Analytical Biochemistry 400 (2010) 69–77
Contents lists available at ScienceDirect
Analytical Biochemistry
journal homepage: www.elsevier.com/locate/yabio
Preimplantation genetic diagnosis of b-thalassemia using real-time polymerase
chain reaction with fluorescence resonance energy transfer hybridization probes
Chia-Cheng Hung a,b,1, Shee-Uan Chen c,1, Shin-Yu Lin c,d, Mei-Ya Fang b, Li-Jung Chang c, Yi-Yi Tsai c,
Li-Ting Lin c, Yu-Shih Yang c, Chien-Nan Lee c,*, Yi-Ning Su a,b,d,*
a
Graduate Institute of Clinical Genomics, National Taiwan University College of Medicine, Taipei, Taiwan
b
Department of Medical Genetics, National Taiwan University Hospital, Taipei, Taiwan
c
Department of Obstetrics and Gynecology, National Taiwan University Hospital, Taipei, Taiwan
d
Graduate Institute of Clinical Medicine, National Taiwan University College of Medicine, Taipei, Taiwan
a r t i c l e i n f o a b s t r a c t
Article history: Preimplantation genetic diagnosis (PGD) is employed increasingly to allow transfer of embryos to the
Received 1 October 2009 uterus in assisted reproduction procedures. There are three stages of biopsy: polar bodies, one or two
Received in revised form 13 December 2009 blastomeres from the cleavage-stage embryos, and trophectoderm cells ($5 cells) from the blastocyst-
Accepted 15 December 2009
stage embryos. Validation of polymerase chain reaction (PCR)-based assays are challenging because only
Available online 24 December 2009
limited genetic material can be obtained for PGD. In the current study, we modified a valid single-cell PCR
protocol for PGD using real-time PCR assay with fluorescence resonance energy transfer (FRET) hybrid-
Keywords:
ization probes followed by melting curve analysis. We optimized and clinically applied the protocol, per-
Preimplantation genetic diagnosis
Fluorescence resonance energy transfer
mitting molecular genetic analysis to amplify a specific region on the beta-globin (HBB) gene for a couple,
Hybridization probes carriers of two mutations: c.-78A>G and c.52A>T. Among a total of eight embryos obtained after ovarian
Melting curve analysis stimulation, a single blastomere per embryo at the six- to eight-cell stage was biopsied. This PGD method
Real-time PCR assay showed that four embryos were unaffected, two embryos were selected for transfer, and one pregnancy
b-Thalassemia was achieved. Finally, a healthy male baby was delivered at 38 weeks’ gestation. The results obtained
HBB gene using the new method, FRET hybridization probes, were compared with findings using an existing
Primer extension minisequencing method, primer extension minisequencing.
Ó 2009 Elsevier Inc. All rights reserved.
b-Thalassemia (OMIM no. 141900), one of the most common tation [5,6]. Recently, preimplantation genetic diagnosis (PGD)2 is
monogenic genetic diseases in the world, is an autosomal reces- becoming widely used for analysis of monogenic diseases before
sive inherited disorder with an estimated carrier frequency of 1 pregnancy [7–11]. It is an alternative allowing prenatal genetic diag-
in 35 resulting from point mutations, small insertions, or dele- nosis of embryos for couples at high risk for inherited disorders be-
tions in the beta-globin (HBB) gene, resulting in either absence fore intrauterine transfer during assisted reproduction. The
or reduced synthesis of beta-globin chains [1]. The HBB gene is advantage of PGD over conventional prenatal diagnosis is the avoid-
located on chromosome 11 (11p15.5) and contains three exons ance of pregnancy termination. There are three main sources of ge-
[2]. Although it is a small gene, more than 200 different muta- netic material for PGD: polar bodies (biopsy of the first polar body
tions of the HBB gene cluster are associated with the b-thalasse- from oocytes before sperm insemination or biopsy of both polar
mia phenotype [3,4]. bodies after fertilization), blastomeres from cleavage-stage embryos,
Typically, prenatal diagnosis for high-risk pregnancies is per- and trophectoderm cells from embryos at the blastocyst stage. In
formed by analysis of fetal cells from chorionic villus sampling at clinical practice, polar body biopsy has been used for aneuploidy
10 to 12 weeks’ gestation or amniocentesis at 15 to 18 weeks’ ges- screening or for the detection of maternal transmission of single-
gene defects. Biopsy of the cleavage embryo at the six- to eight-cell
2
Abbreviations used: PGD, preimplantation genetic diagnosis; PCR, polymerase
chain reaction; ADO, allele dropout; RFLP, restriction fragment length polymorphism;
ARMS, amplification refractory mutation system; WGA, whole genome amplification;
HLA, human leukocyte antigen; FRET, fluorescent resonance energy transfer; GnRH,
* Corresponding authors. Fax: +886 2 2311 8228 (C.-N. Lee), +886 2 2381 6229 gonadotropin-releasing hormone; ICSI, intracytoplasmic sperm injection; PBS, phos-
(Y.-N. Su). phate-buffered saline; BSA, bovine serum albumin; DTT, DL-dithiothreitol; ALB,
E-mail addresses: leecn@ntu.edu.tw (C.-N. Lee), ynsu@ntu.edu.tw (Y.-N. Su). alkaline lysis buffer; dNTP, deoxynucleoside triphosphate; MBS, multiblock system;
1
These authors contributed equally to this study. LC, LightCycler; Tm, melting temperature; ddNTP, dideoxynucleoside triphosphate.
