First successful application of preimplantation genetic diagnosis and haplotyping for congenital hyperinsulinism

Reproductive BioMedicine Online (2011) 22, 72– 79

www.sciencedirect.com www.rbmonline.com

CASE REPORT

First successful application of preimplantation genetic diagnosis and haplotyping for congenital hyperinsulinism Wafa Qubbaj a, Abdulrahman Al-Swaid b, Saad Al-Hassan c, Khalid Awartani c, Hesham Deek a, Serdar Coskun a,* a Department of Pathology and Laboratory Medicine, King Faisal Specialist Hospital and Research Center, 11211 Riyadh, Saudi Arabia; b King Abdulaziz Medical City, 11426 Riyadh, Saudi Arabia; c Department of Obstetrics and Gynecology, King Faisal Specialist Hospital and Research Center, 11211 Riyadh, Saudi Arabia

* Corresponding author. E-mail address: [email protected] (S Coskun). Dr Wafa Qubbaj obtained her PhD in 2003 at the Perinatal Centre, University College London and continued for 2 years to establish the preimplantation genetic diagnosis (PGD) service for haemoglobinopathy disorders for the NHS in UK. She was a research assistant at the Department of Pathology, Kuwait University (1985–1990) and laboratory supervisor at the Genetic Laboratory, Jordan University (1991–1999). Since 2006, she has worked at the King Faisal Specialist Hospital and Research Centre in Riyadh, Saudi Arabia, in charge of providing PGD service for monogenic diseases and chromosomal disorders, supervising more than 300 PGD cycles for single gene disorders and chromosomal aberrations.

Abstract Congenital hyperinsulinism is the most common cause of persistent hypoglycaemia in infancy. Early surgical intervention

is usually required to prevent brain damage. The prevention of the transmission to the offspring is important in families carrying the mutated gene. Preimplantation genetic diagnosis (PGD) is an early genetic testing procedure for couples at risk of transmitting inherited diseases. A 36-year-old Saudi woman married to her first cousin with four affected children was referred for PGD. The hyperinsulinism disease was caused by a novel homozygous mutation in the KCNJ11 gene, an arginine 301 to proline (R301P) substitution. PGD was achieved by whole genome amplification followed by mutation detection combined with short tandem repeat identifier analysis in the first cycle and with haplotyping in the second cycle. The first and second cycles resulted in the births of healthy twin girls and a boy, respectively. As far as is known, this is the first application of PGD to hyperinsulinism. A feasible strategy including whole genome amplification followed by direct mutation detection combined with haplotyping is described. Utilizing haplotyping increases the efficiency of PGD diagnosis as well as confirming the genetic diagnosis. It reveals the parental origin of each inherited chromosome. RBMOnline ª 2010, Reproductive Healthcare Ltd. Published by Elsevier Ltd. All rights reserved. KEYWORDS: haplotyping, preimplantation genetic diagnosis, whole genome amplification

1472-6483/$ - see front matter ª 2010, Reproductive Healthcare Ltd. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.rbmo.2010.09.016

PGD and haplotyping for congenital hyperinsulinism

Introduction Congenital hyperinsulinism (OMIM 256450) is the most common cause of persistent hypoglycaemia in infancy and is due to defective negative feedback regulation of insulin secretion by low glucose concentrations (Thornton et al., 1998). Medical treatments, especially in familial cases, are not always effective, thus pancreatectomy might be necessary as an alternative intervention (Shilyansky et al., 1997). Early surgical intervention is the only current treatment modality to prevent brain damage from recurrent episodes of hypoglycaemia (Schwitzgebel and Gitleman, 1998). Hyperinsulinism is genetically heterogeneous, with mutations in seven different genes accounting for approximately 50% of known cases (James et al., 2009): (i) SUR1 subunit of the pancreatic b-cell inwardly rectifying potassium channel at 11p15.1; (ii) Kir6.2 subunit of the pancreatic b-cell potassium channel at 11p15.1; (iii) glucokinase gene at 7p15-p13; (iv) HADH gene at 4q22-q26; (v) insulin receptor gene INSR at 19p13.2; (vi) GLUD1 gene at 10q23.3; and (vii) SLC16A1 gene at 1p13.2-p12. The mutations in the genes controlling insulin secretion pathways by pancreatic b-cells are involved in the disease manifestation. The inheritance model usually follows an autosomal recessive pattern (Barker et al., 1993; Stanley, 1997). However, dominant hyperinsulinism mutations have also been reported (Glaser et al., 1998; Huopio et al., 2000; Stanley et al., 1998). Mutations in a pair of two adjacent genes on chromosome 11p15.1, SUR1 (ABCC8) and Kir6.2 (KCNJ11) encoding the b-cell plasma membrane ATP-sensitive potassium channel have been widely reported (Nestorowicz et al., 1996, 1997; Thomas et al., 1995, 1996; Gloyn et al., 2004; Sagen et al., 2004; Vaxillaire et al., 2004; Hattersley et al., 1998; Hattersley and Tooke, 1999). The KCNJ11 gene has only one exon that spans about 3.41 kb and encodes a protein of 390 amino acids with transmembrane segments. Preimplantation genetic diagnosis (PGD) is a genetic testing procedure at the early embryonic stage for those couples at risk of transmitting inherited diseases (Handyside et al., 1990). It requires embryo generation by IVF and genetic diagnosis which is usually performed on cleavage-stage embryos. This permits the transfer of disease-free embryos, thereby sparing the need for abortion following prenatal diagnosis in cases of an affected fetus being identified. PGD can be performed, in principle, for any genetic condition with a recognized mutation (Verlinsky and Kuliev, 2000; Verlinsky et al., 1999). Many molecular biology techniques have been introduced in PGD diagnosis to increase its efficiency and to eliminate the possibility of misdiagnosis. However, allele dropout (ADO), the preferential amplification of one allele so that the second allele is amplified below detectable level, is the most common drawback of PGD (Coskun and Alsmadi, 2007). ADO might cause a tolerable misdiagnosis in recessive single-gene disorders since heterozygous embryos are being discarded as abnormal or might result in the unintentional transfer of heterozygous embryos as being normal (Qubbaj et al., 2008). Multiple displacement amplification (MDA) allows the generation of abundant DNA from a single cell which can be used in numerous reactions of many linked markers, a

