counterstained with DAPI.

Table I.

Embryo development, FISH results and interpretation from embryos with concordant results after analysis of two blastomeres on day 3

Case no. Day 3 Day 5 Interpretation of FISH results 
 Morphology scorea/no. of cells FISH results Diagnosis Morphology score/cleavage stage No. of cells analysed FISH results  
2/8 2N/2N Normal 2/hatching blastocyst 48 2N [45] Normalb 
      N [2]  
      4N [1]  
2/12 2N/2N Normal 2/hatching blastocyst 59 2N [32] Mosaic  
      –16 [18]  
      –18 [5]  
      –X [1]  
      –X, –1 [1]  
      –7 [1]  
      –15 [1]  
2/8 2N/2N Normal 1/blastocyst 30 2N [24] Normalb 
      –X [2]  
      –7 [1]  
      –13 [1]  
      –18 [1]  
      –22 [1]  
2/12 2N/2N Normal 3/blastocyst 35 2N [25] Normalb 
      4N [3]  
      –13 [2]  
      –18 [2]  
      –1 [1]  
      +15 [1]  
      –1, –21, –18 [1]  
3/8 2N/2N Normal 5/degenerated –  
1/8 +1/+1 Aneuploid 1/hatching blastocyst 49 +1 [44] Aneuploidb 
      +1, –18 [3]  
      +1, +18 [1]  
      4N [1]  
3/10 –7/–7 Aneuploid  3/blastocyst 13 –7 [10] Mosaicb 
      2N [1]  
      4N [2]  
2/8 –18/–18 Aneuploid 5/arrested day 3 +7, –16, –18 [1] Aneuploid/chaoticb 
      –X, –1, +7, +13 [1]  
      +Y, +1, +15, –16, –18 [1]  
      N [1]  
      –1, –7, –18, –21 [1]  
      –16, –18, +21, +22 [1]  
3/8 3N/3N Triploid 5/degenerated –  
Case no. Day 3 Day 5 Interpretation of FISH results 
 Morphology scorea/no. of cells FISH results Diagnosis Morphology score/cleavage stage No. of cells analysed FISH results  
2/8 2N/2N Normal 2/hatching blastocyst 48 2N [45] Normalb 
      N [2]  
      4N [1]  
2/12 2N/2N Normal 2/hatching blastocyst 59 2N [32] Mosaic  
      –16 [18]  
      –18 [5]  
      –X [1]  
      –X, –1 [1]  
      –7 [1]  
      –15 [1]  
2/8 2N/2N Normal 1/blastocyst 30 2N [24] Normalb 
      –X [2]  
      –7 [1]  
      –13 [1]  
      –18 [1]  
      –22 [1]  
2/12 2N/2N Normal 3/blastocyst 35 2N [25] Normalb 
      4N [3]  
      –13 [2]  
      –18 [2]  
      –1 [1]  
      +15 [1]  
      –1, –21, –18 [1]  
3/8 2N/2N Normal 5/degenerated –  
1/8 +1/+1 Aneuploid 1/hatching blastocyst 49 +1 [44] Aneuploidb 
      +1, –18 [3]  
      +1, +18 [1]  
      4N [1]  
3/10 –7/–7 Aneuploid  3/blastocyst 13 –7 [10] Mosaicb 
      2N [1]  
      4N [2]  
2/8 –18/–18 Aneuploid 5/arrested day 3 +7, –16, –18 [1] Aneuploid/chaoticb 
      –X, –1, +7, +13 [1]  
      +Y, +1, +15, –16, –18 [1]  
      N [1]  
      –1, –7, –18, –21 [1]  
      –16, –18, +21, +22 [1]  
3/8 3N/3N Triploid 5/degenerated –  

aMorphology score: 1 = excellent; 2 = good; 3 = average; 4 = poor quality.

bCases with cytogenetic confirmation.

Table II.