0003-2697/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved.
doi:10.1016/j.ab.2009.12.023
70 Preimplantation diagnosis of b-thalassemia / C.-C. Hung et al. / Anal. Biochem. 400 (2010) 69–77
stage for one or two blastomeres is a common approach for PGD. Re- Genetic counseling
cently, trophectoderm biopsy to retrieve approximately five cells at
the blastocyst stage has been used for PGD. The polar body or blas- A geneticist provided comprehensive genetic counseling to the
tomere biopsy provides a very limited amount of genetic material for couple to explain the risks and that three-quarters of the embryos
analysis and is challenging for accurate diagnosis. The blastocyst would be theoretically unaffected by b-thalassemia. The possibility
biopsy for obtaining more genetic material may have the advantage of a misdiagnosis using single-cell PCR was also discussed. The
of reducing the possibility of polymerase chain reaction (PCR) failure definitive genotype of the fetus would need to be confirmed using
or allele dropout (ADO) [12]. Because only limited genetic material conventional prenatal diagnostics such as chorionic villus sampling
can be obtained for PGD, it is critical and challenging work to estab- or amniocentesis. The informed consent given by the couple was
lish the validity of the PGD protocol to make an accurate molecular written, and the study was approved by the ethics committee of
diagnosis within a limited time. National Taiwan University Hospital.
Recently, DNA-based analyses have been performed based on
different principles to detect single-gene disorders [13–16]. Tradi- Testing of single lymphocytes
tional genotype analysis for PGD takes three steps. Initially, it in-
volves a first round of PCR to preamplify the selected genes. That To evaluate the reliability of the protocol before PGD, the first
is followed by nested PCR to produce sufficient DNA fragments round of PCR and real-time PCR with FRET hybridization were
for subsequent mutation analysis. Restriction fragment length tested on single lymphocytes collected from the husband and wife.
polymorphism (RFLP) [17–19], amplification refractory mutation Lymphocytes were isolated from 5 ml of unclotted blood. The col-
system (ARMS) [20], and primer extension minisequencing lection of single cells was transferred into 0.2-ml PCR tubes for
[21,22] are used for differentiation between mutant and wild-type analysis. Blank controls were also processed under the same condi-
alleles. In our laboratory, we used whole genome amplification tions to check for the presence of contamination. The amplification
(WGA) combined with PCR and primer extension for PGD of sin- efficiency, ADO rate, and contamination rate were measured.
gle-gene disorders and human leukocyte antigen (HLA) haplotype
analysis [23]. After WGA, the quality and quantity of DNA from a Cleavage-stage embryo biopsy
single blastomere can sufficiently facilitate analysis of multiple loci
[14,24]. However, prior to embryo transfer, the time allowed for Ovarian stimulations were performed using a long gonadotro-
PGD is limited. It would be helpful to have simple, rapid, inexpen- pin-releasing hormone (GnRH) agonist protocol as described previ-
sive, and reliable diagnostic methods for PGD. ously [29]. Intracytoplasmic sperm injection (ICSI) was used to
There is increasing need for the development of alternative pre- avoid contamination with sperm and surrounding cumulus cells.