73 strategy termed as haplotyping (Renwick et al., 2006). In PGD cases, preimplantation genetic haplotyping (PGH) can be utilized as a diagnostic tool through the recognition of specific haplotypes associated with DNA sequences encompassing the mutation (Qubbaj et al., 2008). PGH is a novel approach for any mapped single-gene disorders even in the absence of an established mutation. Many PGH cycles have been carried out for different single-gene disorders, resulting in births of many babies (Renwick et al., 2006, 2010). PGH requires careful study of the pedigree with at least one affected family member and identification of several informative linked markers for the mutated gene to be utilized in a PGD cycle. Here a successful PGD strategy is described for hyperinsulinism that included direct mutation analysis combined with PGH.

Materials and methods Case report A 36-year-old Saudi woman married to her first cousin with four affected children, one healthy daughter and one neonatal death, which was secondary to multiple congenital anomalies, was referred for PGD. They were counselled about PGD and agreed to undergo the procedure and signed an informed consent. The first affected daughter is now 18 years old and had hypoglycaemia soon after delivery which was not well controlled by the K(ATP) channel-agonist diazoxide. The disease progressed with episodes of hypoglycaemia-induced seizure and required 80% subtotal pancreatectomy at the age of 6 months that led to cessation of hypoglycaemia. However, on long term follow-up, she developed diabetes mellitus and required insulin therapy by age of 9 years. She developed episodes of unrelated generalized tonic clonic seizure by age of 16 years and is now in remission due to anticonvulsive therapy. The second pregnancy resulted in a healthy female while the third, fourth and fifth pregnancies resulted in three affected children who required subtotal pancreatectomy at 1–2 months of age. One of them is an 11-year-old boy who had the surgery twice because of the persistence of hypoglycaemia and later developed diabetes mellitus. The youngest is a 7-year-old girl who developed diabetes mellitus by age of 3 years and is on insulin therapy. The other affected child is a 10-year-old girl who is free from hypoglycaemia but on anticonvulsive therapy due to partial epilepsy. All the affected siblings have normal development and growth. Both parents are in good health without insulin-related symptoms. Genetic analysis indicated a homozygous 902G-C transversion mutation in the KCNJ11 gene, a novel mutation, resulting in an arginine 301 to proline (R301P) substitution (Figure 1).

Preimplantation genetic diagnosis The couple underwent two PGD cycles. Ovarian stimulation, oocyte retrieval, intracytoplasmic sperm injection (ICSI) procedure, embryo biopsy, cell lysis and MDA were performed as previously described (Coskun et al., 2000; Hellani et al., 2005). For MDA, REPLI-g midi kit (Qiagen) was used

74

W Qubbaj et al.

