Detailed FISH analysis of day 5 human embryos reveals the mechanisms leading to mosaic aneuploidy

D.D. Daphnis1, J.D.A. Delhanty1, S. Jerkovic2, J. Geyer2, I. Craft2 and J.C. Harper1,3

1 UCL Centre for Preimplantation Genetic Diagnosis, Department of Obstetrics and Gynaecology, University College London, 86–96 Chenies Mews, London, WC1E 6HX and 2 The London Fertility Centre, Cozen's House, 112a Harley Street, London W1G 7JH, UK

3 To whom correspondence should be addressed. Email: joyce.harper{at}ucl.ac.uk


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
BACKGROUND: Fluorescence in situ hybridization (FISH) analysis has shown that human embryos display a high level of chromosomal mosaicism at all preimplantation stages. The aim of this study was to investigate the mechanisms involved by the use of two probes for each of three autosomes at different loci and to determine the true level of aneuploid mosaicism by excluding FISH artefacts. METHODS: Embryos were cultured in two different types of medium: group I were cultured in standard cleavage medium for up to day 5 and group II were cultured from day 3 to day 5 in blastocyst medium. Three rounds of FISH were performed. In round 1, the probes used were 1pTel, 11qTel and 18CEP; in round 2, the probes used were 1satII/III, 11CEP and 18qTel; in round 3, the probes used were 18CEP, XCEP and YCEP. RESULTS: A total of 21 embryos were analysed in each group. The FISH results revealed one uniformly diploid and 20 mosaic embryos for group I, and two uniformly diploid and 19 mosaic embryos for group II. The predominant type of mosaicism was diploid/aneuploid. The use of two different probes per autosome was able to distinguish FISH artefacts affecting 5% of nuclei from true single cell anomalies. CONCLUSIONS: Post-zygotic chromosome loss was the most common mechanism leading to aneuploidy mosaicism for both groups, followed by chromosome gain, with fewer examples of mitotic non-disjunction.

Key words: aneuploidy mechanisms/blastocyst/chromosomal mosaicism/FISH/human embryos


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Humans, compared with other species, display poor fecundity since the chance of conception in any one ovulatory cycle is low, being 25% for women under 35 years of age who are trying to conceive and are of proven fertility (Wilcox et al., 1988Go). Despite the advent of IVF (Steptoe and Edwards, 1978Go) and the major advances in reproductive medicine including preimplantation genetic diagnosis (PGD), there still remains a relatively high failure rate after embryo transfer. The high level of early embryonic death must contribute significantly to the observed low fecundity (Wells and Delhanty, 2000Go). A variety of factors have been linked to early pregnancy failures, with chromosome abnormalities, of pre- or post-zygotic origin, being the most important (Munné et al., 1999Go). Embryonic aneuploidy, numerical chromosomal abnormality due to the gain or loss of one or more chromosomes in a gamete, is one of the major causes of reproductive failure in women above 35 years of age, both naturally and following IVF (Hassold, 1986Go; Munné et al., 1995Go; Munné, 2002Go).

Chromosomal analysis of embryos using several techniques including karyotyping, fluorescence in situ hybridization (FISH) and comparative genomic hybridization (CGH) has demonstrated frequent mosaicism that is largely independent of maternal age. Mosaicism mostly arises during cleavage and could have a major effect on embryonic survival. Karyotyping has revealed significant levels of mosaic aneuploidy (Papadopoulos et al., 1989Go; Jamieson et al., 1994Go; Clouston et al., 1997Go, Clouston et al., 2002Go). However, technical difficulties in arresting the chromosomes in metaphase will always result in an underestimate of mosaicism in these studies. The application of single cell CGH provides the opportunity to assess the copy number of all the chromosomes in every cell of an embryo and thus identify the true level of mosaicism (Voullaire et al., 2000Go; Wells and Delhanty, 2000Go). In these two small series (24 embryos in all), 75% of embryos had at least one abnormal cell, but the labour-intensive nature of single cell CGH analysis precludes its widespread application. FISH has allowed chromosome enumeration to be performed on all interphase nuclei within an embryo without the need for culturing cells or preparing metaphase spreads. Thus, the FISH technique overcomes the problems that have arisen during karyotypic analysis and it has been possible to investigate the frequency, stage of onset and the mechanisms leading to chromosomal mosaicism in large groups of embryos. Different types of mosaicism have been reported after FISH analysis including aneuploid mosaics, polyploid and haploid mosaics, chaotic mosaics and multinucleation in cleavage stage embryos (Delhanty et al., 1993Go, 1997Go; Munné et al., 1994Go; Harper et al., 1995Go; Laverge et al., 1997Go; Conn et al., 1998Go; Munné and Cohen, 1998Go; Gianaroli et al., 1999Go; Iwarsson et al., 1999Go, 2000Go) as well as blastocysts (Evsikov and Verlinsky, 1998Go; Veiga et al., 1999Go; Magli et al., 2000Go; Ruangvutilert et al., 2000aGo).

