Articles Diagnosis of trisomy 21 in preimplantation embryos by single-cell DNA fingerprinting

Transcription

1 RBMOnline - Vol 4. No Reproductive BioMedicine Online; on web 6 December 2001 Articles Diagnosis of trisomy 21 in preimplantation embryos by single-cell DNA fingerprinting Mandy Katz is a PhD student at the Monash Institute of Reproduction and Development, Australia. Her thesis centres on understanding the genetic basis of viable human preimplantation embryos. Mandy has a Master s degree in Reproductive Sciences with experience in a range of assisted reproductive technologies and also works as a clinical embryologist at Monash IVF. Mandy Katz MG Katz 1,3, J Mansfield 2, L Gras 2, AO Trounson 1, DS Cram 1,2 1 Monash Institute of Reproduction and Development, Monash University, Clayton, Victoria, Australia 2 Monash IVF, Melbourne, Victoria, Australia 3 Correspondence: Tel: (61) ; Fax: (61) ; mandy.katz@med.monash.edu.au Abstract Many couples presenting for preimplantation genetic diagnosis (PGD) for a single gene disorder are of advanced reproductive age (>35 years) and have a greater chance of producing embryos with chromosomal aneuploidies. The most common chromosomal aneuploidy observed in newborns is trisomy 21, or Down s syndrome. Consequently, the availability of a highly reliable system that simultaneously detects the heritable gene disorder and trisomy 21 would be beneficial to couples at specific risk. A pentaplex chromosome 21 (Ch 21) single-cell DNA fingerprinting system was developed in a multiplex fluorescence polymerase chain reaction (FL-PCR) on single cells. High reliability and accuracy rates were observed, together with low allele dropout (ADO) and preferential amplification rates on diploid buccal cells, trisomy 21 buccal cells and blastomeres derived from Ch 21 aneuploid embryos. A combined multiplex FL-PCR format was optimized with the common cystic fibrosis δ F508 mutation and validated on single buccal cells from a carrier of the cystic fibrosis δ F508 mutation. This new test is a very powerful technique, which also allows confirmation of the embryo parentage and the identification of extraneous DNA contamination that could cause a misdiagnosis in PGD cases. Keywords: cystic fibrosis, DNA fingerprinting, FL-PCR, preimplantation genetic diagnosis, single cell, trisomy 21 Introduction Preimplantation genetic diagnosis (PGD) has now become an acceptable alternative to prenatal diagnosis (PND) for couples at risk for single gene disorders. The main advantage of PGD is that genetic screening for the indicated single gene disorder occurs prior to the establishment of a pregnancy, thereby avoiding the possibility of a therapeutic termination. Many couples presenting for PGD have usually already experienced the birth of an affected child or have undergone PND and therapeutic terminations in previous pregnancies. Hence, at the time of their PGD procedure, the female partners of these couples are usually of advanced reproductive age (ARA) (>36 years). For women of ARA, there is an increased risk of meiotic non-disjunction during gametogenesis, which results in embryos with chromosomal aneuploidies (Damario et al., 1999; Verlinsky and Kuliev, 1999). Down s syndrome is the one of the most common chromosomal aneuploidies found in spontaneous abortions and stillbirths, as well as being observed in 1 in every 700 live births (Hook, 1981; Hook and Cross, 1983; Boue et al., 1985). It has been estimated that 91 95% of trisomy 21 cases have a maternal origin (Mikkelsen et al., 1995), with most maternal meiotic errors initiated in meiosis I (Antonarakis et al., 1992; Petersen and Mikkelsen, 2000). Down s syndrome is the most common known cause of mild to severe mental impairment in the developed world and is also associated with a high risk of congenital heart disease and leukaemia, an abnormal face and simian crease. Hence, PND and screening for Down s 43