Figure 1 The sequence showing the presence of c.DNA 902 G > C (p.R301P) mutation in the KCNJ11 gene along with the normal and heterozygous sequences.

according to the manufacturer’s instructions for cell lysing, neutralization and amplification steps. Upon receiving the kit, the kit was tested by using single and multiple cells prior to use in PGD cases. Negative and positive controls were always included in each PGD. The first cycle was performed by whole genome amplification followed by mutation detection combined with short tandem repeat (STR) identifier analysis, while the second cycle included PGH. The remaining MDA products from the first cycle were utilized to validate the PGH at the single-cell level. The PCR reaction for the region flanking the mutation in the KCNJ11 gene was carried out using MDA products from single blastomeres and blank (medium from the last wash drop) in 20 ll reaction mixture consisting of 1· PCR buffer, 5 ll of 1/10 dilution of MDA product or genomic DNA, 0.5 pmole/ll of each primer set (forward: 50 -ACCATGTCATT GATGCCAAC-30 ; reverse: 50 -TCCAGTAGGCTGTGGTCCTC-30 ), 0.15 mmol/l dNTP, 1.25 mmol/l MgCl2 and 1 unit of Faststart Taq DNA polymerase (Roche, Applied Science). Genomic DNA for normal, carrier and homozygous recessive

were included as controls. Blanks were also included in PCR reactions. Positive reaction resulted in a PCR product of 298 bp. Initial enzyme activation and denaturation was carried out at 95C for 10 min. Thermal cycling was performed under the following conditions: denaturation at 94C for 1 min, annealing at 55C for 1 min and extension at 70C for 1 min, for 30 cycles of amplification. The PCR products were subjected to cycle sequencing using BigDye Terminator version 3.1 Ready Reaction Cycle Sequencing Kit with the forward primer. The amplified products were loaded on an ABI Prism 3100 Genetic Analyzer with a special polymer, POP-6. AmpFlSTR Identifier PCR Amplification kit was utilized to detect any contamination as described earlier (Hellani et al., 2004; all from Applied Biosystems). The remaining MDA product was kept at 20C for future investigations if needed.

Haplotype analysis A panel of 11 STR markers located within the reference interval D11S1307–D11S921 was used: D11S1307, D11S1308,

PGD and haplotyping for congenital hyperinsulinism

75 ware and an ABI PRISM 3100 Genetic Analyzer (Applied Biosystems).

D11S1791, D11S1888, D11S4096, D11S4099, D11S4130, D11S4138, D11S4160, D11S902 and D11S921. The choice of linked markers was based on the UCSC genome browser (http://genome.ucsc.edu/) or the UniSTS database (www.ncbi.nlm.nih.gov/unists/). Among those tested, four markers were uninformative while the remaining seven markers were informative (Figure 2). The remaining MDA product and genomic DNA from the babies born from the first cycle were haplotyped using these seven informative markers. All forward primers were fluorescently labelled at the 50 end either with 6-carboxyfluorescein or 5-hexachloro-fluorescein-CE phosphoramidite. The PCR was performed in 20 ll reactions; containing 5 ll of 1/10 dilution of MDA product or 50 ng/ll genomic DNA. It consists of 1· PCR buffer, 0.5 pmole/ll of each primer set, 0.15 mmol/l dNTP, 1.25 mmol/l MgCl2, 1% dimethylsulphoxide and 1 unit of Faststart Taq DNA polymerase (Roche, Applied Science). Initial enzyme activation and denaturation was carried out at 95C for 10 min, thermal cycling was performed under the following conditions; denaturation at 94C for 1 min, annealing at 50C for 1 min and extension at 70C for 1 min, for 25 cycles of amplification. Fragment analysis was performed using GeneScan soft-

D11S1307 D11S4160 D11S921 KCNJ11-gene D11S902 D11S4130 D11S1888 D11S4096

Affected child

M 125 241 247 M 145 267 214 301

123 265 242 N 141 273 220 313

125 241 247 M 145 267 214 301

125 270 247 N 151 270 210 282

N

M

N

Testing on single cells To assess the reliability of PGH analysis, the amplification efficiency and ADO rates of informative STR markers were evaluated from the MDA product of single lymphocytes and blastomeres using the PCR protocols developed for parental genomic DNA.

Results Amplification efficiency on single cells The amplification efficiency and ADO rates of STR markers were tested on 27 and 20 single lymphocytes and blastomeres, respectively (Table 1). The overall amplification efficiencies were 91% for single lymphocytes (range 85–96%) and 96% for single blastomeres (range 85–100%). ADO rates were measured to be 2–11% with an average of 7% for single lymphocytes and 0–10% with and average of 4% for single

125 241 247 M 145 267 214 301

M

Cycle 1 E1

E2

E3

E4

E5

E6

E7

M 125 241 247 M 145 267 214 301

125 241 247 M 145 267 214 301

123 265 242 N ADO 273 220 313

Cycle 2

125 241 247 M 145 267 214 301

125 241 247 M 145 267 214 301

E1 125 241 247 M 145 267 214 301

125 270 247 N 151 270 210 282

125 241 247 M 145 267 214 301

E2 123 265 242 N 141 273 220 313

123 265 242 N 141 273 220 313

E3 125 241 247 M 145 267 214 301

125 270 247 N 151 270 210 282

125 270 247 N 151 270 210 282

125 241 247 M 145 267 214 301

IVF/PGD child 2

IVF/PGD child 1

125 241 247 M 145 267 214 301

125 270 247 N 151 270 210 282

123 265 242 N 141 273 220 313

125 270 247 N 151 270 210 282

E4 123 265 242 N 141 273 220 313

125 241 247 M 145 267 214 301

Figure 2 Diagram showing the haplotyping analysis of DNA from the parents, the affected child, multiple displacement amplification products remaining from the first PGD-cycle embryos along with DNA from post-PGD children and PGH of second cycle. E = embryo; M = mutated; N = normal; PGD = preimplantation genetic diagnosis.