Aneuploid mosaicism is considered to be the most frequent form of mosaicism originating in the first few embryonic divisions (Munné et al., 1994Go) and it is thought that it arises and persists due to the lack of expression of certain cell cycle checkpoint genes during the early cleavage stage (Delhanty and Handyside, 1995Go). Aneuploid mosaics have been shown to arise due to mitotic non-disjunction (which causes reciprocal chromosome loss or gain in daughter cells); in addition, chromosome duplication and anaphase lag are thought to be likely mechanisms (Delhanty et al., 1997Go). Ploidy mosaics have also been reported frequently (Harper et al., 1995Go; Clouston et al., 1997Go; Delhanty et al., 1997Go; Munné et al., 1997Go; Staessen et al., 1999Go), with tetraploidy predominating and more rarely haploidy and triploidy. It has been proposed that tetraploid cells may be a normal feature of the development of the trophectoderm (Drury et al., 1998Go). Tetraploid trophectoderm cells may arise as a result of endoreduplication or endomitosis and possibly play a role in embryo implantation (Drury et al., 1998Go). The origin of diploid/haploid and diploid/triploid mosaicism is not clear and more difficult to explain. Chaotic mosaic embryos have been reported in many studies and are defined as embryos where all or a majority of nuclei show a randomly different chromosome complement. They have been reported in cleavage stage embryos (Harper et al., 1995Go) as well as at the blastocyst stage (Evsikov and Verlinsky, 1998Go).

When using FISH to analyse chromosome constitutions, several technical obstacles emerge, including failure of hybridization, signal overlapping yielding false-negative results, and split or diffused signals (Munné et al., 1998; Ruangvutilert et al., 2000aGo,bGo). FISH is also limited by the number of probes that can be applied simultaneously due to an increasing risk of FISH artefacts and hybridization failure with increasing numbers of probes (Ruangvutilert et al., 2000bGo). These factors have led to an underestimate of chromosome loss as a cause of mosaicism. The use of differentially labelled probes at two different loci to detect a single chromosome has been shown to increase accuracy of detection by reducing scoring errors and has confirmed that low level mosaicism is not a FISH artefact owing to hybridization failure or overlapping signals (Conn et al., 1999Go; Magli et al., 2001Go).

To avoid these problems, in this study, chromosomes 1, 11, 18, X and Y of day 5 embryos were analysed using three sequential rounds of FISH with two probes for each of the autosomes. The aims were to determine the true level of mosaicism by excluding FISH artefacts and thereby obtain information concerning the mechanisms responsible for generating aneuploidy mosaicism.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
This study was performed on embryos that were not transferred and not suitable for freezing and were donated with written consent from patients undergoing routine IVF treatment at the London Fertility Clinic.

Superovulation
Pituitary downregulation was achieved by the administration of GnRH analogue depot injection (Prostap, leuprorelin acetate; Wyeth Laboratories, UK) 3.75 mg or nasal spray (Synarel, nafarelin acetate; Pharmacia, UK) 2 mg/ml per day starting on cycle day 2 for 10–13 days. Ovarian stimulation was commenced using recombinant FSH (Puregon; Organon, UK) either 150 IU/day if aged <38 years or 225 IU/day if ≥38 years. The patients were monitored with serum estradiol levels and vaginal ultrasound regularly starting on day 6 of stimulation, and the dose of FSH was adjusted accordingly. Between days 12 and 14 of stimulation, patients who had a cohort of follicles of 18–20 mm in diameter were given 10 000 IU of HCG (Pregnyl; Organon, UK) to trigger ovulation. Thirty-six hours later, follicular aspiration was carried out under transvaginal ultrasound guidance.

Embryo culture
Oocytes were retrieved using flushing medium [supplemented with sodium pyruvate, human serum albumin (HSA, heparin 10 IU/ml, penicillin 50 000 IU/l, streptomycin 50 mg/l and HEPES; Medicult UK Ltd], incubated in 6% CO2 in air, at 37°C, inseminated and cultured in 500 µl of IVF medium (bicarbonate-buffered medium containing HSA, penicillin and sodium pyruvate; Vitrolife, Scandinavia). On day 1, oocytes were assessed for the number of pronuclei and transferred into 25 µl microdroplets of cleavage medium (bicarbonate-buffered medium containing HSA, penicillin-G, EDTA, glucose, inorganic salts and amino acids; G-1, Vitrolife, Scandinavia). On day 3, the best embryos were selected for transfer and suitable spare embryos were cryopreserved.

Embryo group categorization
The embryos were divided into two groups depending on the culture medium. Group I embryos were cultured in standard cleavage medium (6.1 Vitrolife, Scandinavia) from day 0 to day 5, and group II embryos were cultured in standard cleavage medium from day 0 to day 3 and then in blastocyst medium (6.2 Vitrolife, Scandinavia) from day 3 to day 5. Only embryos that arose from a bipronucleate zygote were included in the study.