2 44 syndrome for women of ARA is now an established component of antenatal care in most developed countries (Mulvey and Wallace, 2001). All couples where the woman is of ARA are recommended to undergo PND for Down syndrome and other chromosomal abnormalities when a pregnancy is established. It would be tragic for couples at risk for single gene disorders undergoing PGD, if an embryo free of the single gene disorder resulted in a chromosomally abnormal fetus detected later at PND. Hence, the availability of a highly reliable single-cell system that would simultaneously detect at the preimplantation stage the heritable single gene disorder and trisomy 21 would be beneficial to PGD couples at specific genetic and ARA risk. Currently, PGD for detection of chromosomal aneuploidies is performed by fluorescence in-situ hybridization (FISH) (Verlinsky et al., 1998; Munné et al., 1999; Gianaroli et al., 1999). However, FISH is limited to diagnosis only at the chromosomal level and thus is unable to diagnose single gene defects (Coonen et al., 1998). Fluorescent polymerase chain reaction (FL-PCR) is a rapid, highly sensitive and inexpensive method, capable of detecting both single gene defects and chromosomal aneuploidies in a single cell (Findlay et al., 1999). FL-PCR has been successfully used to diagnose numerous single gene defects in embryos, including cystic fibrosis (CF) and Huntington s disease (Delhanty and Harper, 2000; Report of the 11th Annual Meeting of International Working Group on Preimplantation Genetics, 2001). Recently, FL-PCR involving several highly polymorphic microsatellite markers in a multiplex reaction has been performed to detect specific chromosomal aneuploidies in prenatal aneuploidy screening (Mansfield, 1993; Pertl et al., 1994; Cirigliano et al., 2001) and single cells (Sherlock et al., 1998; Blake et al., 1999; Findlay et al., 1995, 1999). Previously published single-cell FL-PCR systems for aneuploidy detection have been plagued with high rates of either allele dropout (ADO, the total amplification failure of one allele) or preferential allele amplification (the under-representation of one allele, resulting in a distortion from the expected 1:1 di-allelic ratio) (Sherlock et al., 1998; Findlay et al., 1999), reducing the degree of reliability for allele quantification and hence aneuploidy diagnosis at the single cell level. Thus, to maximize reliability, the present authors developed a single-cell DNA fingerprinting system incorporating five chromosome 21 (Ch 21) microsatellite markers so that if one locus is affected by either ADO, preferential allele amplification or parental homozygosity, there are still other microsatellite markers available for making a diagnosis. Materials and methods Human buccal cell samples Buccal cells were collected from subjects by twirling a cytology brush (EndoScanPlus; Medico, Melbourne, Victoria, Australia) on the inner cheek for 30 s. The head of the cytology brush was then immersed in 750 μl of phosphate-buffered saline (PBS). Extraction of genomic DNA from human buccal cells Eppendorf tubes (1.5 ml) containing the head of the cytology brush were centrifuged for 1 min at 8000 g and the brush removed. The tubes were centrifuged again for 1 min at 8000 g, the supernatant decanted and the cell pellet resuspended in 600 μl PBS. Following another cycle of centrifugation, the cell pellet was re-suspended in 10 μl lysis buffer (200 mm KOH, 50 mm dithiothreitol (DTT)), vortexed briefly and heated at 65 C for 10 min. The solution was then neutralized by the addition of 10 μl neutralizing buffer (900 mm Tris-HCl ph 8.3, 200 mm HCl, 300 mm KCl), mixed, and centrifuged for 10 min at 10,500 g. The supernatant containing the DNA was transferred to a sterile 0.5 ml microcentrifuge tube and stored at 20 C. Isolation of single human buccal cells Buccal cell samples were collected from a normal female and a patient with Down s syndrome. Cells were collected into a 1.5ml Eppendorf tube containing 750 μl PBS, centrifuged at 6000 g and re-suspended in 500 μl PBS. A 10 μl aliquot was examined for single nucleated cells under an inverted microscope (Leica MS5) and only intact single cells were collected into