Embryo development, FISH results and interpretation from embryos with discordant results after analysis of two blastomeres on day 3

Case no. Day 3 Day 5 Interpretation of FISH results 
 Morphology score/no. of cells FISH results Diagnosis Morphology score/cleavage stage No. of cells analysed FISH results  
10 2/8 2N/+13 Mosaic  1/hatching blastocyst 33 2N [27] Normal 
      4N [1]  
      –X [1]  
      +1 [1]  
      –16 [1]  
      –18 [1]  
      –16, –21 [1]  
11 2/8 2N/–X Mosaic  3/blastocyst 26 2N [8] Mosaic 
      –16 [10]  
      –X [6]  
      –15 [1]  
      –16, +22 [1]  
12 2/8 2N/–1, –16 Mosaic  4/blastocyst 27 2N [14] Mosaic 
      –13 [4]  
      –21 [3]  
      +13, –1 [2]  
      –13, –21 [1]  
      +13 [1]  
      –22 [1]  
      +13, –22 [1]  
13 2/12 2N/–21 Mosaic  1/hatching blastocyst 52 2 N [40] Normal 
      4N [6]  
      –18 [4]  
      +18 [1]  
      –X, –7 [1]  
14 2/10 2N/–21 Mosaic  1/hatching blastocyst 175 2N [155] Normal 
      –18 [11]  
      –15 [7]  
      –16 [2]  
15 2/8 2N/+13, –16 Mosaic  4/arrested day 4 12 2N [9] Normal 
      4N [2]  
      –15 [1]  
16 3/8 2N/–16 Mosaic  5/arrested day 4 14 2N [8] Normal 
      4N [6]  
        
17 3/7 2N/–1, –15 Mosaic  5/arrested day 4 2N [2] Mosaic 
      4N [1]  
      –1 [5]  
      +1 [1]  
18 3/10 2N/–21/–X, +16  Mosaic  5/degenerated –  
19 3/8 4N/ Mosaic 3/blastocyst 20 2N [12] Mosaic 
  +Y, –7, –15, –22    –13 [5]  
      –18 [3]  
20 2/10 3N/+13, –16 Mosaic 5/arrested day 4 14 2N [9] Normal 
      N [1]  
      4N [3]  
      –13, –21 [1]  
21 2/7 –1, –16, +18/ Mosaic 5/degenerated –  
  –1, –7, –18, –22      
Case no. Day 3 Day 5 Interpretation of FISH results 
 Morphology score/no. of cells FISH results Diagnosis Morphology score/cleavage stage No. of cells analysed FISH results  
10 2/8 2N/+13 Mosaic  1/hatching blastocyst 33 2N [27] Normal 
      4N [1]  
      –X [1]  
      +1 [1]  
      –16 [1]  
      –18 [1]  
      –16, –21 [1]  
11 2/8 2N/–X Mosaic  3/blastocyst 26 2N [8] Mosaic 
      –16 [10]  
      –X [6]  
      –15 [1]  
      –16, +22 [1]  
12 2/8 2N/–1, –16 Mosaic  4/blastocyst 27 2N [14] Mosaic 
      –13 [4]  
      –21 [3]  
      +13, –1 [2]  
      –13, –21 [1]  
      +13 [1]  
      –22 [1]  
      +13, –22 [1]  
13 2/12 2N/–21 Mosaic  1/hatching blastocyst 52 2 N [40] Normal 
      4N [6]  
      –18 [4]  
      +18 [1]  
      –X, –7 [1]  
14 2/10 2N/–21 Mosaic  1/hatching blastocyst 175 2N [155] Normal 
      –18 [11]  
      –15 [7]  
      –16 [2]  
15 2/8 2N/+13, –16 Mosaic  4/arrested day 4 12 2N [9] Normal 
      4N [2]  
      –15 [1]  
16 3/8 2N/–16 Mosaic  5/arrested day 4 14 2N [8] Normal 
      4N [6]  
        
17 3/7 2N/–1, –15 Mosaic  5/arrested day 4 2N [2] Mosaic 
      4N [1]  
      –1 [5]  
      +1 [1]  
18 3/10 2N/–21/–X, +16  Mosaic  5/degenerated –  
19 3/8 4N/ Mosaic 3/blastocyst 20 2N [12] Mosaic 
  +Y, –7, –15, –22    –13 [5]  
      –18 [3]  
20 2/10 3N/+13, –16 Mosaic 5/arrested day 4 14 2N [9] Normal 
      N [1]  
      4N [3]  
      –13, –21 [1]  
21 2/7 –1, –16, +18/ Mosaic 5/degenerated –  
  –1, –7, –18, –22      
Table III.