76

W Qubbaj et al. Table 1 Amplification efficiency and allele dropout rates on single lymphocytes and blastomeres following multiple displacement amplification for the STR markers used in preimplantation genetic haplotyping. STR marker

D11S4160 D11S921 D11S902 D11S4130 D11S1888 D11S4096 D11S1307 Total

Single lymphocytes (n = 27)

Single blastomeres (n = 20)

Amplification

ADO

Amplification

ADO

25 (93) 24 (89) 25 (93) 24 (89) 23 (85) 25 (93) 26 (96) 172/189 (91)

5 (10) 3 (6) 1 (2) 6 (13) 5 (11) 2 (4) 1 (2) 23/344 (7)

20 (100) 17 (85) 20 (100) 20 (100) 20 (100) 18 (90) 20 (100) 135/140 (96)

0 (0) 0 (0) 4 (10) 0 (0) 2 (5) 3 (8) 2 (5) 11/270 (4)

Values are n (%) or n/total (%). ADO = allele dropout; STR = single tandem repeat.

blastomeres. In both the retrospective haplotype analysis and the PGH during the second cycle, only one ADO was detected among all markers used (1/126; 1%).

PGD cycle 1 Following controlled ovarian stimulation, 12 oocytes were retrieved and eight were suitable for ICSI. Seven oocytes were fertilized and all were biopsied successfully on day 3 of the culture. One of the embryos failed to amplify following MDA. Of the remaining embryos, two were heterozygous and two were homozygous for the mutation. There were two embryos genotypically normal. The use of AmpFlSTR Identifier PCR Amplification kit excluded the risk of external DNA contamination in all tested embryos. Two of the embryos at morulae stage (embryo 3 as being carrier and embryo 6 as being normal) were transferred back to the mother on day 4. Embryo 7, which was diagnosed as being normal by sequencing, was found to be developmentally arrested and was not considered for embryo transfer. Pregnancy ensued and twin sisters were delivered at week 39 of gestation (Table 2). Genetic testing of the babies born confirmed the PGD results showing that one of the twins was heterozygous for the mutation while the other was normal. They are healthy and have been growing and developing normally. Retrospective haplotype analysis Haplotype analysis of the remaining MDA product from the first cycle using the informative linked markers is shown in Figure 2. One of the embryos failed to amplify as in the

original mutation analysis (embryo 5). Embryos 2 and 4 showed a homozygous haplotype pattern similar to the affected child. Among three carrier embryos, embryos 3 and 7 showed a carrier pattern with the normal maternally inherited allele, while embryo 1 showed a carrier haplotype with the normal paternally inherited allele. Embryo 6 showed normal haplotype pattern. The results of sequencing and haplotyping showed discordance for embryo 7 which had a carrier haplotype pattern and normal sequence probably due to ADO of the mutant allele during sequencing. All other embryos showed concordance between the sequencing and haplotyping. Since only embryos 3 and 6 were transferred, the homozygous normal girl should have developed from embryo 6 and the carrier girl has the same haplotype of embryo 3 (Figure 2).

PGD cycle 2 Following controlled ovarian stimulation, 10 oocytes were retrieved and five were suitable for ICSI. Four oocytes were fertilized and all were biopsied on day 3 of the culture. MDA was successfully performed on all biopsied blastomeres. Direct mutation detection by sequencing along with PGH analysis was performed on all MDA products. Sequencing revealed that both embryos 2 and 4 were affected, embryo 1 showed a carrier sequence, while embryo 3 showed a normal sequence. PGH analysis Embryo 1 showed a carrier haplotype with the normal paternally inherited allele present while both embryos 2 and 4

Table 2

Summary of PGD cycles.

Cycle

Mutation analysis

STR identifier kit

PGH

No. of embryos biopsied

MDA failure

Embryos transferred

Pregnancy

1

Yes

Yes

No

7

1

2

2

Yes

Yes

Yes

3

0

1

Delivered twin girls Delivered a boy

MDA = multiple displacement amplification; PGH = preimplantation genetic haplotyping; STR = single tandem repeat.

