Preimplantation genetic diagnosis for single gene disorders: experience with five single gene disorders Joyce C. Harper 1 *, Dagan Wells 1 , Wirawit Piyamongkol 1 , Patrick Abou-Sleiman 1 , Angela Apessos 1 , Antonis Ioulianos 1 , Mary Davis 2 , Alpesh Doshi 3 , Paul Serhal 3 , Massimo Ranieri 3 , Charles Rodeck 1 and Joy D. A. Delhanty 1 1 Department of Obstetrics and Gynaecology, University College London, London, UK 2 Institute of Neurology, University College London, London, UK 3 Assisted Conception Unit, University College London, London, UK We report our experience of 14 preimplantation genetic diagnosis (PGD) cycles in eight couples carrying five different single gene disorders, during the last 18 months. Diagnoses were performed for myotonic dystrophy (DM), cystic fibrosis (CF) [DF508 and exon 4 (621+1G>T)], fragile X and CF simultaneously, and two disorders for which PGD had not been previously attempted, namely neurofibromatosis type 2 (NF2) and Crouzon syndrome. Diagnoses for single gene disorders were carried out on ideally two blastomeres biopsied from Day 3 embryos. A highly polymorphic marker was included in each diagnosis to control against contamination. For the dominant disorders, where possible, linked polymorphisms provided an additional means of determining the genotype of the embryo hence reducing the risk of misdiagnosis due to allele dropout (ADO). Multiplex fluorescent polymerase chain reaction (F-PCR) was used in all cases, followed by fragment analysis and/or single-stranded conformation polymorphism (SSCP) for genotyping. Embryo transfer was performed in 13 cycles resulting in one biochemical pregnancy for CF, three normal deliveries (a twin and a singleton) and one early miscarriage for DM and a singleton for Crouzon syndrome. In each case the untransferred embryos were used to confirm the diagnoses performed on the biopsied cells. The results were concordant in all cases. The inclusion of a polymorphic marker allowed the detection of extraneous DNA contamination in two cells from one case. Knowing the genotype of the contaminating DNA allowed its origin to be traced. All five pregnancies were obtained from embryos in which two blastomeres were biopsied for the diagnosis. Our data demonstrate the successful strategy of using multiplex PCR to simultaneously amplify the mutation site and a polymorphic locus, fluorescent PCR technology to achieve greater sensitivity, and two-cell biopsy to increase the efficiency and success of diagnoses. Copyright # 2002 John Wiley & Sons, Ltd. KEY WORDS: PGD; single gene; PCR INTRODUCTION For more than a decade preimplantation genetic diagnosis (PGD) has been used as an alternative to prenatal diagnosis for a small proportion of patients at risk of transmitting an inherited disease to their children (Wells and Delhanty, 2001; Delhanty and Harper, 2002). In the majority of centres offering PGD worldwide, embryo biopsy is performed on Day 3, with the removal of one or two cells from the cleavage- stage embryo (ESHRE PGD Consortium Steering Committee, 1999, 2000, 2002). Since performing a diagnosis on a single cell is technically difficult, some centres prefer to make a diagnosis when two cells are available (Van de Velde et al., 2000; De Vos and Van Steirteghem, 2001). For the analysis of single gene defects, the polymerase chain reaction (PCR) is used (Wells and Sherlock, 1998). For the analysis of chro- mosomes (Conn et al., 1998; Munne ´ et al., 1998a; Van Assche et al., 1999; Fridstrom et al., 2001) and to determine embryo sex (Harper et al., 1994; Staessen et al., 1999), interphase fluorescent in situ hybridis- ation (FISH) is the method of choice. More recently, the PGD procedure has been used as a form of preimplantation aneuploidy screening (PGD-AS) to try to improve in vitro fertilisation (IVF) pregnancy rates (Munne ´ et al., 1998b; Verlinsky et al., 1998). PGD and PGD-AS have been hampered by the reports of several misdiagnoses (reviewed in Harper and Delhanty, 2000). In the first series of PGD per- formed in the late 1980s a misdiagnosis of sex occurred. In this case diagnosis had relied on the efficient amplification of a Y chromosome-specific sequence from all male embryos (Handyside et al., 1990). Since this time, phenomena such as allele dropout (ADO), contamination, chromosomal mosaicism and hybrid- isation failure have revealed PGD to be more techni- cally difficult than was originally imagined. ADO and contamination are important for PCR diagnosis (Wells and Sherlock, 1998; Sermon, 2002). ADO is observed when only one of the two alleles in a heterozygous cell is detected after PCR (i.e. ampli- fication failure affecting just one allele) (Ray and Handyside, 1996). In most circumstances this does not *Correspondence to: J. C. Harper, Department of Obstetrics and Gynaecology, University College London, 86–96 Chenies Mews, London WC1E 6HX, UK. E-mail: [email protected] PRENATAL DIAGNOSIS Prenat Diagn 2002; 22: 525–533. Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002 / pd.394 Copyright # 2002 John Wiley & Sons, Ltd. Received: 1 February 2002 Accepted: 27 March 2002
Preimplantation genetic diagnosis for single gene disorders: experience with five single gene disorders
Jan 14, 2023
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
PND12404 525..533Joyce C. Harper1*, Dagan Wells1, Wirawit Piyamongkol1, Patrick Abou-Sleiman1, Angela Apessos1, Antonis Ioulianos1, Mary Davis2, Alpesh Doshi3, Paul Serhal3, Massimo Ranieri3, Charles Rodeck1 and Joy D. A. Delhanty1
1Department of Obstetrics and Gynaecology, University College London, London, UK 2Institute of Neurology, University College London, London, UK 3Assisted Conception Unit, University College London, London, UK