Embryo development, FISH results and interpretation from embryos after analysis of one blastomere on day 3

Case No. Day 3 Day 5 Interpretation of FISH results 
 Morphology score/no. of cells FISH results Diagnosis Morphology score/cleavage stage No. of cells analysed FISH results  
22 1/8 2N Normal 1/hatching blastocyst 98 2N [54] Mosaic 
      –18 [32]  
      +18 [11]  
      +X [1]  
23 2/10 2N Normal 5/arrested day 3 17 2N [11] Mosaic 
      +16 [2]  
      –16 [2]  
      –15 [1]  
      –X [1]  
24 3/12 +18 Aneuploid 4/blastocyst 31 2N [5] Mosaic 
      +18 [22]  
      +15[1]  
      –1, +18 [1]  
      +1, +X, +18 [1]  
      –1, +15, +18 [1]  
25 3/8 –18 Aneuploid 5/arrested day 3 2N [2] Mosaic 
      –7 [1]  
26 3/7 –X, –16 Aneuploid 5/arrested day 4 2N [9] Normal 
27 3/5 3N Triploid 5/degenerated –  
28 3/6 –7, –16 Aneuploid 5/degenerated –  
29 4/5 –X Aneuploid 5/degenerated –  
Case No. Day 3 Day 5 Interpretation of FISH results 
 Morphology score/no. of cells FISH results Diagnosis Morphology score/cleavage stage No. of cells analysed FISH results  
22 1/8 2N Normal 1/hatching blastocyst 98 2N [54] Mosaic 
      –18 [32]  
      +18 [11]  
      +X [1]  
23 2/10 2N Normal 5/arrested day 3 17 2N [11] Mosaic 
      +16 [2]  
      –16 [2]  
      –15 [1]  
      –X [1]  
24 3/12 +18 Aneuploid 4/blastocyst 31 2N [5] Mosaic 
      +18 [22]  
      +15[1]  
      –1, +18 [1]  
      +1, +X, +18 [1]  
      –1, +15, +18 [1]  
25 3/8 –18 Aneuploid 5/arrested day 3 2N [2] Mosaic 
      –7 [1]  
26 3/7 –X, –16 Aneuploid 5/arrested day 4 2N [9] Normal 
27 3/5 3N Triploid 5/degenerated –  
28 3/6 –7, –16 Aneuploid 5/degenerated –  
29 4/5 –X Aneuploid 5/degenerated –  

References

Almeida PA and Bolton VN (
1996
) The relationship between chromosomal abnormality in the human preimplantation embryo and development in vitro.
Reprod Fertil Dev
 
8
,
235
–241.
Angel RR, Sumner AT, West JD, Thatcher SS, Glasier AF and Baird DT (
1987
) Post‐fertilization polyploidy in human preimplantation embryos fertilized in vitro.
Hum Reprod
 
2
,
721
–727.
Bahçe 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.
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.
Bielanska M, Tan SL and Ao A (
2002
) Chromosomal mosaicism throughout human preimplantation development in vitro: incidence, type and relevance to embryo outcome.
Hum Reprod
 
17
,
413
–419.
Choo KH, Earle E, Vissel B and Filby RG (
1990
) Identification of two distinct subfamilies of alpha satellite DNA that are highly specific for human chromosome 15.
Genomics
 
7
,
143
–151.
Cooke HJ and Hindley J (
1979
) Cloning of human satellite III DNA: different components are on different chromosomes.
Nucleic Acids Res
 
6
,
3177
–3197.
DeJong D, Eijkemans MJ, Beckers NG, Pruijsten RV, Fauser BC and Macklon NS (
2002
) The added value of embryo cryopreservation to cumulative ongoing pregnancy rates per IVF treatment: is cryopreservation worth the effort?
J Assist Reprod Genet
 
19
,
561
–568.
Delhanty JD (
2001
) Preimplantation genetics: an explanation for poor human fertility?
Ann Hum Genet
 
65
,
331
–338.
Dozortsev DI and McGinnis KT (
2001
) An improved fixation technique for fluorescence in situ hybridization for preimplantation genetic diagnosis.
Fertil Steril
 