implantation genetic diagnostic tools in medical laboratories. Real- Embryos that had six to eight cells on day 3 after fertilization
time PCR with molecular beacons provides a highly accurate assay were placed into 5-ll droplets of human tubal fluid medium consist-
for detection in single cells [25,26]. Moreover, real-time PCR in ing of 0.01 M Hepes buffer (Sigma–Aldrich, St. Louis, MO, USA) under
combination with hybridization probes designed for fluorescent mineral oil. Micromanipulation was performed with the aid of two
resonance energy transfer (FRET) is a powerful tool for detection micromanipulators (Narishige, Tokyo, Japan) mounted on an in-
of genetic variants with high efficiency and sensitivity [27]. The verted microscope (Nikon, Tokyo, Japan). The microscope was fitted
principle is based on fluorescence-labeled oligonucleotide probes with a noncontact microdrill laser system (Fertilase Medical Tech-
used for sequence variant detection. When the two probes are in nologies, Montreux, Switzerland). The laser system emitted an invis-
close proximity, FRET leads to excitation and thus fluorescence of ible diode beam at a wavelength of 1.48 lm. The embryo was kept
the second fluorophore by the energy emitted by the first using a holding pipette (Humagen Fertility Diagnostics, Charlottes-
fluorophore. ville, VA, USA), and zona was drilled by the laser at a power of
Here we present an alternative method with a single-cell PCR 47 mW with 10 ms exposure. A hole of approximately 30 lm was
protocol for PGD of b-thalassemia using real-time PCR with FRET made. The biopsy pipette (Humagen Fertility Diagnostics) was then
hybridization probes followed by melting curve analysis. We opti- inserted to remove a blastomere from each embryo. Following the
mized and clinically applied the protocol, permitting molecular ge- biopsy, embryos were further cultured until day 5 after fertilization,
netic analysis to amplify a specific region on the HBB gene by and unaffected blastocysts were transferred to the patient’s uterus.
analyzing a single blastomere of eight embryos for a couple, carri- Each blastomere was washed with wash buffer (phosphate-buf-
ers of two mutations: c.-78A>G and c.52A>T. The results obtained fered saline [PBS] without calcium and magnesium [Sigma Chem-
using the new method, FRET hybridization probes, were compared ical, St. Louis, MO, USA] and 0.5% bovine serum albumin [BSA,
with findings using an existing method, primer extension minis- Sigma Chemical]) and was transferred to a separate 0.2-ml PCR
equencing. tube containing 3 ll of lysis buffer (200 mM KOH [Sigma Chemical]
and 50 mM DL-dithiothreitol [DTT, Sigma Chemical]). The last wash
drop of each blastomere served as a blank control. Embryo transfer
Materials and methods was carried out at the blastocyst stage.
Patients Lysis of blastomeres and PCR
The couple, a 36-year-old man and a 33-year-old woman (both A singe blastomere was placed in the alkaline lysis buffer (ALB,
with b-thalassemia trait), was referred to our laboratory. The father containing 50 mM DTT and 200 mM KOH) [30] and was incubated
carried the c.52A>T mutation (CD17), and the mother carried the at 65 °C for 10 min, followed by incubation at 94 °C for 10 min. The
c.-78A>G mutation (À28), of the HBB gene. In her first and second subsequent PCR encompassing both mutation sites of the HBB gene
pregnancies, the woman underwent dilatation and curettage be- generated a fragment of 364 bp. Primer sequences were 50 -ACG
cause both fetuses had b-thalassemia major, as identified on chori- GCT GTC ATC ACT TAG AC-30 (sense) and 50 -GTC AGT GCC TAT
onic villus sampling for DNA testing [28]. To avoid a pregnancy CAG AAA CC-30 (antisense). PCR was performed in a total volume
with an affected fetus, the couple requested that PGD be done in of 25 ll containing 3 ll of lysis buffer with the lysed blastomere,
the Reproductive Medicine Laboratory (Department of Obstetrics 0.12 lM of each primer, 2.5 mM deoxynucleoside triphosphates
and Gynecology, National Taiwan University Hospital). (dNTPs), 0.5 U of AmpliTaq Gold enzyme (Applied Biosystems, Fos-
Preimplantation diagnosis of b-thalassemia / C.-C. Hung et al. / Anal. Biochem. 400 (2010) 69–77 71
ter City, CA, USA), and 2.5 ll of GeneAmp 10Â buffer II (10 mM Tris– 10 ll containing 1 ll of nested PCR product, 1 mM primer, and
HCl [pH 8.3] and 50 mM KCl) in 2 mM MgCl2, as provided by the 1Â SNP Start Master Mix (Beckman Coulter). The primer extension
manufacturer. PCR was performed in a multiblock system (MBS) primers were used to detect the mutation sites causing b-thalasse-
thermocycler (ThermoHybaid, Ashford, UK) using a touchdown mia (50 -atg gct ctg ccc tga ctt-30 for the c.-78A>G mutation and 50 -
PCR program with an initial denaturation at 95 °C for 10 min fol- aac ttc atc cac gtt cac ct-30 for the c.52A>T mutation). The reaction
lowed by 14 cycles consisting of denaturation at 94 °C for 30 s, was performed in an MBS thermocycler (ThermoHybaid) with 25
annealing at 64 °C for 30 s (decreased by 0.5 °C/cycle), and extension cycles of 90 °C for 10 s and 45 °C for 20 s. Postreaction sample
at 72 °C for 30 s. Immediately following the initial program, an addi- cleanup was done to avoid contamination by unincorporated dye
tional program was run and consisted of 26 cycles of denaturation at following the manufacturer’s protocol. Primer extension products
94 °C for 30 s, annealing at 57 °C for 30 s, and extension at 72 °C for were analyzed using the Beckman CEQ-8000 genetic analysis sys-
30 s, followed by a final extension at 72 °C for 10 min. tem (Beckman Coulter) with CEQ software using D-1-labeled 80
DNA size standard (Beckman Coulter).