finely pulled 22.9 cm long glass Pasteur pipettes (Becton Dickinson, Lincoln Park, NJ, USA). Each cell was washed through three further 5 μl drops of PBS before transfer into sterile 0.2 ml PCR tubes on ice. Tubes were immediately frozen at 80 C prior to PCR analysis. Embryo biopsy and FISH Poor prognosis couples on the PGD programme underwent standard IVF treatment, including ovulation induction, surgical aspiration of the oocytes and sperm collection. Oocytes were fertilized by intracytoplasmic sperm injection (ICSI), and on day 3 of embryonic development, cleavage-stage embryos with 5 8 cells were subjected to biopsy. Embryos were incubated in Ca ++ /Mg ++ -free medium prior to zona drilling using acid Tyrode s solution, and one or two cells were removed with a fine glass pipette. Isolation of human blastomeres from aneuploid embryos Biopsied embryos diagnosed as aneuploid by FISH on day 3 are regarded as genetically abnormal. Under guidelines established by the Infertility Treatment Authority in Victoria, aneuploid embryos deemed to be unsuitable for transfer must be left to succumb on the bench for 24 h. Succumbed aneuploid embryos for chromosome 21 were treated with pronase (2 mg/ml in HEPES-buffered human tubal fluid (HTF) culture medium) for 1 min to dissolve the zona pellucida and transferred into Ca ++ /Mg ++ -free medium to dissociate the blastomeres. Single blastomeres were carefully washed through three 5 μl drops of PBS and transferred into a sterile 0.2 ml PCR tube on ice. Microsatellite markers The six tetranucleotide microsatellite markers investigated for aneuploidy detection of chromosome 21 (Ch 21) were selected from Human Genome and Cooperative Human Linkage Centre (CHLC) databases for high heterozygosity (>0.8): D21S11, D21S1411, D21S1437, D21S1413, D21S1412 and D21S1442. The cystic fibrosis δ F508 primers have been described previously (Tsai, 1999). Primers were synthesized and

3 fluorescently labelled (6-FAM, HEX or TET) by Applied Biosystems, Melbourne, Victoria, Australia. All primer pairs were diluted in molecular biology grade H 2 O (Sigma, Melbourne, Victoria, Australia) to 200 pmol/μl stock solutions under sterile conditions and stored in aliquots of 100 pmol/μl at 20 C until use. Single-cell multiplex FL-PCR The optimized single-cell multiplex FL-PCR consisted of the following: 2.5 μl of 10 Taq PCR Buffer (500 mm KCl, 100mM Tris-HCl, ph 9.0 and 15 mm MgCl 2 ), 0.5 μl of 10 mm dntps (200 μm), 0.3μl of Taq polymerase (5 units/μl) (Amersham Pharmacia Biotech, Sydney, New South Wales, Australia), 12.7 μl MQ-H 2 O and 9 μl of primer mix (5 25 pmol of each primer pair), making a final volume of 25 μl. Manual Hot Start and multiplex FL-PCR were performed using the 9700 Thermocycler PCR machine (Applied Biosystems) situated in a secluded room to avoid laboratory DNA contamination. A total of 36 thermal cycles of denaturation for 45 s at 94 C, annealing for 45 s at 60 C, and extension for 1 min at 72 C were performed. Positive control tubes contained cells in 1 2 μl of PBS, whereas negative control tubes contained either 1 2 μl of PBS from the last wash droplet or no cell. Genescan analysis of DNA fingerprints Each PCR product (1.0 μl) was mixed with 1.54 μl of formamide, 0.15 μl loading buffer and 0.31 μl of Genescan TAMRA internal standard (Applied Biosystems). Samples were denatured at 95 C for 3 min, placed on ice and 2.5 μl loaded into the pre-formed wells of a 6% denaturing polyacrylamide gel run on a ABI Prism 377 DNA Sequencer (Applied Biosystems). All fragments were automatically sized by Genescan software (Applied Biosystems) and the fluorescent product yield was calculated from integration of the peak area. PCR products were identified as coloured peaks dependent on the fluorescent dye used: TET (green), HEX (black) and 6-FAM (blue). Results Identification of highly polymorphic Ch 21 microsatellite markers Fluorescent primers spanning the repeat regions were designed for six Ch 21 microsatellite markers with high heterozygosity (>0.8) and a multiplex PCR format optimized on low