We report our experience of 14 preimplantation genetic diagnosis (PGD) cycles in eight couples carrying five different single gene disorders, during the last 18 months. Diagnoses were performed for myotonic dystrophy (DM), cystic fibrosis (CF) [DF508 and exon 4 (621+1 G>T)], fragile X and CF simultaneously, and two disorders for which PGD had not been previously attempted, namely neurofibromatosis type 2 (NF2) and Crouzon syndrome. Diagnoses for single gene disorders were carried out on ideally two blastomeres biopsied from Day 3 embryos. A highly polymorphic marker was included in each diagnosis to control against contamination. For the dominant disorders, where possible, linked polymorphisms provided an additional means of determining the genotype of the embryo hence reducing the risk of misdiagnosis due to allele dropout (ADO). Multiplex fluorescent polymerase chain reaction (F-PCR) was used in all cases, followed by fragment analysis and/or single-stranded conformation polymorphism (SSCP) for genotyping. Embryo transfer was performed in 13 cycles resulting in one biochemical pregnancy for CF, three normal deliveries (a twin and a singleton) and one early miscarriage for DM and a singleton for Crouzon syndrome. In each case the untransferred embryos were used to confirm the diagnoses performed on the biopsied cells. The results were concordant in all cases. The inclusion of a polymorphic marker allowed the detection of extraneous DNA contamination in two cells from one case. Knowing the genotype of the contaminating DNA allowed its origin to be traced. All five pregnancies were obtained from embryos in which two blastomeres were biopsied for the diagnosis. Our data demonstrate the successful strategy of using multiplex PCR to simultaneously amplify the mutation site and a polymorphic locus, fluorescent PCR technology to achieve greater sensitivity, and two-cell biopsy to increase the efficiency and success of diagnoses. Copyright # 2002 John Wiley & Sons, Ltd.
KEY WORDS: PGD; single gene; PCR
INTRODUCTION
For more than a decade preimplantation genetic diagnosis (PGD) has been used as an alternative to prenatal diagnosis for a small proportion of patients at risk of transmitting an inherited disease to their children (Wells and Delhanty, 2001; Delhanty and Harper, 2002). In the majority of centres offering PGD worldwide, embryo biopsy is performed on Day 3, with the removal of one or two cells from the cleavage- stage embryo (ESHRE PGD Consortium Steering Committee, 1999, 2000, 2002). Since performing a diagnosis on a single cell is technically difficult, some centres prefer to make a diagnosis when two cells are available (Van de Velde et al., 2000; De Vos and Van Steirteghem, 2001). For the analysis of single gene defects, the polymerase chain reaction (PCR) is used (Wells and Sherlock, 1998). For the analysis of chro- mosomes (Conn et al., 1998; Munne et al., 1998a; Van Assche et al., 1999; Fridstrom et al., 2001) and to
determine embryo sex (Harper et al., 1994; Staessen et al., 1999), interphase fluorescent in situ hybridis- ation (FISH) is the method of choice. More recently, the PGD procedure has been used as a form of preimplantation aneuploidy screening (PGD-AS) to try to improve in vitro fertilisation (IVF) pregnancy rates (Munne et al., 1998b; Verlinsky et al., 1998).
PGD and PGD-AS have been hampered by the reports of several misdiagnoses (reviewed in Harper and Delhanty, 2000). In the first series of PGD per- formed in the late 1980s a misdiagnosis of sex occurred. In this case diagnosis had relied on the efficient amplification of a Y chromosome-specific sequence from all male embryos (Handyside et al., 1990). Since this time, phenomena such as allele dropout (ADO), contamination, chromosomal mosaicism and hybrid- isation failure have revealed PGD to be more techni- cally difficult than was originally imagined.
ADO and contamination are important for PCR diagnosis (Wells and Sherlock, 1998; Sermon, 2002). ADO is observed when only one of the two alleles in a heterozygous cell is detected after PCR (i.e. ampli- fication failure affecting just one allele) (Ray and Handyside, 1996). In most circumstances this does not
*Correspondence to: J. C. Harper, Department of Obstetrics and Gynaecology, University College London, 86–96 Chenies Mews, London WC1E 6HX, UK. E-mail: [email protected]
PRENATAL DIAGNOSIS
Prenat Diagn 2002; 22: 525–533. Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002 /pd.394
Copyright # 2002 John Wiley & Sons, Ltd. Received: 1 February 2002 Accepted: 27 March 2002
pose a major problem for single cell PCR, but it is very important for dominant disorders, where an affected embryo carries only one copy of the mutant allele. In this case if the affected allele does not amplify, the embryo will be diagnosed as normal, when actually it is affected. Contamination is also a major problem for all PCR diagnoses. Sperm and cumulus cells embedded in the zona pellucida of the embryo may become dislodged during the biopsy procedure and therefore lead to paternal or maternal contamination, respectively. To overcome this problem, intracyto- plasmic sperm injection (ICSI) is used to achieve fer- tilisation for PCR cycles and the cumulus cells must be carefully removed for all PGD cases. It is possible that paternal or maternal contamination has caused some of the misdiagnoses reported (Sermon et al., 1998; Harper and Delhanty, 2000). To overcome the problem of contamination, either maternal, paternal or from other sources, it has been advisable to per- form multiplex PCR reactions, including ideally linked (Rechitsky et al., 1998; Dreesen et al., 2000; Piyamongkol et al., 2001a,b) or unlinked polymorphic markers, which are informative for the family under- going PGD. For these markers the mother and father should have four different alleles, so when examining the embryo it should have inherited one allele from the father and one from the mother. Any other pattern would indicate the presence of additional DNA, suggesting that contamination had occurred. The use of fluorescent PCR is ideal for multiplex PCR from a single cell.