Real-time PCR with FRET
Results
The first-round PCR product was further amplified by real-time
PCR in a LightCycler 480 instrument (Roche, Branchburg, NJ, USA). Results from genomic DNA
Real-time PCR was performed using two primers and two adjacent
fluorescent hybridization probes for the known HBB genotypes. The maternal mutation (c.-78A>G) and the paternal mutation
The primers, 50 -ACT TAG ACC TCA CCC TGT GGA-30 (sense) and (c.52A>T) were detected by FRET hybridization probes and melting
50 -AGG TTG CTA GTG AAC ACA GTT GT-30 (antisense), were used temperature (Tm) analysis based on genomic DNA. In the genomic
for the c.-78A>G mutation. One probe was labeled with fluorescein DNA sample of the mother, the peak at the lower temperature
at the 30 end (50 -TGG GCA TAA AAG TCA GGG C-30 ) as the donor, (54 °C) corresponded to the mutated allele (G), whereas the peak
and another probe (acceptor) was labeled with the LightCycler at the higher temperature (62 °C) corresponded to the wild-type
(LC) Red fluorophore at the 50 end (50 -GAG CCA TCT ATT GCT TAC allele (A), as shown in Fig. 1A. Similarly, the genomic DNA of the
ATT TGC TTC TGA-30 ), which phosphorylated at the 30 end. Primers father showed a peak at the lower temperature (61 °C) correspond-
for the c.52A>T mutation were 50 -ACT TAG ACC TCA CCC TGT GGA- ing to the wild-type allele (A) and a peak at the higher temperature
30 (sense) and 50 -GTT CTA TTG GTC TCC TTA AAG GTG-30 (anti- (66 °C) belonging to the mutated allele (T), as shown in Fig. 2A.
sense). The FRET donor probe was fluorescein labeled (50 -TCC
ACG TTC ACC TAG CCC CA-30 ), and the FRET acceptor probe was
Amplification efficiency on single lymphocytes
LC Red labeled (50 -AGT CTG CCG TTA CTG CTC TGT-30 ).
Real-time PCR was performed in a total volume of 10 ll con-
Prior to performing the PGD cycle, 50 single lymphocytes were
taining 1 ll of PCR products, 3 mM MgCl2, 0.25 lM of each primer,
isolated and the complete procedure was performed for the two
0.25 lM fluorescein probe, 0.25 lM LC Red fluorophore probe, and
mutated loci. The amplification efficiency of the HBB gene was 92
1Â LightCycler FastStart DNA Master Hybridization Probes Mix
to 96% with a mean of 95%. ADO was measured at 6 to 8% with a
(Roche), as provided by the manufacturer. The cycling conditions
mean of 7%. There was no contamination of the 100 reactions.
for real-time PCR in the LiC480 were as follows: 95 °C for 10 min,
The duration of the procedures was 4 h and included the first-
followed by 50 cycles of denaturation at 95 °C for 10 s with a tem-
round PCR (3 h) and real-time PCR with FRET hybridization plus
perature transition rate of 4.4 °C/s, annealing at 53 °C for 10 s with
melting curve analysis (1 h).
a temperature transition rate of 2.2 °C/s, and extension at 72 °C for
10 s, with a temperature transition rate of 4.4 °C/s.