template DNA (2 8 ng) to simultaneously amplify the respective alleles of each microsatellite marker. In order to identify which markers had a high probability of producing a tri-allelic pattern in a trisomic embryo, buccal cell DNA from 20 couples was genotyped (data not shown). Analysis of the DNA fingerprints enabled microsatellite markers to be evaluated with respect to the probability that the partner s alleles were different. The microsatellite markers that displayed the broadest allelic size ranges (D21S1442 and D21S1411) were also the markers that showed the greatest potential for observing a tri-allelic pattern (70%). D21S1437, D21S1413 and D21S11 were slightly less informative, with a 50 60% chance of observing a tri-allelic pattern. However, a tri-allelic pattern would be observed only 30% of the time for marker D21S1412, rendering it the least informative. The final marker set selected for DNA fingerprinting, therefore, comprised the five most informative for a tri-allelic pattern: D21S1437, D21S1413, D21S1442, D21S11 and D21S1411. Validation of the performance of the multiplex PCR at the single cell level The pentaplex DNA fingerprinting system was optimized in a multiplex FL-PCR reaction on single buccal cells by manipulating primer pair concentrations to produce strong and consistent amplification of all alleles (Figure 1a). The optimized single-cell Ch 21 DNA fingerprinting system was applied to 100 individual buccal cells (50 cells diploid for chromosome 21 and 50 cells from a patient with Down s syndrome) to assess reliability and accuracy. The diploid status of chromosome 21 was represented by di-allelic patterns (ratio 1:1) for three of the five microsatellite markers and monoallelic (homozygous) patterns for the other two microsatellite markers (see Figure 1a, for example). The trisomy status of chromosome 21 was represented by tri-allelic patterns for three of the five microsatellite markers, as well as double-dosage diallelic patterns (ratio ) for the remaining two microsatellite markers (see Figure 1b, for example). Reliability was defined as the number of successful allelic amplifications, calculated at 91% for the diploid cells and 93% for the trisomy 21 cells. Accuracy was defined as the number of correct allelic amplifications, calculated at 96% for the diploid cells and 94% for the trisomy cells. ADO was observed at less than 8% for both cell types, while preferential allele amplification was observed at ~10%. No false positives or false negatives were detected during the analysis of diploid and trisomy 21 buccal cells. Combined CF and Ch 21 aneuploidy detection CF is one of the most common single gene autosomal recessive disorders (~1 in 2500 births). The most common CF mutation, which accounts for 70% of the CF alleles, is a 3 bp deletion in exon 10 of the CF transmembrane conductance regulator gene, named δ F508 (Goossens et al., 2000). As allelic sizing can separate these two alleles, δf508 mutation detection is therefore compatible with the DNA fingerprinting format and thus was selected as the first mutation candidate for combined diagnosis. A multiplex combining the CF δf508 primer pairs with four compatible Ch 21 microsatellite markers (D21S1413, D21S11, D21S1411 and D21S1412) was optimized. Analysis of 25 buccal cells from a carrier of the CF δf508 mutation indicated comparable high reliability and accuracy (>90%) and low background interference, as well as low ADO and preferential allele amplification rates (<10%), as observed with the pentaplex Ch 21 DNA fingerprinting system (data not shown). The CF carrier status was represented by two alleles sized at 153 bp and 156 bp, whilst the diploid status of chromosome 21 was represented by di-allelic patterns (ratio 1:1) for D21S1413 and D21S1412 (see Figure 1c, for example). 45

4 a. b. c. Figure 1. DNA fingerprints of single buccal cells: (a) a diploid subject; (b) a subject with Down s syndrome; (c) a diploid subject heterozygous for CF δf Figure 2. Strategy for the analysis of sister blastomeres from aneuploid embryos.