Here we report on the use of PCR for PGD of five different single gene disorders: myotonic dystrophy (DM), cystic fibrosis (CF) [DF508 and exon 4 (621+ 1 G>T)], fragile X and CF simultaneously, and two disorders for which PGD had not been previously attempted, namely neurofibromatosis type 2 (NF2) and Crouzon syndrome.
DM or Steinert’s disease is a progressive autosomal dominant muscular dystrophy. It is caused by an unstable CTG repeat expansion within exon 15 of the DMPK gene on chromosome 19q. Expanded (mutant) CTG repeat sequences are refractory to conventional PCR, but alleles with a number of repeats within the normal range can be readily amplified and detected. PGD, with analysis based on the detection of the nor- mal alleles, has been successfully performed (Sermon et al., 1997, 2001) but a misdiagnosis was reported (Sermon et al., 1998) which could have been due to maternal contamination. Therefore, we have devel- oped two new PGD protocols for DM, which utilise multiplex fluorescent PCR (Piyamongkol et al., 2001a,b). Ideally a linked polymorphic marker, APOC2, is ampli- fied in addition to the normal DMPK alleles, thus providing a back-up diagnostic result. We report on PGD for three couples where the females were carriers using a single-step duplex fluorescent (F)- PCR which will detect the DM triplet repeat with normal alleles. However, two of the couples were not fully informative at the APOC2 locus and so an unlinked STR marker, D21S1414, was substituted.
This locus provides no direct diagnostic information, but is still useful for contamination detection.
CF, an autosomal recessive disease caused by mu- tation in the CFTR gene, is located on chromosome 7q and was the first single gene defect for which PGD was offered (Handyside et al., 1992). The most common mutation in the Caucasian population is a three base pair deletion known as DF508 and the majority of PGD cycles for CF have been for this mutation (ESHRE PGD Consortium Steering Committee, 1999). Here we report on PGD for this mutation using a single- step triplex F-PCR with primers for the mutation, a linked marker and a contamination marker. A couple also presented with both partners carrying a mu- tation in exon 4 of CFTR (621+1 G>T). A triplex PCR with direct mutation detection, a linked marker and a contamination marker was used using nested PCR and SSCP for the mutation detection.
Fragile X syndrome is the most common cause of familial mental retardation. The most common mu- tation is an expansion of a triplet (CGG)n repeat in the 5k untranslated region of the FMR1 gene on Xq27.3. The expansion is refractory to PCR due to preferential amplification of the smaller allele in heterozygous cells and the high GC content of the repeat and surround- ing sequences. Detection of the normal and premu- tation alleles has been proposed for use in PGD of fragile X syndrome (Black et al., 1995, Sermon et al., 1999, 2001), an approach which has also been used in the diagnosis of two other triplet repeat disorders, DM (Sermon et al., 1997) and Huntington’s disease (Sermon et al., 1998). We investigated the use of linked polymorphic markers flanking the mutation to track the normal and premutation carrying maternal chro- mosomes in preimplantation embryos (Apessos et al., 2001). One couple presented where the female was carrying a large FMR1 premutation for fragile X syndrome and the couple were both carriers for CF, DF508. A single-step tetraplex F-PCR was performed employing amplification of two linked markers flank- ing FMR1, a sequence from the amelogenin gene for sexing, and a DNA fragment encompassing the DF508 mutation.
Mutations in the NF2 gene predispose to neuro- fibromatosis type 2 (NF2), a dominantly inherited cancer predisposition syndrome with an incidence of 1 in 33 000–40 000 (Evans et al., 1992). NF2 is charac- terised by the development of histologically benign tumours in the central nervous system. A couple pre- sented where the male carried a splice site mutation in intron 4: 514-2 (A to G) of NF2. The couple needed fertility treatment due to ovulatory problems in the female. The molecular diagnosis was performed by duplex F-PCR, followed by analysis of an informative intragenic single nucleotide polymorphism (SNP) and detection of the causative mutation using SSCP (Abou-Sleiman et al., 2002a).
Crouzon syndrome is a dominantly inherited cranio- synostosis syndrome that is caused by mutations in the fibroblast growth factor receptor 2 gene (FGFR2) on chromosome 10q (Reardon et al., 1994). It has an estimated birth prevalence of 15–16 per 1 million
J. C. HARPER ET AL.526
Copyright # 2002 John Wiley & Sons, Ltd. Prenat Diagn 2002; 22: 525–533.
births (Cohen and Kreiborg, 1992). The spectrum of mutations in FGFR2 is relatively limited compared to other genes. To date only 46 mutations have been identified, the majority of which are missense, with a smaller number of splice mutations or small insertions/ deletions, all of which remain in-frame. For PGD we developed DNA amplification using a multiplex nested PCR, for the simultaneous amplification of a fragment of the FGFR2 gene encompassing the mutation and a highly polymorphic short tandem repeat (STR) locus (unlinked, D21S11). SSCP was used to detect the mutation. The genotype of the embryos at the STR was determined by fragment analysis on an automated DNA sequencer by fluorescent polyacrylamide gel electrophoresis (PAGE). Two intragenic SNP had been described in FGFR2 at the time of preparation for the diagnosis. However, neither was suitable for use in this case as the parents were uninformative (Abou Sleiman et al., 2002b).