Mutation analysis of blastomeres
Melting curve assay
After ovarian stimulation and ICSI, a total of eight embryos were
The subsequent melting curve analysis was appended to the biopsied. Results for amplification and genotyping of the c.-78A>G
amplification step, which was performed by completely denatur- site from the single blastomeres of the eight embryos are illus-
ing the PCR products at 95 °C for 1 min, with a ramping rate of trated in Fig. 1B. The mutated allele lowered the Tm of the probe
4.4 °C/s, cooling to 40 °C, with a ramping rate of 1.5 °C/s, and then by 8 °C. Genotyping analyses for the c.52A>T site of the eight blas-
heating to 95 °C, with a ramping rate of 2.2 °C/s, with continuous tomeres from the embryos are shown in Fig. 2B. The mutated allele
fluorescence monitoring [31]. The melting curve data were ana- resulted in a Tm shift of 5 °C. The overall results are shown in
lyzed using gene scanning module software (version 1.3, Roche). Table 1, and each blastomere was analyzed in triplicate. Based on
real-time PCR with FRET hybridization probes, four embryos
Primer extension minisequencing (embryos 3, 4, 5, and 7) were diagnosed definitively unaffected
by b-thalassemia. Two were not diagnosed due to amplification
The PCR–FRET results were compared with primer extension failure. The remaining two embryos were diagnosed as affected
minisequencing as described previously [32]. Briefly, nested PCR cases. The amplification rates were 81.25% (13/16), 87.50% (14/
was performed for further amplification of the specific regions on 16), and 93.75% (15/16), respectively (Table 1). There was no con-
the HBB genes. Nested PCR was performed in a total volume of tamination of the 24 reactions.
25 ll containing 2 ll of first-round PCR product, 0.12 lM of each
primer (sense: 50 -gac agg tac ggc tgt ca-30 ; antisense: 50 -gtc tcc Primer extension minisequencing
aca tgc cca gtt tc-30 ), 2.5 mM dNTPs, 0.5 U of AmpliTaq Gold en-
zyme (Applied Biosystems), and 2.5 ll of GeneAmp 10Â buffer II All embryos were processed for genetic analysis by primer exten-
(10 mM Tris–HCl [pH 8.3] and 50 mM KCl) in 2 mM MgCl2, as pro- sion minisequencing to confirm and compare with the melting anal-
vided by the manufacturer. Nested PCR was performed under the ysis results, as described previously. Using the primer extension
same conditions as first-round PCR. The subsequent primer exten- assay with different dideoxynucleoside triphosphate (ddNTP)-la-
sion reaction was performed using a GenomeLab SNPStart Primer beled dyes, we accurately determined the mutated/wild-type alleles
Extension Kit (Beckman Coulter, Brea, CA, USA) in a volume of of the primer extension products by their different dyes.
72 Preimplantation diagnosis of b-thalassemia / C.-C. Hung et al. / Anal. Biochem. 400 (2010) 69–77
Fig. 1. Melting curve analysis of real-time hybridization with FRET probes for c.-78 mutation of the HBB gene. (A) Upper panel: For the wild-type control with homozygous c.-
78A, the Tm of the matched DNA duplex was 62 °C. Lower panel: For the heterozygous c.-78A and c.-78G, the peak Tm of 54 °C was from mismatched DNA duplex of the
mutant allele with c.-78G. The other peak of Tm at 62 °C resulted from the wild-type allele with c.-78A. (B) Embryos 1 to 8.
Preimplantation diagnosis of b-thalassemia / C.-C. Hung et al. / Anal. Biochem. 400 (2010) 69–77 73
Fig. 2. Genotyping of the c.52 mutation site by FRET hybridization probing. (A) Upper panel: For the wild-type control with homozygous c.52A, the Tm of the mismatched
DNA duplex was 61 °C. Lower panel: For the heterozygous c.52T and c.52A, the peak Tm occurred at 66 °C was from matched DNA duplex of the mutant allele with c.52T. The
other peak of Tm at 61 °C resulted from the wild-type allele with c.52A. (B) Results of the FERT hybridization for embryos 1 to 8.
74 Preimplantation diagnosis of b-thalassemia / C.-C. Hung et al. / Anal. Biochem. 400 (2010) 69–77
Table 1
Summary of results of the PGD for the HBB gene with FRET hybridization probes and primer extension.