5 a. b. Figure 3. Monosomy 21 aneuploid embryo: (a) DNA fingerprint of a sister blastomere, showing a consistent mono-allelic pattern across all five microsatellite markers; (b) fluorescence in-situ hybridization (FISH) of a sister blastomere, showing only one green signal for chromosome 21. a. b. Figure 4. Trisomy 21 aneuploid embryo: (a) DNA fingerprint of a sister blastomere, showing a tri-allelic pattern for four of the five microsatellite markers. D21S1413 displayed only a mono-allelic pattern. (b) fluorescence in-situ hybridization (FISH) of a sister blastomere, showing three green signals for chromosome

6 Diagnosis of Ch 21 aneuploidy in IVF embryos by DNA fingerprinting Nine Ch 21 aneuploid embryos identified by FISH were left to succumb on the bench for 24 h. Individual blastomeres were disassociated, coded and DNA fingerprinted (Figure 2). Allelic profiles were generated and decoded, and results from FISH and PCR were combined. Figures 3 and 4 depict examples of two sister blastomeres analysed by these two different techniques. In one case, both techniques indicated that this embryo has only one copy of Ch 21, defined by the single green FISH signal and a consistent mono-allelic pattern across all five microsatellite loci by PCR (Figure 3). In the second case, both techniques indicated that this embryo had three copies of Ch 21, defined by three green FISH signals and a tri-allelic pattern for four of the five microsatellite markers (Figure 4). A total of 67 blastomeres were disassociated from the nine Ch 21 aneuploid embryos and fingerprinted. Of these, 11 failed to amplify, indicating either the absence of a nucleus (some of the embryos were highly fragmented) or failure to transfer the cell into the PCR tube successfully. Out of a possible 280 loci, 232 loci successfully amplified, resulting in 83% reliability. Table 1 summarizes the FISH/PCR results of these Ch 21 aneuploid embryos. Analysis of all sister blastomeres from embryos 1, 3, 4 and 6 revealed consistent and concordant results. Embryo 7 appeared to be a diploid mosaic, with the majority of the sister blastomeres being diploid, whereas the remaining embryos (2, 5, 8 and 9) were aneuploid mosaics, with more than half of the sister blastomeres being aneuploid. These results suggested that different types of errors have occurred during cell division in these embryos, resulting in sister blastomeres with different Ch 21 chromosomal complements. Discussion A reliable and accurate Ch 21 single-cell PCR DNA fingerprinting system based on the fluorescent amplification of Ch 21 microsatellite markers was successfully developed. Extensive analysis of six potential markers on low template DNA extracted from 20 consenting couples identified five informative markers suitable for aneuploidy detection of Ch 21. These informative markers displayed high heterozygosity, broad distribution of alleles, identifiable allelic size ranges and low stutter artefact (smaller allelic peaks, which are thought to be generated by slippage of the DNA polymerase). A pentaplex single-cell DNA fingerprinting was successfully adapted to single buccal cells and blastomeres for the detection of Ch 21 aneuploidy. The DNA fingerprints obtained showed strong allelic amplification, with virtually no non-specific background interference to confound interpretation of allelic status. A high reliability and accuracy rate was observed (>91%) for this Ch 21 single-cell DNA fingerprinting system, together with low ADO and preferential allele amplification rates (~8 10%) on single buccal cells from both a normal and Down s syndrome subject. The performance of this system is significantly better than previously published systems (Sherlock et al., 1998; Findlay et al., 1998, 1999), which reported ADO rates as high as 56%, preferential allele amplification rates as high as 25%, and reliability rates of multiplex systems in the order of 75% for buccal cells. However, the 83% reliability of this single-cell