76
,
186
–188.
Emiliani S, Merino EG, Van den BM, Abramowicz M, Vassart G, Englert Y and Delneste D (
2000
) Re‐analysis by fluorescence in situ hybridisation of spare embryos cultured until day 5 after preimplantation genetic diagnosis for a 47, XYY infertile patient demonstrates a high incidence of diploid mosaic embryos: a case report.
Prenat Diagn
 
20
,
1063
–1066.
ESHRE PGD Consortium Steering Committee (
2002
) ESHRE Preimplantation Genetic Diagnosis Consortium: data collection III (May 2001).
Hum Reprod
 
17
,
233
.
Evsikov S and Verlinsky Y (
1998
) Mosaicism in the inner cell mass of human blastocysts.
Hum Reprod
 
13
,
3151
–3155.
Fauser BC, Devroey P, Yen SS, Gosden R, Crowley WF Jr, Baird DT and Bouchard P (
1999
) Minimal ovarian stimulation for IVF: appraisal of potential benefits and drawbacks.
Hum Reprod
 
14
,
2681
–2686.
Gianaroli L, Magli MC, Ferraretti AP and Munné S (
1999
) Preimplantation diagnosis for aneuploidies in patients undergoing in vitro fertilization with a poor prognosis: identification of the categories for which it should be proposed.
Fertil Steril
 
72
,
837
–844.
Gianaroli L, Magli MC and Ferraretti AP (
2001
) The in vivo and in vitro efficiency and efficacy of PGD for aneuploidy.
Mol Cell Endocrinol
 
183
(Suppl 1),
S13
–S18.
Hardy K (
1997
) Cell death in the mammalian blastocyst.
Mol Hum Reprod
 
3
,
919
–925.
Hardy K (
1999
) Apoptosis in the human embryo.
Rev Reprod
 
4
,
125
–134.
Hardy K, Spanos S, Becker D, Iannelli P, Winston RML and Stark J (
2001
) From cell death to embryo arrest: Mathematical models of human preimplantation embryo development.
Proc Natl Acad Sci USA
 
98
,
1655
.
Harper JC and Delhanty JD (
1996
) Detection of chromosomal abnormalities in human preimplantation embryos using FISH.
J Assist Reprod Genet
 
13
,
137
–139.
Hohmann FP, Macklon NS and Fauser BC (
2003
) A randomized comparison of two ovarian stimulation protocols with gonadotropin‐releasing hormone (GnRH) antagonist cotreatment for in vitro fertilization commencing recombinant follicle‐stimulating hormone on cycle day 2 or 5 with the standard long GnRH agonist protocol.
J Clin Endocrinol Metab
 
88
,
166
–173.
Huisman GJ, Fauser BC, Eijkemans MJ and Pieters MH (
2000
) Implantation rates after in vitro fertilization and transfer of a maximum of two embryos that have undergone three to five days of culture.
Fertil Steril
 
73
,
117
–122.
Hunault CC, Eijkemans MJ, Pieters MH, te Velde ER, Habbema JD, Fauser BC and Macklon NS (
2002
) A prediction model for selecting patients undergoing in vitro fertilization for elective single embryo transfer.
Fertil Steril
 
77
,
725
–732.
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.
Jamieson ME, Coutts JR and Connor JM (
1994
) The chromosome constitution of human preimplantation embryos fertilized in vitro.
Hum Reprod
 
9
,
709
–715.
Jurisicova A, Varmuza S and Casper RF (
1996
) Programmed cell death and human embryo fragmentation.
Mol Hum Reprod
 
2
,
93
–98.
Jurisicova A, Antenos M, Varmuza S, Tilly JL and Casper RF (
2003
) Expression of apoptosis‐related genes during human preimplantation embryo development: potential roles for the Harakiri gene product and Caspase‐3 in blastomere fragmentation.
Mol Hum Reprod
 
9
,
133
–141.
Lau YF (
1985
) Detection of Y‐specific repeat sequences in normal and variant human chromosomes using in situ hybridization with biotinylated probes.
Cytogenet Cell Genet
 
39
,
184
–187.
Macklon NS, Geraedts JP and Fauser BC (
2002
) Conception to ongoing pregnancy: the ‘black box’ of early pregnancy loss.
Hum Reprod Update
 