PGD and haplotyping for congenital hyperinsulinism showed a homozygous haplotype pattern similar to the affected child. Embryo 3 showed only the normally derived paternal allele which suggested a haploid pattern and was not considered for embryo transfer (Figure 2). There was no discordance between the PGH analysis and the sequencing results from embryos. Embryo 1 was transferred at hatching blastocyst stage on day 5. A healthy boy was delivered.

Discussion As far as is known, this is the first report to apply PGD to hyperinsulinism with haplotyping analysis. The first cycle resulted in the birth of healthy twins and the second cycle which included PGH analysis resulted in the delivery of a healthy boy. The main objective of PGD is to provide an alternative option for preventing the birth of an affected baby and to spare the need for therapeutic abortion following prenatal diagnosis for those couples at risk of transmitting inherited disorders. PGD could be applied to any genetic disorders as long as disease-causing mutations are identified. In many single-gene disorders, treatment options are very limited and, in many cases, they are in terms of treating the symptoms. Successful management also has cost implications. Furthermore, as new patients continue to be born and the number of cases requiring treatment rises, the costs to the healthcare system rise cumulatively. Therefore, prevention becomes important to contain the number of affected people. PGD has been shown to provide net economic benefits in families of cystic fibrosis compared with natural conception followed by prenatal testing and termination of affected pregnancies as an alternative prevention method (Davis et al., 2009; Tur-Kaspa et al., 2010). Moreover, prenatal diagnosis may not be acceptable to certain groups due to the involvement of abortions. Prior to a PGD cycle, couples should be counselled regarding treatment and the likelihood of misdiagnosis, which has been estimated to be at <1% for single-gene disorders following PGD (Wilton et al., 2009). The hyperinsulinism disease in the present family was caused by a novel mutation, a homozygous transversion 902G-C, in the KCNJ11 gene resulting in an arginine 301 to proline (R301P) substitution. This mutation is different from the previously reported 902G-A transition in the KCNJ11 gene, resulting in an arginine 301 to histidine (R301H) substitution in an infant with focal hyperinsulinism hypoglycaemia familial 2 (Henwood et al., 2005; Stanley et al., 2004). The availability of only a small quantity of DNA from a single cell obtained during PGD presents challenges to molecular analysis. Many strategies have been introduced to increase the efficiency and reliability of the diagnosis for mutation detection and to overcome conventional problems associated with single-cell PCR such as amplification failure, ADO and extraneous DNA contamination. The PGD strategy in the first cycle is based on whole genome amplification using MDA, followed by direct mutation detection and the use of informative unlinked markers to exclude the risk of external contamination. In the second PGD cycle, direct mutation detection was performed simultaneously with PGH analysis. Although whole genome amplification

77 enables the performance of many assays from small quantities of DNA derived from single cells, it does not eliminate the problems associated with single-cell analysis including amplification failure and ADO (Coskun and Alsmadi, 2007). Overall in the study centre’s experience, ADO rates following MDA were 14.7% and 9.1% when one and two cells were analysed from each embryo, respectively (Coskun et al., 2008). The validation of informative STR markers following MDA for the PGH analysis of hyperinsulinism on single cells showed acceptable amplification efficiency and ADO rates. In the Saudi population, 51% of births are to parents who are first or second cousins (Mathew et al., 1988). As a result of consanguinity, rare diseases with the pure homozygous genotypes and haplotypes were present in this family and both the paternal and maternal mutant-derived haplotypes were identical. Fortunately, normal alleles had enough informative markers for haplotyping analysis (Lander and Botstein, 1987). Haplotyping analysis on genomic DNA from both parents and their affected child allows detection of the informative linked markers to be included in the clinical PGD cycle and reveals the haplotypes associated with both normal and mutant chromosomes for such couples (Dreesen et al., 2000; Renwick et al., 2006). Moreover, applying PGH to extended family members reveals the incidence of crossover via the presence of recombinant haplotypes (Renwick et al., 2007). Linked markers for PGH analysis in this report included informative STR located at both sides of the gene of interest to reduce errors caused by recombination. PGH was validated via haplotyping analysis using the remaining MDA samples of embryos from the first cycle and the genomic DNA from the babies born as a result of that cycle. This allowed the study centre to determine which embryos generated each baby born following PGD and confirm intact haplotype segregation. When applying PGH, the presence of sufficient informative linked markers and at least one affected individual is needed to ensure a reliable diagnosis. Retrospective haplotyping analysis using the remaining MDA from previously biopsied blastomeres might allow the determination of possible haplotypes, including recombinants, as an alternative to extended family testing for couples who underwent previous PGD cycles. In classical PGD, it is difficult to determine which embryo has been implanted, especially in cases where two embryos have been transferred and resulted in a singleton pregnancy. Although in the first cycle, it was obvious that the carrier embryo generated the carrier baby and the normally diagnosed embryo resulted in the normal baby. The use of remaining MDA provided sufficient DNA for further investigation and retrospective testing for evaluating new techniques such as PGH. It allows for testing the efficiency of such linked markers and to estimate the risk of crossing over, if any. In both the retrospective haplotype analysis and the second PGD cycle, only one ADO was detected among all markers used. Similarly, overall ADO rates were less than 10% on single lymphocytes and blastomeres for STR markers used in PGH analysis. ADO following MDA has been reported to occur in 0.8–36% of the tested alleles (Handyside et al., 2004; Hellani et al., 2005; Renwick et al., 2006; Spits et al., 2006). One of the recommendations to minimize the risk of any misdiagnosis due to ADO is to test two cells