Here we report on 14 PGD cycles for eight patients carrying five single gene defects, using multiplex F-PCR incorporating polymorphic markers for con- tamination detection.
PATIENTS AND METHODS
All patients had previous genetic counselling before presenting to the PGD centre. Two IVF/PGD consul- tations were conducted during which the limitations of the PGD procedure were outlined, such as the need for IVF, risk of misdiagnosis, possibility that all embryos could be affected, PGD pregnancy rates, etc. Blood, DNA or buccal cell samples [obtained by scraping the inside of the cheek with a sterile wooden stick with a cotton swab tip and suspending in 3 ml phosphate- buffered saline (PBS) 4% (v/v) bovine serum albumin] were obtained from all couples and affected relatives prior to treatment. This enabled the work-up of a specific diagnosis for each couple, i.e. analysis of linked or unlinked markers for contamination detec- tion, development of multiplex PCR using these markers and optimisation of the mutation detection. The necessary investigations required prior to an IVF cycle were conducted, i.e. G test (Ranieri et al., 2001), HyCoSy, dummy embryo transfer, semen analysis, etc.
The time taken to work-up a specific diagnosis varied from 6 months to over a year. All work-ups were conducted on DNA, single buccal cells from the parents and affected relatives, and single blastomeres donated from our IVF programme. The results from the work-ups were submitted to the Human Fertilis- ation and Embryology Authority (HFEA) for licen- sing purposes and permission was granted for each disorder.
The patients underwent routine IVF procedures as outlined previously (Ranieri et al., 2001). ICSI was performed in all cases to reduce the risk of sperm contamination. Oocytes and embryos were cultured in IVF medium (Cook IVF, Brisbane, Australia). Prior to biopsy, all cumulus cells were mechanically removed to reduce the risk of maternal contamination. On Day
3, embryo biopsy was performed on all embryos that were considered suitable (depending on number of cells, grade and the number of embryos available), in Ca2+Mg2+-free embryo biopsy medium (Medicult, Surrey, UK), using Research Instrument (Cornwall, UK) micromanipulators. Zona drilling was performed using acid Tyrode’s solution as described previously (Piyamongkol et al., 2001a). Two blastomeres were removed from embryos containing six cells or more by aspiration and washed in PBS. In cases where only a small number of 6–8-cell embryos are available, we remove a single cell from embryos with less than six cells. However, in all the cases reported here, embryos were transferred after a normal result was obtained from two cells. All cells were washed at least three times in droplets of PBS/BSA before their transfer in 2 ml PBS/BSA to microcentrifuge tubes containing 3 ml 125 mg/ml proteinase K and 4r10x4% (w/v) sodium dodecyl sulphate (SDS). A control blank was pro- cessed for each blastomere using 2 ml of fluid from the final PBS wash drop. Cells were lysed by incubation at 37uC for 1 h and the proteinase K inactivated by incubation for 15 min at 95uC. After biopsy, embryos were stored in Cook extended culture medium.
A polymorphic marker was included in each diag- nosis to control against contamination. For the dominant disorders, where possible, linked poly- morphisms provided an additional means of determin- ing the genotype of the embryo hence reducing the risk of misdiagnosis due to ADO. Multiplex PCR was used in all cases, followed by fragment analysis using fluorescent DNA sequencers and/or SSCP and silver staining for genotyping. All diagnoses were performed within 24 h and embryo transfers undertaken on Day 4 post-insemination. Untransferred embryos were used to check the original diagnosis.
DM
The method and details of Patients 1 and 2 have been reported previously (Piyamongkol et al., 2001a). In brief, amplification of the expanded allele is not possible as it is refractory to single cell PCR. Therefore the diagnosis depends on the identification of the normal alleles (Brook et al., 1992). In the three patients, the females carried an expanded allele and so it was important to check for maternal contami- nation. We have previously reported on the use of the linked marker APOC2, but this marker was only informative for one of our PGD couples (Piyamongkol et al., 2001a,b). Therefore an unlinked STR marker, D21S1414 (which is on chromosome 21), was sub- stituted for contamination detection (Sherlock et al., 1998) for two couples. The multiplex amplified pro- ducts from single cells were each tagged with two dif- ferent fluorochromes using labelled primers. This allowed analysis to be performed on an automated laser fluorescent sequencer (ALFExpress1) (Pharmacia Biotech, Herts, UK) and also an ABI Prism1310 (PE Applied Biosystems, Warrington, UK).
PGD FOR SINGLE GENE DISORDERS 527
Copyright # 2002 John Wiley & Sons, Ltd. Prenat Diagn 2002; 22: 525–533.
CF DF508
One couple presented where both partners were carrying DF508. A single-step, triplex F-PCR was employed which allowed simultaneous amplification of the mutation locus together with two polymorphisms used as internal controls. An informative STR (D13S631) and the intron 6-linked polymorphism were used to detect any contamination and ADO, respectively. Fragment (size) analysis was carried out on the ALFExpress1. The primers used for the mu- tation were as described by Handyside et al. (1992), whereas the primers for the intron 6-linked poly- morphism were modified from the original described by Chehab et al. (1991).