Embryo spreading
Embryos were spread as described previously (Harper et al., 1994Go; Ruangvutilert et al., 2000aGo). Each embryo was washed in phosphate-buffered saline (PBS) to remove excess culture medium. The embryo was transferred to a small drop of spreading solution (0.01 mol/l HCl, 0.1% Tween-20) on a poly-L-lysine-coated slide. The spreading solution was used to dissolve the zona pellucida and cytoplasm. Subsequently, the embryonic nuclei were washed by gentle agitation of the spreading solution using a polycarbonate capillary, until free of cytoplasm. The embryo was constantly observed under an inverted microscope. The slides were left to air dry, washed in PBS for 5 min and dehydrated through an ethanol series (70, 90 and 100%). All embryonic nuclei were mapped with the aid of an England Finder (Graticules Ltd, UK) under a phase-contrast microscope.

Control analysis
For each FISH procedure, a control slide from a normal male lymphocyte culture was used to assess the efficiency of the FISH by randomly scoring 100 interphase nuclei. Moreover, a set of interphase nuclei (10–20) and metaphases (2–5) were mapped as representative results from each control slide, using an England Finder, and analysed in the same way as the embryos using three rounds of FISH.

FISH procedure
FISH was performed in three sequential rounds. The first and second rounds were performed with probes for chromosomes 1, 11 and 18, whilst the third round used probes for the sex chromosomes and chromosome 18 (Table I). All probes except one were obtained from Vysis (UK) Ltd. The first round included the following probes: 1p SpectrumGreen (telomere CEB108/T7), which hybridizes to the subtelomere region of the short arm of chromosome 1; 11q SpectrumOrange (telomere VIJyRM2072), which hybridizes to the subtelomere region of the long arm of chromosome 11; and CEP18 SpectrumAqua ({alpha}-satellite D18Z1), which hybridizes to the centromere region of chromosome 18. The second round included the following probes: 1sat-III (satellite II/III), which hybridizes to the heterochromatic region of chromosome 1 (laboratory prepared); CEP11 SpectrumGreen (satellite D11Z1), which hybridizes to the centromere region of chromosome 11; and 18q SpectrumOrange (telomere VIJyRM2050), which hybridizes to the subtelomere region of the long arm of chromosome 18. The third round included the following probes: repeated use of CEP18 SpectrumAqua (as an internal control); CEPX SpectrumGreen ({alpha}-satellite DXZ1), which hybridizes to the centromere region of chromosome X; and CEPY SpectrumOrange ({alpha}-satellite DYZ3), which hybridizes to the centromere region of chromosome Y (Table I).


View this table:
[in this window]
[in a new window]
 
Table I. Probe information for sequential rounds of FISH analysis

 
All three rounds of FISH were performed essentially as described previously (Harper et al., 1994Go). The slides were pre-treated by incubation in 100 µg/ml pepsin in 0.01 mol/l HCl at 37°C for 20 min, washing in bi-distilled water and PBS, and further incubated in 1% paraformaldehyde in PBS at 4°C for 10 min. The slides were washed in PBS and twice in water, dehydrated through an ethanol series (70, 90 and 100%) and air dried. Aliquots of 5 µl of the first hybridization solution were applied to a glass slide containing the fixed cells and covered with a 13 x 13 mm coverslip. Slide and probes were co-denatured at 75°C for 5 min and were incubated at 37°C in a moist chamber overnight. They were washed at 45°C in 50% formamide/2x standard saline citrate (SSC) for 10 min (three times), 2x SSC for 10 min and 2x SSC/0.1% NP-40 for 5 min (all in the dark). Finally, the slides were air dried and mounted in Vectashield (Vector Laboratories, CA) with 4',6-diamidino-2-phenylindole (DAPI). The signals of every cell were visualized and assessed using a Zeiss Axioscope fluorescence microscope, captured using a cooled charged coupled device camera (photometrics) and analysed using ‘Smartcapture 2’ (Digital Scientific, UK) software.

After the analysis of the first round, all probes were washed off using 4x SSC/0.05% Tween-20 for 5 min (x2) and in PBS for 10 min at room temperature. Aliquots of 5 µl of the second hybridization solution were applied to the glass slide, covered with a 13 x 13 mm coverslip and co-denatured at 75°C for 5 min. The slides subsequently were left to hybridize at 37°C in a moist chamber overnight. The post-hybridization washes consisted of 3 x 10 min in 50% formamide/2x SSC, 1 x 10 min in 2x SSC and 1 x 5 min in 2x SSC/0.1% NP-40. The slides were air dried, mounted and analysed in the same way as in the first round. The embryonic nuclei were specifically identified with the aid of a map previously drawn using an England Finder.