DNA fingerprinting system on blastomeres was significantly lower than the >91% obtained from buccal cells (chi-squared test, P < 0.05). The blastomeres in this study were dissociated from arrested aneuploid embryos that were left to succumb on the bench for up to 24 h. Previous studies of arrested and fragmented embryos consistently demonstrated much lower PCR amplification efficiencies compared with freshly isolated cells (Ray et al., 1998; Findlay et al., 1999). Therefore, higher amplification efficiency on fresh biopsied blastomeres from clinical PGD cases would be expected with this DNA fingerprinting system. A diploidy status at any particular microsatellite marker locus was determined by the presence of two alleles with an expected allelic ratio of 1:1. Monosomy 21 was defined by a consistent mono-allelic pattern across all five microsatellite markers, while trisomy 21 was diagnosed by a tri-allelic pattern or a double dosage di-allelic pattern. Unfortunately it is impossible to distinguish a double dosage di-allelic pattern from the occurrence of preferential allele amplification. Nevertheless, since the calculated preferential allele amplification rate for this Ch 21 single cell DNA fingerprinting system is only in the order of 10%, in a situation where three or more microsatellite markers display a double dosage di-allelic pattern, it is more Table 1. Chromosome 21 status of IVF embryos by fluorescence in-situ hybridization (FISH) and polymerase chain reaction (PCR). Embryo PGD diagnosis (FISH) DNA fingerprints Embryo status 1 1 cell (3x21) 4 cells (3x21), 1 cell a Trisomy cell (1x21) 3 cells (1x21), 2 cells (2x21), 2 cells a Aneuploid mosaic cell (1x21) 8 cells (1x21) Monosomy cell (3x21) 5 cells (3x21) Trisomy cell (3x21) 3 cells (3x21), 1 cell (1x21), 4 cells (2x21), 2 cells a Aneuploid mosaic cell (1x21) 6 cells (1x21) Monosomy cell (3x21) 4 cells (2x21), 1 cell a Diploid mosaic cell (1x21) 4 cells (3x21), 4 cells (2x21), 2 cells a Aneuploid mosaic cells (1x21) 1 cell (3x21), 1 cell (1x21), 2 cells (2x21) Aneuploid mosaic a No DNA

7 likely that this is due to an extra copy of chromosome 21 rather than all the markers being affected by preferential allele amplification. By successfully combining the CF δ F508 mutation detection and a Ch 21 single-cell DNA fingerprinting system, it is now possible to offer PGD couples at genetic risk for both CF δ F508 and trisomy 21 a simultaneous diagnosis on a single biopsied blastomere. These results also indicate the possibility of combining other single gene mutation detection systems with a Ch 21 single-cell DNA fingerprinting system. For example, previous preliminary studies suggest that allele refractory mutation system (ARMS)-PCR can be applied to single cells for the diagnosis of point mutations (Sherlock et al., 1998). Considering that many PGD couples are presenting with ARA and would like to reduce the requirement for PND, a test to diagnose both the single gene disorder and Ch 21 aneuploidy would be of enormous benefit. In future studies, the present authors are aiming to combine single gene mutation detection with a more complex DNA fingerprinting system to include aneuploidy detection of chromosomes X, Y, 13, 18 and 21. Embryonic chromosomal mosaicism has been observed in over 50% of all human cleavage stage embryos (Delhanty, 1994; Munné et al., 1994; Harper et al., 1995; Kuo et al., 1998; Iwarsson et al., 1999; Bielanska et al., 2000; Magli et al., 2000; Márquez et al., 2000; Voullaire et al., 2000; Wells and Delhanty, 2000). These mosaic embryos can be classified as either diploid mosaic, in which the majority of the blastomeres are diploid, or aneuploid mosaics, in which the majority of the blastomeres are aneuploid. From the results of the aneuploid embryos analysed in this study, concordance between sister blastomeres was observed in four of the nine aneuploid embryos. Only one aneuploid embryo was defined as a diploid mosaic, while the remaining four were