CF exon 4
One couple presented where both partners carried a mutation in exon 4 (621+1 G>T). A single-step triplex F-PCR was performed for the mutation locus and two polymorphisms (the informative HumTh01 STR and the intron 6-linked polymorphism) followed by a second (nested) PCR round for the mutation locus only. Fragment (size) analysis was performed on the ALFExpress1 for the two polymorphisms. The mutation was detected by silver-stained SSCP on the semi-automated GenePhor1 electrophoresis system (Pharmacia Biotech, Herts, UK) using 12.5% polyacryl- amide gels at 15uC. The primers used for the mutation were designed by Ioulianos et al. (unpublished data), whereas the primers for the intron 6-linked poly- morphism were as described by Chehab et al. (1991).
CF and fragile X
A couple presented where the female carried an expansion that ranged from the upper permutation to the full mutation (133–300 repeats) and both partners were carrying the CF mutation DF508. A single-step, tetraplex F-PCR was used with two linked markers flanking to FMR1 [dinucleotide microsatellite repeat, DXS998 and (CA)n microsatellite polymorph- ism, p39] (Wehnert et al., 1993; Apessos et al., 2001), primers for amelogenin for embryo sexing (Sullivan et al., 1993), and primers for DF508 (Liu et al., 1992).
NF2
The causative mutation was identified as a single base pair substitution affecting a splice site in intron 4 of the gene (517-2 A/G). The female partner suffered from polycystic ovary syndrome and had never ovulated spontaneously; this case has previously been reported (Abou-Sleiman et al., 2002a). Two sets of fluores- cently labelled primers were designed to amplify the region of the gene harbouring the mutation, and a fragment of the 5k untranslated end of the gene encompassing a linked single nucleotide polymorph- ism, G/C substitution at nucleotide 8240. As the linked polymorphism was found to be informative in
the couple it could be used to infer the presence or absence of the mutation providing a secondary, inde- pendent means of detecting the mutation, thus reduc- ing the risk of misdiagnosis due to ADO. A duplex F-PCR was developed with simultaneous SSCP used for the mutation and SNP.
Crouzon syndrome
The causative mutation was identified in the female partner as a de novo G/A single base pair substitution at codon 568 situated in the alternative coding domain for the 3k half of the IgIII domain (numbering based on the sequence published by Zhang et al., 1999). The patient was diagnosed with Crouzon syndrome at 16 years of age and was mildly affected. This case has previously been reported (Abou-Sleiman et al., 2002b). The primers used for the mutation were as described by Sharma and Litt (1992). The mutation was detected by silver-stained SSCP on the semi-automated Phast- System1 electrophoresis system using 20% polyacryl- amide gels at 4uC. The…
1Department of Obstetrics and Gynaecology, University College London, London, UK 2Institute of Neurology, University College London, London, UK 3Assisted Conception Unit, University College London, London, UK
We report our experience of 14 preimplantation genetic diagnosis (PGD) cycles in eight couples carrying five different single gene disorders, during the last 18 months. Diagnoses were performed for myotonic dystrophy (DM), cystic fibrosis (CF) [DF508 and exon 4 (621+1 G>T)], fragile X and CF simultaneously, and two disorders for which PGD had not been previously attempted, namely neurofibromatosis type 2 (NF2) and Crouzon syndrome. Diagnoses for single gene disorders were carried out on ideally two blastomeres biopsied from Day 3 embryos. A highly polymorphic marker was included in each diagnosis to control against contamination. For the dominant disorders, where possible, linked polymorphisms provided an additional means of determining the genotype of the embryo hence reducing the risk of misdiagnosis due to allele dropout (ADO). Multiplex fluorescent polymerase chain reaction (F-PCR) was used in all cases, followed by fragment analysis and/or single-stranded conformation polymorphism (SSCP) for genotyping. Embryo transfer was performed in 13 cycles resulting in one biochemical pregnancy for CF, three normal deliveries (a twin and a singleton) and one early miscarriage for DM and a singleton for Crouzon syndrome. In each case the untransferred embryos were used to confirm the diagnoses performed on the biopsied cells. The results were concordant in all cases. The inclusion of a polymorphic marker allowed the detection of extraneous DNA contamination in two cells from one case. Knowing the genotype of the contaminating DNA allowed its origin to be traced. All five pregnancies were obtained from embryos in which two blastomeres were biopsied for the diagnosis. Our data demonstrate the successful strategy of using multiplex PCR to simultaneously amplify the mutation site and a polymorphic locus, fluorescent PCR technology to achieve greater sensitivity, and two-cell biopsy to increase the efficiency and success of diagnoses. Copyright # 2002 John Wiley & Sons, Ltd.
KEY WORDS: PGD; single gene; PCR
INTRODUCTION
For more than a decade preimplantation genetic diagnosis (PGD) has been used as an alternative to prenatal diagnosis for a small proportion of patients at risk of transmitting an inherited disease to their children (Wells and Delhanty, 2001; Delhanty and Harper, 2002). In the majority of centres offering PGD worldwide, embryo biopsy is performed on Day 3, with the removal of one or two cells from the cleavage- stage embryo (ESHRE PGD Consortium Steering Committee, 1999, 2000, 2002). Since performing a diagnosis on a single cell is technically difficult, some centres prefer to make a diagnosis when two cells are available (Van de Velde et al., 2000; De Vos and Van Steirteghem, 2001). For the analysis of single gene defects, the polymerase chain reaction (PCR) is used (Wells and Sherlock, 1998). For the analysis of chro- mosomes (Conn et al., 1998; Munne et al., 1998a; Van Assche et al., 1999; Fridstrom et al., 2001) and to
determine embryo sex (Harper et al., 1994; Staessen et al., 1999), interphase fluorescent in situ hybridis- ation (FISH) is the method of choice. More recently, the PGD procedure has been used as a form of preimplantation aneuploidy screening (PGD-AS) to try to improve in vitro fertilisation (IVF) pregnancy rates (Munne et al., 1998b; Verlinsky et al., 1998).