Probes of the second round were washed off as described previously. The third round X/Y/18 probe mix was applied to the slides and co-denatured at 75°C for 5 min. The slides were incubated in a moist chamber at 37°C for 2 h and subsequently were washed, mounted and analysed as described previously.

Criteria for signal scoring
The criteria for signal scoring were as used by Hopman et al. (1988)Go, which proposed that signals had to be a minimum of a signal's width apart to be scored as two separate signals.

Mosaicism and events
All embryonic chromosome patterns were classified using the same criteria as Delhanty et al. (1997)Go described for cleavage stage embryos.

Diploid mosaic embryos with aneuploid cells were considered to have arisen through three different mechanisms: (i) when the embryo contained cells with monosomies, then the mechanism was classed as ‘chromosome loss’ (CL); (ii) when the embryo contained cells with trisomies, then the mechanism was classed as ‘chromosome gain’ (CG); and (iii) when the embryo had monosomies and trisomies of the same chromosome(s) in different cells, this was classified as mitotic non-disjunction (MND). Nuclei with multiple abnormalities affecting at least three chromosomes were classed as chaotic and not included in the analysis of the events. This included nullisomies and tetrasomies.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Twenty-seven couples donated a total of 47 embryos for this research project. The overall mean maternal age was 33.7 years (range 19–41). The mean maternal age for group I embryos was 35.2 years (range 27–41) and for group II was 32.1 years (range 19–41).

Five arrested embryos were not included, since the signals were not analysable due to loss of most nuclei during sequential rounds of FISH and poor quality nuclei. Embryos from both groups included nuclei which showed contradictory information for the two probes used for each autosome in the different rounds of FISH. These cells were excluded from the analysis of events and classed as inconsistent results.

FISH analysis of controls
Each FISH experiment included a control male lymphocyte slide with mapped nuclei in order to assess the efficiency of probe hybridization in the sequential rounds. Overall, 87.1% (range 78–96) of the control nuclei showed normal signals for all eight probes used. Subtelomeric probes for chromosomes 1p, 11q and 18q showed a higher incidence of one signal per chromosome per nucleus, 7.8 (range 4–11), 3.1 (range 1.2–4.5) and 3.9% (range 2–5), respectively. Heterochromatic region or centromeric probes, 1satII/III, 11CEP and 18CEP (in both the first and third rounds), displayed a lower incidence of one signal per chromosome per nucleus, 6.2 (range 4.4–7.5), 2.4 (range 0.9–3.3) and 2.2% (1.5–3.4), respectively. Nuclei with one signal for the X chromosome and no signal for chromosome Y and nuclei with three or more signals for autosomes comprised <1%. Furthermore, all three subtelomeric probes, 1p, 11q and 18q, showed the occurrence of split signals (i.e. replicated DNA) in 2.2, 1.4 and 1.9% of nuclei, respectively, which was considered normal due to the position of the probes.

FISH analysis of embryos
Twenty-one embryos were analysed for each group. FISH results for group I and group II embryos are shown in Tables II and III, respectively.


View this table:
[in this window]
[in a new window]
 
Table II. Results from FISH analysis of group I embryos after three sequential rounds of hybridization with probes for chromosomes 1, 11, 18, X and Y

 

View this table:
[in this window]
[in a new window]
 
Table III. Results from FISH analysis of group II embryos after three sequential rounds of hybridization with probes for chromosomes 1, 11, 18, X and Y

 
In group I, the total number of nuclei analysed was 401 out of 410 (nine were lost during sequential rounds of FISH). Thus, the mean number of nuclei analysed per embryo was 20 (range 4–45). In all, 72% (±20 SD) of the nuclei analysed were diploid for the chromosomes tested (Figure 1a–); however, only a single embryo was uniformly diploid for the tested chromosomes (embryo 9.1). One embryo was aneuploid mosaic, the remainder (14) were diploid mosaics.



View larger version (38K):
[in this window]
[in a new window]
 
Figure 1. Results of three sequential rounds of FISH analysis on a single female embryonic nucleus. (a) Using probes for 1p, 11q and 18CEP; (b) using probes for 1cep, 11cep and 18q; and (c) using probes for XCEP, YCEP and 18CEP. In all three rounds, the cell appears to be diploid. The probe for 18CEP is present in rounds 1 and 3 and shows identical results.

 
In group I, there were 32 post-zygotic errors (16 CL, 14 CG and two instances of MND) (Table IV). Table V summarizes the breakdown by chromosome, with chromosome 1 showing the most losses and the Y chromosome the most gains. Three embryos were mosaic diploid/polyploid (6.1, 17.2 and 21.1) and although most polyploid cells were tetraploid, embryo 17.2 had hexaploid and octaploid cells. Embryos 8.1, 12.2 and 17.2 were the only embryos with chaotic cells. Inconsistent results affected 26 out of 401 (6.5%) nuclei and were considered to be FISH artefacts.