78 from each embryo if possible, since the probability of ADO affecting the same allele in both cells is low (Kontogianni et al., 1996). Additionally, the use of an informative linked polymorphism has been suggested to detect any ADO in the gene of interest (Kuliev et al., 1998). Even with high ADO, testing multiple linked markers located at the 30 and 50 ends of the gene of interest will allow haplotypes which are associated with the mutant allele to be identified (Renwick et al., 2006, 2010; Ogilvie et al., 2009). The retrospective haplotyping analysis on the first PGD cycle described in this paper proves the efficiency of utilizing PGH in future PGD cases, as there was no discordance between the sequencing results retrieved earlier on the day of PGD and on the predicted results from PGH analysis, except for embryo 7. The latter discordance was explained by the presence of ADO for the mutant sequence. This shows the significance of utilizing PGH analysis alongside mutation analysis, as it helps to exclude the risk of misdiagnosis due to ADO. Another advantage of fully informative linked markers is identifying the origin of each inherited chromosome which assists in the detection of any DNA contamination. In the second cycle, embryo 1 was diagnosed as being a carrier by sequencing and was confirmed using PGH. This embryo inherited the normal paternal allele along with the mutated maternal allele. In addition, the absence of any normal maternal allele excludes maternal contamination for such an embryo. On the other hand, embryo 3 showed normal sequencing pattern and this embryo should have been considered for embryo transfer. However, PGH analysis revealed the presence of only one chromosome, the normal paternal allele. It is unlikely that there were ADO at seven loci for the maternal allele. Consequently this embryo was diagnosed as being either haploid or monosomy for that chromosome and was not considered for embryo transfer. PGH analysis has been extended to 12 different diseases and 127 cycles have been performed in 101 couples with 13 ongoing pregnancies and 26 babies born (Renwick et al., 2010). It can even be applied in the absence of known mutations (Renwick et al., 2006). Moreover, recent advances in microarray technology and their application to single cell following whole genome amplification allowed the genome-wide linkage-based analysis of inheritance and detection of chromosome imbalances which covers aneuploidies of all the chromosomes (Handyside et al., 2009; Johnson et al., 2010). In conclusion, this study describes a feasible strategy for PGD for hyperinsulinism syndrome including whole genome amplification by MDA using direct mutation detection along with PGH. Utilizing PGH increases the efficiency of PGD diagnosis as well as it confirms the genetic diagnosis. It reveals the paternal origin of each inherited chromosome and also discloses the presence of any haploid cell or monosomic chromosome and helps to detect the presence of any DNA contamination. This information provided will benefit centres performing PGD not only for hyperinsulinism but also other single-gene disorders by applying similar strategies reported in this paper. In addition to medical, ethical and psychosocial benefits of PGD, it has also been found to be more cost-effective. Recent studies showed a clear cost benefit of PGD compared with the cost of prenatal diagnosis followed by pregnancy termination or treating

W Qubbaj et al. patients affected with cystic fibrosis (Davis et al., 2009; Tur-Kaspa et al., 2010). Implementing a PGD strategy for the prevention of genetic disease should have a remarkable implication on reducing the financial burden on the healthcare system as it helps to contain the number of affected people (Tur-Kaspa et al., 2010).