PGD and PGD-AS have been hampered by the reports of several misdiagnoses (reviewed in Harper and Delhanty, 2000). In the first series of PGD per- formed in the late 1980s a misdiagnosis of sex occurred. In this case diagnosis had relied on the efficient amplification of a Y chromosome-specific sequence from all male embryos (Handyside et al., 1990). Since this time, phenomena such as allele dropout (ADO), contamination, chromosomal mosaicism and hybrid- isation failure have revealed PGD to be more techni- cally difficult than was originally imagined.
ADO and contamination are important for PCR diagnosis (Wells and Sherlock, 1998; Sermon, 2002). ADO is observed when only one of the two alleles in a heterozygous cell is detected after PCR (i.e. ampli- fication failure affecting just one allele) (Ray and Handyside, 1996). In most circumstances this does not
*Correspondence to: J. C. Harper, Department of Obstetrics and Gynaecology, University College London, 86–96 Chenies Mews, London WC1E 6HX, UK. E-mail: [email protected]
PRENATAL DIAGNOSIS
Prenat Diagn 2002; 22: 525–533. Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002 /pd.394
Copyright # 2002 John Wiley & Sons, Ltd. Received: 1 February 2002 Accepted: 27 March 2002
pose a major problem for single cell PCR, but it is very important for dominant disorders, where an affected embryo carries only one copy of the mutant allele. In this case if the affected allele does not amplify, the embryo will be diagnosed as normal, when actually it is affected. Contamination is also a major problem for all PCR diagnoses. Sperm and cumulus cells embedded in the zona pellucida of the embryo may become dislodged during the biopsy procedure and therefore lead to paternal or maternal contamination, respectively. To overcome this problem, intracyto- plasmic sperm injection (ICSI) is used to achieve fer- tilisation for PCR cycles and the cumulus cells must be carefully removed for all PGD cases. It is possible that paternal or maternal contamination has caused some of the misdiagnoses reported (Sermon et al., 1998; Harper and Delhanty, 2000). To overcome the problem of contamination, either maternal, paternal or from other sources, it has been advisable to per- form multiplex PCR reactions, including ideally linked (Rechitsky et al., 1998; Dreesen et al., 2000; Piyamongkol et al., 2001a,b) or unlinked polymorphic markers, which are informative for the family under- going PGD. For these markers the mother and father should have four different alleles, so when examining the embryo it should have inherited one allele from the father and one from the mother. Any other pattern would indicate the presence of additional DNA, suggesting that contamination had occurred. The use of fluorescent PCR is ideal for multiplex PCR from a single cell.
Here we report on the use of PCR for PGD of five different single gene disorders: myotonic dystrophy (DM), cystic fibrosis (CF) [DF508 and exon 4 (621+ 1 G>T)], fragile X and CF simultaneously, and two disorders for which PGD had not been previously attempted, namely neurofibromatosis type 2 (NF2) and Crouzon syndrome.
DM or Steinert’s disease is a progressive autosomal dominant muscular dystrophy. It is caused by an unstable CTG repeat expansion within exon 15 of the DMPK gene on chromosome 19q. Expanded (mutant) CTG repeat sequences are refractory to conventional PCR, but alleles with a number of repeats within the normal range can be readily amplified and detected. PGD, with analysis based on the detection of the nor- mal alleles, has been successfully performed (Sermon et al., 1997, 2001) but a misdiagnosis was reported (Sermon et al., 1998) which could have been due to maternal contamination. Therefore, we have devel- oped two new PGD protocols for DM, which utilise multiplex fluorescent PCR (Piyamongkol et al., 2001a,b). Ideally a linked polymorphic marker, APOC2, is ampli- fied in addition to the normal DMPK alleles, thus providing a back-up diagnostic result. We report on PGD for three couples where the females were carriers using a single-step duplex fluorescent (F)- PCR which will detect the DM triplet repeat with normal alleles. However, two of the couples were not fully informative at the APOC2 locus and so an unlinked STR marker, D21S1414, was substituted.
This locus provides no direct diagnostic information, but is still useful for contamination detection.
CF, an autosomal recessive disease caused by mu- tation in the CFTR gene, is located on chromosome 7q and was the first single gene defect for which PGD was offered (Handyside et al., 1992). The most common mutation in the Caucasian population is a three base pair deletion known as DF508 and the majority of PGD cycles for CF have been for this mutation (ESHRE PGD Consortium Steering Committee, 1999). Here we report on PGD for this mutation using a single- step triplex F-PCR with primers for the mutation, a linked marker and a contamination marker. A couple also presented with both partners carrying a mu- tation in exon 4 of CFTR (621+1 G>T). A triplex PCR with direct mutation detection, a linked marker and a contamination marker was used using nested PCR and SSCP for the mutation detection.