8
,
333
–343.
Magli MC, Jones GM, Gras L, Gianaroli L, Korman I and Trounson AO (
2000
) Chromosome mosaicism in day 3 aneuploid embryos that develop to morphologically normal blastocysts in vitro.
Hum Reprod
 
15
,
1781
–1786.
Magli MC, Gianaroli L and Ferraretti AP (
2001
) Chromosomal abnormalities in embryos.
Mol Cell Endocrinol
 
183
(Suppl 1),
S29
–S34.
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.
Munné S, Magli C, Bahçe M, Fung J, Legator M, Morrison L, Cohert J and Gianaroli L (
1998
) Preimplantation diagnosis of the aneuploidies most commonly found in spontaneous abortions and live births: XY, 13, 14, 15, 16, 18, 21, 22.
Prenat Diagn
 
18
,
1459
–1466.
Munné S, Magli C, Cohen J, Morton P, Sadowy S, Gianaroli L, Tucker M, Marquez C, Sable D, Ferraretti AP et al (
1999
) Positive outcome after preimplantation diagnosis of aneuploidy in human embryos.
Hum Reprod
 
14
,
2191
–2199.
Munné S, Sandalinas M, Escudero T, Velilla E, Walmsley R, Sadowy S, Cohen J and Sable D (
2003
) Improved implantation after preimplantation genetic diagnosis of aneuploidy.
Reprod Biomed Online
 
7
,
91
–97.
Noomen P, van den BC, de Ruyter JL, Van Opstal D and Los FJ (
2001
) Prevalence of tetraploid metaphases in semidirect and cultured chorionic villi.
Fetal Diagn Ther
 
16
,
129
–132.
Pehlivan T, Rubio C, Rodrigo L, Romero J, Remohí J, Simón C and Pellicer A (
2003
) Impact of preimplantation genetic diagnosis on IVF outcome in implantation failure patients.
Reprod Biomed Online
 
6
,
232
–237.
Ruangvutilert P, Delhanty JD, Serhal P, Simopoulou M, Rodeck CH and Harper JC (
2000
) FISH analysis on day 5 post‐insemination of human arrested and blastocyst stage embryos.
Prenat Diagn
 
20
,
552
–560.
Rubio C, Simón C, Vidal F, Rodrigo L, Pehlivan T, Remohí J and Pellicer A (
2003
) Chromosomal abnormalities and embryo development in recurrent miscarriage couples.
Hum Reprod
 
18
,
182
–188.
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.
Staessen C and Van Steirteghem AC (
1997
) The chromosomal constitution of embryos developing from abnormally fertilized oocytes after intracytoplasmic sperm injection and conventional in‐vitro fertilization.
Hum Reprod
 
12
,
321
–327.
Staessen C and Van Steirteghem A (
1998
) The genetic constitution of multinuclear blastomeres and their derivative daughter blastomeres.
Hum Reprod
 
13
,
1625
–1631.
Veiga A, Gil Y, Boada M, Carrera M, Vidal F, Boiso I, Ménézo Y and Barri PN (
1999
) Confirmation of diagnosis in preimplantation genetic diagnosis (PGD) through blastocyst culture: preliminary experience.
Prenat Diagn
 
19
,
1242
–1247.
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.
Voullaire L, Wilton L, McBain J, Callaghan T and Williamson R (
2002
) Chromosome abnormalities identified by comparative genomic hybridization in embryos from women with repeated implantation failure.
Mol Hum Reprod
 
8
,
1035
–1041.
Waye JS, England SB and Willard HF (
1987
) Genomic organization of alpha satellite DNA on human chromosome 7: evidence for two distinct alphoid domains on a single chromosome.
Mol Cell Biol
 
7
,
349
–356.
Wells D and Delhanty JD (
2000
) Comprehensive chromosomal analysis of human preimplantation embryos using whole genome amplification and single cell comparative genomic hybridization.
Mol Hum Reprod
 
6
,
1055
–1062.
Willard HF, Smith KD and Sutherland J (
1983
) Isolation and characterization of a major tandem repeat family from the human X chromosome.
Nucleic Acids Res
 
11
,
2017
–2033.
Wilton L (
2002
) Preimplantation genetic diagnosis for aneuploidy screening in early human embryos: a review.
Prenat Diagn
 
22
,
512
–518.