References Barker, D.J., Gluckman, P.D., Godfrey, K.M., Harding, J.E., Owens, J.A., Robinson, J.S., 1993. Fetal nutrition and cardiovascular disease in adult life. Lancet 341, 938–941. Coskun, S., Alsmadi, O., 2007. Whole genome amplification from a single cell: a new era for preimplantation genetic diagnosis. Prenat. Diagn. 27, 297–302. Coskun, S., Hollanders, J., Al Hassan, S., Al Sufyan, H., Al Mayman, H., Jaroudi, K., 2000. Day 5 versus day 3 embryo transfer: a controlled randomized trial. Hum. Reprod. 15, 1947–1952. Coskun, S., Qubbaj, W., Al-Kabra, M., Awartani, K., Al-Rejjal, R., Al-Hassan, S., 2008. Multiple displacement amplification efficiency could be improved by removing 2 blastomeres during preimplantation genetic diagnosis. Fertil. Steril. 90, S302–S303. Davis, L.B., Champion, S.J., Fair, S.O., Baker, V.L., Garber, A.M., 2009. A cost-benefit analysis of preimplantation genetic diagnosis for carrier couples of cystic fibrosis. Fertil. Steril. 93, 1793–1804. Dreesen, J.C., Jacobs, L.J., Bras, M., et al., 2000. Multiplex PCR of polymorphic markers flanking the CFTR gene; a general approach for preimplantation genetic diagnosis of cystic fibrosis. Mol. Human Reprod. 6, 391–396. Glaser, B., Kesavan, P., Heyman, M., et al., 1998. Familial hyperinsulinism caused by an activating glucokinase mutation. N. Engl. J. Med. 338, 226–230. Gloyn, A.L., Pearson, E.R., Antcliff, J.F., et al., 2004. Activating mutations in the gene encoding the ATP-sensitive potassium-channel subunit Kir6.2 and permanent neonatal diabetes. N. Engl. J. Med. 350, 1838–1849. Handyside, A.H., Kontogianni, E.H., Hardy, K., Winston, R.M., 1990. Pregnancies from biopsied human preimplantation embryos sexed by Y-specific DNA amplification. Nature 344, 768–770. Handyside, A.H., Robinson, M.D., Simpson, R.J., et al., 2004. Isothermal whole genome amplification from single and small numbers of cells: a new era for preimplantation genetic diagnosis of inherited disease. Mol. Human Reprod. 10, 767–772. Handyside, A.H., Harton, G.L., Mariani, B. et al. 2009. Karyomapping: a universal method for genome wide analysis of genetic disease based on mapping crossovers between parental haplotypes. J. Med. Genet. Oct 25 (Epub ahead of print). Hattersley, A.T., Beards, F., Ballantyne, E., Appleton, M., Harvey, R., Ellard, S., 1998. Mutations in the glucokinase gene of the fetus result in reduced birth weight. Nat. Genet. 19, 268–270. Hattersley, A.T., Tooke, J.E., 1999. The fetal insulin hypothesis: an alternative explanation of the association of low birthweight with diabetes and vascular disease. Lancet 353, 1789–1792. Hellani, A., Coskun, S., Benkhalifa, M., et al., 2004. Multiple displacement amplification on single cell and possible PGD applications. Mol. Human Reprod. 10, 847–852. Hellani, A., Coskun, S., Tbakhi, A., Al Hassan, S., 2005. Clinical application of multiple displacement amplification in preimplantation genetic diagnosis. Reprod. BioMed. Online 10, 376–380. Henwood, M.J., Kelly, A., Macmullen, C., et al., 2005. Genotype-phenotype correlations in children with congenital hyperinsulinism due to recessive mutations of the adenosine triphosphate-sensitive potassium channel genes. J. Clin. Endocrinol. Metab. 90, 789–794.

PGD and haplotyping for congenital hyperinsulinism Huopio, H., Reimann, F., Ashfield, R., et al., 2000. Dominantly inherited hyperinsulinism caused by a mutation in the sulfonylurea receptor type 1. J. Clin. Invest. 106, 897–906. James, C., Kapoor, R.R., Ismail, D., Hussain, K., 2009. The genetic basis of congenital hyperinsulinism. J. Med. Genet. 46, 289–299. Johnson, D.S., Gemelos, G., Baner, J. et al., 2010. Preclinical validation of a microarray method for full molecular karyotyping of blastomeres in a 24-h protocol. Human Reprod. Jan 24 (Epub ahead of print). Kontogianni, E.H., Griffin, D.K., Handyside, A.H., 1996. Identifying the sex of human preimplantation embryos in X-linked disease: amplification efficiency of a Y-specific alphoid repeat from single blastomeres with two lysis protocols. J. Assist. Reprod. Genet. 13, 125–132. Kuliev, A., Rechitsky, S., Verlinsky, O., et al., 1998. Preimplantation diagnosis of thalassemias. J. Assist. Reprod. Genet. 15, 219–225. Lander, E.S., Botstein, D., 1987. Homozygosity mapping: a way to map human recessive traits with the DNA of inbred children. Science 236, 1567–1570. Mathew, P.M., Young, J.M., Abu-Osba, Y.K., et al., 1988. Persistent neonatal hyperinsulinism. Clin. Pediatr. Phila. 27, 148–151. Nestorowicz, A., Wilson, B.A., Schoor, K.P., et al., 1996. Mutations in the sulonylurea receptor gene are associated with familial hyperinsulinism in Ashkenazi Jews. Hum. Mol. Genet. 5, 1813–1822. Nestorowicz, A., Inagaki, N., Gonoi, T., et al., 1997. A nonsense mutation in the inward rectifier potassium channel gene, Kir6.2, is associated with familial hyperinsulinism. Diabetes 46, 1743–1748. Ogilvie, C.M., Renwick, P.J., Khalaf, Y., Braude, P.R., 2009. First use of preimplantation genotyping in prevention of recurrent diandric complete hydatidiform mole. Reprod. Biomed. Online 19, 224–227. Qubbaj, W., Al Aqeel, A.I., Al Hassnan, Z., et al., 2008. Preimplantation genetic diagnosis of Morquio disease. Prenat. Diagn. 28, 900–903. Renwick, P.J., Trussler, J., Ostad-Saffari, E., et al., 2006. Proof of principle and first cases using preimplantation genetic haplotyping–a paradigm shift for embryo diagnosis. Reprod. BioMed. Online 13, 110–119. Renwick, P.J., Lewis, C.M., Abbs, S., Ogilvie, C.M., 2007. Determination of the genetic status of cleavage-stage human embryos by microsatellite marker analysis following multiple displacement amplification. Prenat. Diagn. 27, 206–215. Renwick, P., Trussler, J., Lashwood, A., Braude, P., Ogilvie, C., 2010. Preimplantation Genetic Haplotyping: 127 diagnostic cycles demonstrating a robust, efficient alternative to direct mutation testing on single cells. Reprod. Biomed. Online 20, 470–476. Sagen, J.V., Raeder, H., Hathout, E., et al., 2004. Permanent neonatal diabetes due to mutations in KCNJ11 encoding Kir6.2: patient characteristics and initial response to sulfonylurea therapy. Diabetes 53, 2713–2718.