Fragile X syndrome is the most common cause of familial mental retardation. The most common mu- tation is an expansion of a triplet (CGG)n repeat in the 5k untranslated region of the FMR1 gene on Xq27.3. The expansion is refractory to PCR due to preferential amplification of the smaller allele in heterozygous cells and the high GC content of the repeat and surround- ing sequences. Detection of the normal and premu- tation alleles has been proposed for use in PGD of fragile X syndrome (Black et al., 1995, Sermon et al., 1999, 2001), an approach which has also been used in the diagnosis of two other triplet repeat disorders, DM (Sermon et al., 1997) and Huntington’s disease (Sermon et al., 1998). We investigated the use of linked polymorphic markers flanking the mutation to track the normal and premutation carrying maternal chro- mosomes in preimplantation embryos (Apessos et al., 2001). One couple presented where the female was carrying a large FMR1 premutation for fragile X syndrome and the couple were both carriers for CF, DF508. A single-step tetraplex F-PCR was performed employing amplification of two linked markers flank- ing FMR1, a sequence from the amelogenin gene for sexing, and a DNA fragment encompassing the DF508 mutation.
Mutations in the NF2 gene predispose to neuro- fibromatosis type 2 (NF2), a dominantly inherited cancer predisposition syndrome with an incidence of 1 in 33 000–40 000 (Evans et al., 1992). NF2 is charac- terised by the development of histologically benign tumours in the central nervous system. A couple pre- sented where the male carried a splice site mutation in intron 4: 514-2 (A to G) of NF2. The couple needed fertility treatment due to ovulatory problems in the female. The molecular diagnosis was performed by duplex F-PCR, followed by analysis of an informative intragenic single nucleotide polymorphism (SNP) and detection of the causative mutation using SSCP (Abou-Sleiman et al., 2002a).
Crouzon syndrome is a dominantly inherited cranio- synostosis syndrome that is caused by mutations in the fibroblast growth factor receptor 2 gene (FGFR2) on chromosome 10q (Reardon et al., 1994). It has an estimated birth prevalence of 15–16 per 1 million
J. C. HARPER ET AL.526
Copyright # 2002 John Wiley & Sons, Ltd. Prenat Diagn 2002; 22: 525–533.
births (Cohen and Kreiborg, 1992). The spectrum of mutations in FGFR2 is relatively limited compared to other genes. To date only 46 mutations have been identified, the majority of which are missense, with a smaller number of splice mutations or small insertions/ deletions, all of which remain in-frame. For PGD we developed DNA amplification using a multiplex nested PCR, for the simultaneous amplification of a fragment of the FGFR2 gene encompassing the mutation and a highly polymorphic short tandem repeat (STR) locus (unlinked, D21S11). SSCP was used to detect the mutation. The genotype of the embryos at the STR was determined by fragment analysis on an automated DNA sequencer by fluorescent polyacrylamide gel electrophoresis (PAGE). Two intragenic SNP had been described in FGFR2 at the time of preparation for the diagnosis. However, neither was suitable for use in this case as the parents were uninformative (Abou Sleiman et al., 2002b).
Here we report on 14 PGD cycles for eight patients carrying five single gene defects, using multiplex F-PCR incorporating polymorphic markers for con- tamination detection.
PATIENTS AND METHODS
All patients had previous genetic counselling before presenting to the PGD centre. Two IVF/PGD consul- tations were conducted during which the limitations of the PGD procedure were outlined, such as the need for IVF, risk of misdiagnosis, possibility that all embryos could be affected, PGD pregnancy rates, etc. Blood, DNA or buccal cell samples [obtained by scraping the inside of the cheek with a sterile wooden stick with a cotton swab tip and suspending in 3 ml phosphate- buffered saline (PBS) 4% (v/v) bovine serum albumin] were obtained from all couples and affected relatives prior to treatment. This enabled the work-up of a specific diagnosis for each couple, i.e. analysis of linked or unlinked markers for contamination detec- tion, development of multiplex PCR using these markers and optimisation of the mutation detection. The necessary investigations required prior to an IVF cycle were conducted, i.e. G test (Ranieri et al., 2001), HyCoSy, dummy embryo transfer, semen analysis, etc.
The time taken to work-up a specific diagnosis varied from 6 months to over a year. All work-ups were conducted on DNA, single buccal cells from the parents and affected relatives, and single blastomeres donated from our IVF programme. The results from the work-ups were submitted to the Human Fertilis- ation and Embryology Authority (HFEA) for licen- sing purposes and permission was granted for each disorder.
The patients underwent routine IVF procedures as outlined previously (Ranieri et al., 2001). ICSI was performed in all cases to reduce the risk of sperm contamination. Oocytes and embryos were cultured in IVF medium (Cook IVF, Brisbane, Australia). Prior to biopsy, all cumulus cells were mechanically removed to reduce the risk of maternal contamination. On Day
3, embryo biopsy was performed on all embryos that were considered suitable (depending on number of cells, grade and the number of embryos available), in Ca2+Mg2+-free embryo biopsy medium (Medicult, Surrey, UK), using Research Instrument (Cornwall, UK) micromanipulators. Zona drilling was performed using acid Tyrode’s solution as described previously (Piyamongkol et al., 2001a). Two blastomeres were removed from embryos containing six cells or more by aspiration and washed in PBS. In cases where only a small number of 6–8-cell embryos are available, we remove a single cell from embryos with less than six cells. However, in all the cases reported here, embryos were transferred after a normal result was obtained from two cells. All cells were washed at least three times in droplets of PBS/BSA before their transfer in 2 ml PBS/BSA to microcentrifuge tubes containing 3 ml 125 mg/ml proteinase K and 4r10x4% (w/v) sodium dodecyl sulphate (SDS). A control blank was pro- cessed for each blastomere using 2 ml of fluid from the final PBS wash drop. Cells were lysed by incubation at 37uC for 1 h and the proteinase K inactivated by incubation for 15 min at 95uC. After biopsy, embryos were stored in Cook extended culture medium.