View this table:
[in this window]
[in a new window]
 
Table IV. Aneuploidy mechanisms for group I embryos

 

View this table:
[in this window]
[in a new window]
 
Table V. Mechanisms of mosaicism by chromosome for group I

 
Table III shows the results of group II embryos. The total number of nuclei analysed was 1143 out of 1171 (28 nuclei were lost during sequential rounds of FISH). The mean number of nuclei per embryo was 55, ranging from 20 to 100. Seventy-eight percent (±15 SD) of the nuclei analysed were found to be diploid, although only two embryos were found to be uniformly diploid (16.1 and 23.1) as tested. The majority (16 out of 21) of embryos were found to have some tetraploid cells (confirmed in all rounds), which totalled ~7% of the abnormal cells analysed (Figure 2a–). Six out of 21 embryos included a chaotic complement. Those embryos included at least one cell which had more than two abnormalities per cell, including nullisomies and tetrasomies. Furthermore, there was one embryo (15.4) which was a diploid/aneuploid/haploid/chaotic mosaic. This embryo contained only one cell which was haploid; however, the results were consistent in all rounds.



View larger version (48K):
[in this window]
[in a new window]
 
Figure 2. Results of three sequential rounds of FISH analysis on a single male embryonic nucleus. (a) Using probes for 1p, 11q and 18CEP; (b) using probes for 1cep, 11cep and 18q; and (c) using probes for XCEP, YCEP and 18CEP. This nucleus displayed a tetraploid chromosome complement in all three rounds of FISH. In the third round, there are two signals for X and two signals for Y.

 
In group II, 37 post-zygotic errors took place, which gave rise to 14 mosaic aneuploid embryos (Table VI). Some of those embryos also contained polyploid, chaotic and haploid cells. Moreover, in group II, there were nine MND events, mostly affecting chromosome X, in comparison with two in group I embryos (Table VII). Once again, chromosome loss was the predominant mechanism leading to mosaic aneuploidy (Table VII), but in this group it involved mainly chromosome 18, while an additional copy of chromosome 11 was the most frequent gain. FISH artefacts (inconsistent results) affected 50 out of 1143 (4.4%) nuclei.


View this table:
[in this window]
[in a new window]
 
Table VI. Aneuploidy mosaicism mechanisms for group II

 

View this table:
[in this window]
[in a new window]
 
Table VII. Mechanisms of mosaicism by chromosome for group II

 
Of the nuclei showing inconsistent results, 34 out of 76 (45%) were due to failure of hybridization of the subtelomeric probe for chromosome 18 (18q). The probe for the satellite II/III region of chromosome 1 exhibited the highest failure rate of the centromeric probes (14.5%).


    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The main aim of this study was to gain information about the mechanisms leading to aneuploid mosaicism in human embryos. This was achieved by the use of triple colour FISH in three sequential rounds of hybridization. The use of two probes per autosome and the dual use of a chromosome 18 centromere probe allowed the distinction between true single cell anomalies and artefacts, which showed up as inconsistent results. Of the 42 day 5 embryos, only three were uniformly diploid for the tested chromosomes, a single embryo was the product of a meiotic error and the remainder were mosaic.

Prevalence and type of mosaicism
The opportunity also arose during the study to compare the effects of transferring some embryos on day 3 of culture from the standard cleavage medium to the more enriched blastocyst development medium (group II). Both groups showed a high prevalence of mosaicism; 95% (20 out of 21) of embryos were mosaic for group I and 90% (19 out of 21) of embryos were mosaic for group II. These findings are similar to the findings from other studies, which have analysed blastocysts either by karyotyping (Clouston et al., 1997Go, 2002Go) or by FISH analysis (Evsikov and Verlinsky, 1998Go; Veiga et al., 1999Go; Magli et al., 2000Go; Ruangvutilert et al., 2000aGo). In the present study, the proportion of cells testing normal (diploid) per embryo amounted to 73% for group I and 78% for group II. These results are also comparable with those of other studies, which have shown a mean of 72% of all analysable nuclei per blastocyst normal for the tested chromosomes (Coonen et al., 2004Go).

The predominant type of mosaicism in this study was that of diploid/aneuploid, affecting 32 of the 42 embryos. However, in 20 of these embryos, other cell lines were also present including polyploid, haploid and chaotic types. The available evidence suggests that aneuploid mosaicism detected at the cleavage stage is generated during one of the first three divisions, typically at the first or second division (Delhanty et al., 1997Go; Munné et al., 2002Go; Katz-Jaffe et al., 2004Go). Data from these day 5 embryos suggest that if the aneuploid cells arose during cleavage, they are at a selective disadvantage compared with the diploid cells which have outgrown them, since both groups were comprised of embryos which had grown beyond the cleavage stage. Alternatively, anomalies affecting a few cells or a single cell may have been generated in later divisions since for most embryonic genes, maximum expression does not occur until the blastocyst stage (D.Wells et al., in preparation). Reduced numbers of transcripts from certain of the cell cycle checkpoint genes could allow errors to arise during days 4 and 5. Although there are many single cell anomalies, in some cases the total number of aneuploid cells per embryo is significant and, bearing in mind that only a few chromosome pairs were tested, there are likely to be many more undetected abnormalities in these embryos. The general consensus is that embryos with less than half of their cells truly normal diploid would be unlikely to survive beyond the implantation stage (Munne et al., 2004).