aneuploid mosaics (44%). It would appear that several different types of errors have occurred during the first few cell divisions of these mosaic embryos, including mitotic non-disjunction. These results are similar to what has been observed in previous studies investigating chromosomal status of sister blastomeres from aneuploid embryos. However, DNA fingerprinting also enables the identification of the origin and nature of the chromosomal aneuploidy. A continuation of this study, involving DNA fingerprinting of a larger number of embryos and sister blastomeres, is essential to establish the origin and nature of aneuploidy and embryonic mosaicism in cleavage stage human embryos. As well as specific aneuploid detection, single-cell DNA fingerprinting has other unique advantages, including confirmation of parental allelic contribution to the embryo and the identification of extraneous DNA contamination that could cause a misdiagnosis in PGD cases (Piyamongkol et al., 2001). Considering a single cell contains only 6 pg of DNA and two target sequences per locus, any external DNA contamination can be detrimental. DNA fingerprinting is the ultimate method of biological individualization, enabling the identification of the origin of any DNA sample. It is also possible to identify each sibling embryo in a cohort by its unique allelic DNA fingerprint, and thus it would be possible to track IVF embryos from the time of transfer to term. Tracking of IVF embryos to term would be powerful technology, enabling the viable embryo from a multiple transfer that produced the pregnancy to be identified. With this knowledge it could be possible to correlate embryonic viability with morphological, biochemical or genetic indices. A marker for embryonic viability could lead to an increase in IVF pregnancy rates and the possibility of only singleton embryo transfers to reduce the incidence of multiple gestations. Acknowledgments This research was funded in part by a grant from Monash IVF Pty Ltd. References Antonarakis SE, Petersen MB, McInnis MG et al The meiotic stage of non-disjunction in trisomy 21: Determination by using DNA polymorphisms. American Journal of Human Genetics 50, Bielanska MM, Tan SL, Ao A 2000 Significance of mosaicism in the human preimplantation embryo. Human Reproduction 15 (Abstract Book 1), 12. Blake D, Tan SL, Ao A 1999 Assessment of multiplex fluorescent PCR for screening single cells for trisomy 21 and single gene defects. Molecular Human Reproduction 5, Boué A, Boué J, Grapp A 1985 Cytogenetics of pregnancy wastage. Advances in Human Genetics 14, Cirigliano V, Ejarque M, Canadas MP et al Clinical application of multiplex quantitative fluorescent polymerase chain reaction (QF-PCR) for the rapid prenatal detection of common chromosome aneuploidies. Molecular Human Reproduction 7, Coonen E, Hopman AH, Geraedts JP et al Application of in-situ hybridization techniques to study human preimplantation embryos: a review. Human Reproduction Update 4, Damario MA, Davis OK, Rosenwaks Z 1999 The role of maternal age in the assisted reproductive technologies. Reproductive Medicine Review 7, Delhanty JD 1994 Preimplantation diagnosis. Prenatal Diagnosis 14, Delhanty JD, Harper JC 2000 Pre-implantation genetic diagnosis. Baillieres Best Practice Research. Clinical Obstetrics and Gynaecology 14, Findlay I, Matthews P, Quirke P 1999 Preimplantation genetic diagnosis using fluorescent polymerase chain reaction: results and future developments. Journal of Assisted Reproduction and Genetics 16, Findlay I, Toth T, Matthews P et al Rapid trisomy diagnosis (21, 18 and13) using fluorescent PCR and short tandem repeats: applications for prenatal diagnosis and preimplantation genetic diagnosis. Journal of Assisted Reproduction and Genetics 15, Findlay I, Urquhart A, Quirke P et al Simultaneous DNA fingerprinting, diagnosis of sex and single-gene defect status from single cells. Molecular Human Reproduction 10, Gianaroli L, Magli MC, Ferraretti AP et