79 Schwitzgebel, V.M., Gitelman, S.E., 1998. Neonatal hyperinsulinism. Clin. Perinatol. 25, 1015–1038, viii. Shilyansky, J., Fisher, S., Cutz, E., Perlman, K., Filler, R.M., 1997. Is 95% pancreatectomy the procedure of choice for treatment of persistent hyperinsulinemic hypoglycemia of the neonate? J. Pediatr. Surg. 32, 342–346. Spits, C., De Rycke, M., Verpoest, W., et al., 2006. Preimplantation genetic diagnosis for Marfan syndrome. Fertil. Steril. 86, 310–320. Stanley, C.A., 1997. Hyperinsulinism in infants and children. Pediatr. Clin. North Am. 44, 363–374. Stanley, C.A., Lieu, Y.K., Hsu, B.Y., et al., 1998. Hyperinsulinism and hyperammonemia in infants with regulatory mutations of the glutamate dehydrogenase gene. N. Engl. J. Med. 338, 1352–1357. Stanley, C.A., Thornton, P.S., Ganguly, A., et al., 2004. Preoperative evaluation of infants with focal or diffuse congenital hyperinsulinism by intravenous acute insulin response tests and selective pancreatic arterial calcium stimulation. J. Clin. Endocrinol. Metab. 89, 288–296. Thomas, P.M., Cote, G.J., Wohllk, N., et al., 1995. Mutations in the sulfonylurea receptor gene in familial persistent hyperinsulinemic hypoglycemia of infancy. Science 268, 426–429. Thomas, P., Ye, Y., Lightner, E., 1996. Mutation of the pancreatic islet inward rectifier Kir6.2 also leads to familial persistent hyperinsulinemic hypoglycemia of infancy. Hum. Mol. Genet. 5, 1809–1812. Thornton, P.S., Satin-Smith, M.S., Herold, K., et al., 1998. Familial hyperinsulinism with apparent autosomal dominant inheritance clinical and genetic differences from the autosomal recessive variant. J. Pediatr. 132, 9–14. Tur-Kaspa, I., Aljadeff, G., Rechitsky, S., Grotjan, H.E., Verlinsky, Y., 2010. PGD for all cystic fibrosis carrier couples: novel strategy for preventive medicine and cost analysis. Reprod. Biomed. Online 21, 186–195. Vaxillaire, M., Populaire, C., Busiah, K., et al., 2004. Kir6.2 mutations are a common cause of permanent neonatal diabetes in a large cohort of French patients. Diabetes 53, 2719–2722. Verlinsky, Y., Rechitsky, S., Verlinsky, O., et al., 1999. Prepregnancy testing for single-gene disorders by polar body analysis. Genet. Test. 3, 185–190. Verlinsky, Y., Kuliev, A., 2000. An Atlas of Preimplantation Genetic Diagnosis. The Parthenon Publishing Group, New York. Wilton, L., Thornhill, A., Traeger-Synodinos, J., Sermon, K.D., Harper, J.C., 2009. The causes of misdiagnosis and adverse outcomes in PGD. Hum. Reprod. 24, 1221–1228. Declaration: The authors report no financial or commercial conflicts of interest. Received 21 April 2010; refereed 20 September 2010; accepted 22 September 2010.