Tetraploidy
The phenomenon of tetraploidy, especially in blastocyst studies, has been well documented. It is considered as a normal feature of human embryonic development (Evsikov and Verlinsky, 1998Go). In our study, there was a mean of 1.9 and 7% of tetraploid cells per embryo for group I and group II, respectively. The difference in frequency of tetraploid cells as a result of changed culture medium was not unexpected, but also raises the question of whether the development of tetraploid cells is favoured by certain culture conditions. It has been proposed that pure polyploid embryos or diploid/polyploid embryos might arise from failure of cytokinesis (Harper et al., 1995Go; Munné and Cohen, 1998Go) or from cell fusion (Benkhalifa et al., 1993). The existence in our study of a few triploid cells and cells of higher ploidy suggests that these may have been formed by cell fusion. It is of interest in this context that a recently established line of human trophectoderm cells exhibits syncytium formation but the nuclei remain diploid (H.Moore, personal communication).

Distinction between artefacts and true results
The use of two probes per autosome allowed this study to detect an error rate of 5% per nucleus and to exclude those artefacts from the analysed results. For both groups, a true chromosomal error in a total of 76 nuclei would have been missed if only one probe per chromosome had been used. Therefore, by adopting our chosen strategy, we were able to detect the true levels of mosaicism for the three autosomes studied. The probe that showed the highest rate of failure of hybridization on the embryonic material was the subtelomere probe for chromosome 18 (Table VIII). Subtelomeric probes generally have lower hybridization efficiency than probes that detect repeat sequences, but in this case the situation possibly was exacerbated because the probe was used in the second treatment round. However, in the lymphocyte control nuclei, the probes for chromosome 1, both for the subtelomeric (used in the first round) and the heterochromatic regions, showed the highest failure rates.


View this table:
[in this window]
[in a new window]
 
Table VIII. Failure rates for each probe in the embryonic nuclei for both groups

 
Many of the aneuploidies detected in our study were of a single chromosome and confined to a single cell. It is clear from CGH data on full analysis of individual blastomeres that single cell anomalies affecting one chromosome are a common feature at the cleavage stage (Voullaire et al., 2000Go; Wells and Delhanty, 2000Go). In these two small series, nine of the 24 embryos fully analysed showed such single cell anomalies.

Mechanisms of aneuploidy mosaicism
In the current study, CL was the predominant mechanism leading to mosaicism in both groups of embryos, being responsible for 50% of aneuploid cells in group I and 40.5% in group II. CG followed, with 44% in group I and 35.1% in group II. However, only 6% of aneuploid cells occurred due to MND in group I, while four times this percentage arose by this mechanism in group II The differential involvement of chromosomes in MND (predominantly chromosome X in group II which was not involved at all in group I) is interesting but may well be due to chance.

Similarly, it is of interest that in group I embryos, in which growth had slowed or arrested, chromosome 1 showed a high incidence of loss but in group II embryos, which had continued dividing, chromosome 1 was not affected by loss at all. This finding is in accordance with findings of Sandalinas et al. (2001)Go which revealed that embryos fully monosomic for the larger chromosomes fail to reach the blastocyst stage. This might possibly be due to the fact that the presence of cells with monosomies of such a large chromosome would have a detrimental effect on development and had been selected against in the more rapidly dividing group II embryos. Trisomy 18 caused by CG was seldom found in either group, which is a similar finding to that reported recently by Coonen et al. (2004)Go. Furthermore, the occurrence of monosomy 18 was shown to be high (especially in group II embryos), indicating that chromosome 18 might be more prone to CL compared with the other autosomes and gonosomes tested.

The current study shows that CL and CG are the mechanisms mostly affecting human day 5 embryos. CL is the predominant mechanism in this study leading to mosaicism; furthermore, analysis of a larger pool of embryos may prove such a finding statistically. This reinforces data on day 3 embryos obtained earlier by our group using dual locus-specific yeast artificial chromosome (YAC) and plasmid probe combinations for various autosomes (C.Conn and J.D.A.Delhanty, unpublished observations). Chromosome loss is presumed to occur via anaphase lag, in this case during mitosis. Coonen et al. (2004)Go concluded in their study on a much larger number of blastocysts that anaphase lagging is the major cause of chromosomal mosaicism. However, since they were using a single probe for each chromosome, they were only able to count as valid those abnormalities affecting at least two cells. In relation to the findings in these two studies, it is of considerable interest that aneuploidy screening of cleavage stage embryos has shown that CL is more common than CG as a cause of constitutional aneuploidy arising during meiosis (Munné et al., 2004Go).