al Preimplantation diagnosis for aneuploidies in patients undergoing in vitro fertilization with a poor prognosis: identification of the categories for which it should be proposed. Fertility and Sterility 72, Goossens V, Sermon K, Lissens W et al. (2000) Clinical application of preimplantation genetic diagnosis for cystic fibrosis. Prenatal Diagnosis 20, Harper JC, Coonen E, Handyside AH et al Mosaicism of autosomes and sex chromosomes in morphologically normal monospermic preimplantation human embryos. Prenatal Diagnosis 15, Hook EB 1981 Rates of chromosome abnormalities at different maternal ages. Obstetrics and Gynecology 58, Hook EB, Cross PK 1983 Spontaneous abortion and subsequent 49

8 Down syndrome livebirth. Human Genetics 64, Iwarsson E, Lundqvist M, Inzunza J et al A high degree of aneuploidy in frozen-thawed human preimplantation embryos. Human Genetics 104, Kuo HC, Ogilvie CM, Handyside AH 1998 Chromosomal mosaicism in cleavage-stage human embryos and the accuracy of single-cell genetic analysis. Journal of Assisted Reproduction and Genetics 15, Magli MC, Jones GM, Gras L et al Chromosome mosaicism in day 3 aneuploid embryos that develop to morphologically normal blastocysts in vitro. Human Reproduction 15, Mansfield ES 1993 Diagnosis of Down syndrome and other aneuploidies using quantitative polymerase chain reaction and small tandem repeat polymorphisms. Human Molecular Genetics 2, Márquez C, Sandalinas M, Bahçe M et al Chromosome abnormalities in 1255 cleavage-stage human embryos. Reproductive BioMedicine Online 1, Mikkelsen M, Hallberg A, Poulsen H et al Epidemiological study of Down s syndrome in Denmark, including family studies of chromosomes and DNA markers. Developmental Brain Dysfunction 8, Mulvey S, Wallace EM 2001 Levels of knowledge of Down syndrome and Down syndrome testing in Australian women. Australian and New Zealand Journal of Obstetrics and Gynaecology 41, Munné S, Magli C, Cohen J et al Positive outcome after preimplantation diagnosis of aneuploidy in human embryos. Human Reproduction 14, Munné S, Weier H, Grifo J et al Chromosome mosaicism in human embryos. Biology of Reproduction 51, Pertl B, Yau SC, Sherlock J et al Rapid molecular method for prenatal detection of Down s syndrome. Lancet 14, Petersen MB, Mikkelsen M 2000 Nondisjunction in trisomy 21: Origin and mechanisms. Cytogenetics and Cellular Genetics 91, Piyamongkol W, Harper JC, Delhanty JD et al Preimplantation genetic diagnostic protocols for alpha- and beta-thalassaemias using multiplex fluorescent PCR. Prenatal Diagnosis 21, Ray PF, Ao A, Taylor DM et al Assessment of the reliability of single blastomere analysis for preimplantation diagnosis of the DF508 deletion causing cystic fibrosis in clinical practice. Prenatal Diagnosis 18, Report of the 11th Annual Meeting of International Working Group on Preimplantation Genetics 2001 Preimplantation Genetic Diagnosis: experience of 3000 clinical cycles. Reproductive BioMedicine Online 3, Sherlock J, Cirigliano V, Petrou M et al Assessment of diagnostic quantitative fluorescent multiplex polymerase chain reaction assays performed on single cells. Annals of Human Genetics 62, Tsai YH (1999) Cost-effective one-step PCR amplification of cystic fibrosis delta F508 fragment in a single cell for preimplantation genetic diagnosis. Prenatal Diagnosis 19, Verlinsky Y, Kuliev A 1999 Preimplantation Genetic Diagnosis. Reproductive Medicine Review 7, Verlinsky Y, Cieslak J, Ivakhnenko V et al Implantation diagnosis of common aneuploidies by the first and second-polar body FISH analysis. Journal of Assisted Reproduction and Genetics 15, Voullaire L, Slater H, Williamson R et al Chromosome analysis of blastomeres from human embryos by using comparative genomic hybridization. Human Genetics 106, Wells D, Delhanty JD 2000 Comprehensive chromosomal analysis of human preimplantation embryos using whole genome amplification and single cell comparative genomic hybridization. Molecular Human Reproduction 6,