A polymorphic marker was included in each diag- nosis to control against contamination. For the dominant disorders, where possible, linked poly- morphisms provided an additional means of determin- ing the genotype of the embryo hence reducing the risk of misdiagnosis due to ADO. Multiplex PCR was used in all cases, followed by fragment analysis using fluorescent DNA sequencers and/or SSCP and silver staining for genotyping. All diagnoses were performed within 24 h and embryo transfers undertaken on Day 4 post-insemination. Untransferred embryos were used to check the original diagnosis.
DM
The method and details of Patients 1 and 2 have been reported previously (Piyamongkol et al., 2001a). In brief, amplification of the expanded allele is not possible as it is refractory to single cell PCR. Therefore the diagnosis depends on the identification of the normal alleles (Brook et al., 1992). In the three patients, the females carried an expanded allele and so it was important to check for maternal contami- nation. We have previously reported on the use of the linked marker APOC2, but this marker was only informative for one of our PGD couples (Piyamongkol et al., 2001a,b). Therefore an unlinked STR marker, D21S1414 (which is on chromosome 21), was sub- stituted for contamination detection (Sherlock et al., 1998) for two couples. The multiplex amplified pro- ducts from single cells were each tagged with two dif- ferent fluorochromes using labelled primers. This allowed analysis to be performed on an automated laser fluorescent sequencer (ALFExpress1) (Pharmacia Biotech, Herts, UK) and also an ABI Prism1310 (PE Applied Biosystems, Warrington, UK).
PGD FOR SINGLE GENE DISORDERS 527
Copyright # 2002 John Wiley & Sons, Ltd. Prenat Diagn 2002; 22: 525–533.
CF DF508
One couple presented where both partners were carrying DF508. A single-step, triplex F-PCR was employed which allowed simultaneous amplification of the mutation locus together with two polymorphisms used as internal controls. An informative STR (D13S631) and the intron 6-linked polymorphism were used to detect any contamination and ADO, respectively. Fragment (size) analysis was carried out on the ALFExpress1. The primers used for the mu- tation were as described by Handyside et al. (1992), whereas the primers for the intron 6-linked poly- morphism were modified from the original described by Chehab et al. (1991).
CF exon 4
One couple presented where both partners carried a mutation in exon 4 (621+1 G>T). A single-step triplex F-PCR was performed for the mutation locus and two polymorphisms (the informative HumTh01 STR and the intron 6-linked polymorphism) followed by a second (nested) PCR round for the mutation locus only. Fragment (size) analysis was performed on the ALFExpress1 for the two polymorphisms. The mutation was detected by silver-stained SSCP on the semi-automated GenePhor1 electrophoresis system (Pharmacia Biotech, Herts, UK) using 12.5% polyacryl- amide gels at 15uC. The primers used for the mutation were designed by Ioulianos et al. (unpublished data), whereas the primers for the intron 6-linked poly- morphism were as described by Chehab et al. (1991).
CF and fragile X
A couple presented where the female carried an expansion that ranged from the upper permutation to the full mutation (133–300 repeats) and both partners were carrying the CF mutation DF508. A single-step, tetraplex F-PCR was used with two linked markers flanking to FMR1 [dinucleotide microsatellite repeat, DXS998 and (CA)n microsatellite polymorph- ism, p39] (Wehnert et al., 1993; Apessos et al., 2001), primers for amelogenin for embryo sexing (Sullivan et al., 1993), and primers for DF508 (Liu et al., 1992).
NF2
The causative mutation was identified as a single base pair substitution affecting a splice site in intron 4 of the gene (517-2 A/G). The female partner suffered from polycystic ovary syndrome and had never ovulated spontaneously; this case has previously been reported (Abou-Sleiman et al., 2002a). Two sets of fluores- cently labelled primers were designed to amplify the region of the gene harbouring the mutation, and a fragment of the 5k untranslated end of the gene encompassing a linked single nucleotide polymorph- ism, G/C substitution at nucleotide 8240. As the linked polymorphism was found to be informative in
the couple it could be used to infer the presence or absence of the mutation providing a secondary, inde- pendent means of detecting the mutation, thus reduc- ing the risk of misdiagnosis due to ADO. A duplex F-PCR was developed with simultaneous SSCP used for the mutation and SNP.
Crouzon syndrome
The causative mutation was identified in the female partner as a de novo G/A single base pair substitution at codon 568 situated in the alternative coding domain for the 3k half of the IgIII domain (numbering based on the sequence published by Zhang et al., 1999). The patient was diagnosed with Crouzon syndrome at 16 years of age and was mildly affected. This case has previously been reported (Abou-Sleiman et al., 2002b). The primers used for the mutation were as described by Sharma and Litt (1992). The mutation was detected by silver-stained SSCP on the semi-automated Phast- System1 electrophoresis system using 20% polyacryl- amide gels at 4uC. The…
Related Documents