During this study, the average maternal age was 34 years but there was still a high prevalence of mosaicism. The three embryos that were uniformly diploid for the tested chromosomes were donated by three women <30 years of age, whereas six women whose age was >38 years donated nine embryos, which were either diploid/aneuploid mosaic or diploid/aneuploid/chaotic mosaic. While these observations may indicate a trend, clearly the numbers in our study are too small to enable analysis of a maternal age effect. However, a report of a study of a large number of cleavage stage embryos from women in an older age group showed a significant association of mitotic aneuploidy with advanced maternal age (Munné et al., 2002Go). The association was particularly marked for the MND category. In that study, CL was only a third as frequent as MND and consequently mitotic anaphase lag failed to show a significant increase with maternal age. The increased frequency of MND may be due to the age of the group studied but, since only a single probe per chromosome was used, a cut off point of 10% was used to avoid ‘FISH error’, i.e. embryos with <10% abnormal cells were considered ‘normal’, almost certainly leading to an underestimate of anomalies affecting a single cell. Possibly a study of a large number of embryos using two probes per chromosome might show a significant association of mitotic anaphase lag with increasing maternal age.


    Acknowledgements
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
We would like to express our gratitude to the whole London Fertility Clinic (LFC) team and the University College London team of preimplantation genetic diagnosis. We are exceptionally grateful to Life-Force Research Ltd for funding this study.


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Benkhalifa M, Janny L, Vye P, Malet P, Boucher D and Mennezzo Y (1993) Assessment of polyploidy in human morulae and blastocysts using co-culture and fluorescent in situ hybridisation. Hum Reprod 8, 895–902.[Abstract]

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

Clouston HJ, Herbert M, Fenwick J, Murdoch AP and Wolstenholme J (2002) Cytogenetic analysis of human blastocysts. Prenat Diagn 12, 1143–1152.[CrossRef]

Conn CM, Harper JC, Winston RML and Delhanty JDA (1998) Infertile couples with Robertsonian translocations: preimplantation genetic analysis of embryos reveals chaotic cleavage divisions. Hum Genet 102, 117–123.[CrossRef][ISI][Medline]

Conn CM, Cozzi J, Harper JC, Winston RM and Delhanty JD (1999) Preimplantation genetic diagnosis for couples at high risk of Down syndrome pregnancy owing to parental translocation or mosaicism. J Med Genet 36, 45–50.[Abstract/Free Full Text]

Coonen E, Derhaag JG, Dumoulin JC, Van Wissen LC, Bras M, Janssen M, Evers JL and Geraedts JP (2004) Anaphase lagging mainly explains chromosomal mosaicism in human preimplantation embryos. Hum Reprod 19, 316–324.[Abstract/Free Full Text]

Delhanty JDA and Handyside AH (1995) The origin of genetic defects in the human and their detection in preimplantation embryo. Hum Reprod Update 1, 201–215.[CrossRef][ISI][Medline]

Delhanty JDA, Griffin DK and Handyside AH (1993) Detection of aneuploidy and chromosomal mosaicism in human embryos during preimplantation sex determination by FISH. Hum Mol Genet 2, 1185.

Delhanty JDA, Harper JC, Asangla A, Handyside AH and Winston RML (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]

Drury K, Kovalinskaya R and William S (1998) Polyploidy as a normal function of the trophoblastic development in human preimplantation embryos observed by fluorescent in situ hybridisation analysis. Fertil Steril 70 (Suppl 1), 0–25.

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

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.[CrossRef][ISI][Medline]

Harper JC, Coonen E, Ramaekers FC, Delhanty JD, Handyside AH, Winston RM and Hopman AH (1994) Identification of the sex of human preimplantation embryos in two hours using an improved spreading method and FISH using directly-labelled probes. Hum Reprod 9, 721–724.[Abstract]

Harper JC, Coonen E, Handyside AH, Winston RML and Delhanty JDA (1995) Mosaicism of autosomes and sex chromosomes in morphologically normal, monospermic preimplantation human embryos. Prenat Diagn 15, 41–49.[ISI][Medline]

Hassold TJ (1986) Chromosome abnormalities in human reproductive wastage. Trends Genet 2, 105–110.[CrossRef][ISI]

Hopman AHN, Raemakers FCS, Raap AK, Beck JLM, Devilee P, Ploeg van der M and Vooijis GP (1988) In-situ hybridisation as a tool to study numerical chromosome aberrations in solid bladder tumours. Histochemistry 89, 307–316.[ISI][Medline]

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

Iwarsson E, Malmgren H, Inzunza J, Ahrlund-Richter L, Sjoblom P, Rosenlund B, Fridstrom M, Hovatta O, Nordenskjold M and Blennow E (2000) Highly abnormal cleavage divisions in preimplantation embryos from translocation carriers. Prenat Diagn 20, 1038–1047.[CrossRef][ISI][Medline]

Jamieson ME, Coutts JRT and Connor JM (1994) The chromosome constitution of human preimplantation embryos fertilised in vivo. Hum Reprod 9, 709–715.[Abstract]

Katz-Jaffe MG, Trounson AO and Cram DS (2004) Mitotic errors in chromosome 21 of human preimplantation embryos are associated with non-viability. Mol Hum Reprod 10, 143–147.[Abstract/Free Full Text]

Laverge H, De Sutter P, Verschraegen-Spae MR, De Paepe A and Dhont M (1997) Triple colour fluorescent in-situ hybridization for chromosomes X, Y and 1 on spare human embryos. Hum Reprod 12, 809–814.[Abstract]

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.[Abstract/Free Full Text]

Magli MC, Sandalinas M, Escudero T, Morrison L, Ferraretti AP, Gianaroli L and Munné S (2001) Double locus analysis of chromosome 21 for preimplantation genetic diagnosis of aneuploidy. Prenat Diagn 21, 1080–1085.[CrossRef][ISI][Medline]

Munné S (2002) Preimplantation genetic diagnosis of numerical and structural chromosome abnormalities. Reprod Biomed Online 4, 183–196.[Medline]

Munné S and Cohen J (1998) Chromosome abnormalities in human embryos. Hum Reprod Update 4, 842–855.[Abstract/Free Full Text]

Munné S, Weier HU, Grifo J and Cohen J (1994) Chromosome mosaicism in human embryos. Biol Reprod 51, 373–379.[Abstract]

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

Munné S, Marquez C, Reing A, Garrisi J and Alikani M (1998) Chromosome abnormalities in embryos obtained after conventional in vitro fertilisation and intracytoplasmic sperm injection. Fertil Steril 69, 904–908.[CrossRef][ISI][Medline]

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

Munné S, Magli C, Cohen J, Morton P, Sadowy S, Gianaroli L, Tucker M, Marquez C, Sable D, Ferraretti AP, Massey JB and Scott R (1999) Positive outcome after preimplantation diagnosis of aneuploidy in human embryos. Hum Reprod 14, 2192–2199.

Munné S, Sandalinas M, Escudero T, Marquez C and Cohen J (2002) Chromosome mosaicism in cleavage-stage human embryos: evidence of a maternal age effect. Reprod Biomed Online 4, 223–232.[Medline]

Munné S, Bahce M, Sandalinas M, Escudero T, Marquez C, Velilla E, Colls P, Oter M, Alikani M and Cohen J (2004) Differences in chromosome susceptibility to aneuploidy and survival to first trimester. Reprod Biomed Online 8, 81–890.[Medline]

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

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

Ruangvutilert P, Delhanty JD, Rodeck CH and Harper JC (2000b) Relative efficiency of FISH on metaphase and interphase nuclei from non-mosaic trisomic or triploid fibroblast cultures. Prenat Diagn 20, 159–162.[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]

Staessen, Van Assche E, Joris H, Bonduelle M, Vandervorst M, Liebaers I and Van Steirteghem A (1999) Clinical experience of sex determination by fluorescent in situ hybridisation for preimplantation genetic diagnosis. Mol Hum Reprod 5, 382–389.[Abstract/Free Full Text]

Steptoe PC and Edwards RG (1978) Birth after the reimplantation of a human embryo. Lancet 2, 366.

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

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

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.[Abstract/Free Full Text]

Wells D and Delhanty JDA (2000) Comprehensive chromosomal analysis of human reproduction using whole genome amplification and single cell comparative genomic hybridisation. Mol Hum Reprod 6, 1055–1062.[Abstract/Free Full Text]

Wells D, Escudero T, Levy B, Hirschhorn K, Delhanty JD and Munné S (2002) First clinical application of comparative genomic hybridization and polar body testing for preimplantation genetic diagnosis of aneuploidy. Fertil Steril 78, 543–549.[CrossRef][ISI][Medline]

Wilcox AJ, Weinberg CR, O'Connor JF, Baird DD, Schlatterer JP, Canfield RE, Armstrong EG and Nisula BC (1988) Incidence of early loss pregnancy. N Engl J Med 319, 189–194.[Abstract]

Wilton L, Williamson R, McBain J, Edgar D and Voullaire L (2001) Birth of a healthy infant after preimplantation confirmation of euploidy by comparative genomic hybridization. N Engl J Med. 345, 1537–1541.[Free Full Text]

Submitted on May 13, 2004; accepted on September 17, 2004.