FRET hybridization probes Primer extension
First time Second time Third time
Embryo Mutation site Mutation site Mutation site Mutation site
c.-78 c.52 c.-78 c.52 c.-78 c.52 c.-78 c.52
1 G A G A G A G A
2 No signal A No signal A No signal A No signal No signal
3 A A A A A A A A
4 A A A A A A A A
5 A A A A A A A A
6 A T A T A T A T
7 A A A A A A A A
8 A A A No signal No signal No signal No signal No signal
In Fig. 3, for each example of an electropherogram from the pri- genotype sequencing showed that the fetus was a carrier of the
mer extension analysis, the first peak corresponded to the unex- maternal mutation (c.-78A>G). Finally, a healthy male baby was
tended primer extension primer and always presented the same delivered at 38 weeks’ gestation.
retention time in each electropherogram. There were two extended
peaks that were identified for individuals with A (wild-type) and G
Discussion
(mutation) alleles. Likewise, in Fig. 4, there were two extended
peaks that were identified for individuals with A (wild-type) and
A schematic representation of the single-cell PCR protocol used
T (mutation) alleles. Embryos 2 and 8 failed at amplification on
for genotyping the region of interest of the HBB gene is shown in
both mutation sites. The amplification rate for primer extension
Fig. 5. A schematic diagram of the gene structure showing positions
was 75% (12/16). Embryos 3, 4, 5, and 7 were confirmed as unaf-
of amplification primers, mutation sites, and hybridization probes
fected by b-thalassemia.
is shown in Fig. 6. After embryo biopsy, a single blastomere cell
was lysed. The template was a 364-bp fragment generated from
Embryo transfer and follow-up of pregnancy PCR amplification of the HBB gene using the appropriate forward
and reverse primers. The mutation sites of the HBB gene (c.-
Two unaffected embryos (embryos 5 and 7) were transferred 78A>G and c.52A>T) were within the region. The FRET hybridiza-
into the uterus at the blastocyst stage. The other two unaffected tion probes lie immediately adjacent to the mutation target, which
embryos (embryos 3 and 4) were cryopreserved for future use. is detected as the emission wavelength of the acceptor fluoro-
The remaining four embryos stopped growing and were discarded. phore. Finally, the melting curve was used to analyze the PCR prod-
The patient had a singleton pregnancy. Prenatal diagnosis was per- ucts for mutation analysis of the HBB genes. The strategy for direct
formed by chorionic villus sampling at 11 weeks’ gestation. Direct analysis of embryo mutations by FRET hybridization probes and
Fig. 3. Electropherograms of primer extension minisequencing corresponding to negative control, wild-type, c.-78 A>G/WT, c.-78 A>G/WT, and each of embryos 1 to 8. For
each example of an electropherogram from the primer extension analysis, the first peak corresponded to the unextended primer extension primer and always presented the
same retention time in each electropherogram. There were two extended peaks that were identified for individuals with A (wild-type) and G (mutation) alleles.
Preimplantation diagnosis of b-thalassemia / C.-C. Hung et al. / Anal. Biochem. 400 (2010) 69–77 75
Fig. 4. Minisequencing electropherograms of the c.52 A>T mutation corresponding to negative control, wild-type, c.-52 A>T/c.-52 A>T mutation, c.-52 A>T /WT, and each of
embryos 1 to 8. For each example of an electropherogram from the primer extension analysis, the first peak corresponded to the unextended primer extension primer and
always presented the same retention time in each electropherogram. There were two extended peaks that were identified for individuals with A (wild-type) and T (mutation)
alleles.
were used in the LightCycler (Roche), which is a PCR-based instru-
ment that provides real-time and post-PCR analysis in a closed
tube environment. It can also provide high-speed throughput that
could analyze 96 or 384 samples per hour. We compared the lab-
oratory costs of PCR combined with FRET hybridization probes,
nested PCR followed by primer extension, and WGA. The costs of
diagnosis were estimated to be U.S. $6, $10, and $20 for FRET
hybridization, primer extension, and WGA, respectively. However,
it should be noted that the expensive fluorescent probes of FRET
assays cannot be used for PGD of other mutations; thus, unless
there are many other PGD cases involving the same mutations
within the next 1 or 2 years, the entire cost of the synthesized
probes (costing hundreds of dollars for each fluorophore–quencher
pair) must be borne by the patients regardless of how much probe
was used during the PGD cycle. This is unlike the case where pri-
mer extension reagents are used because the kits are generic and
can be used for any mutation in any gene and the unlabeled min-
isequencing primers are comparatively cheap. In terms of length of
time, the WGA method requires 12 h because it involves a hybrid-
ization procedure, primer extension requires 9 h because it in-
volves DNA cleanup, and FRET hybridization can be completed
within 4 h, which includes the time it takes to perform the PCR
amplification, FRET hybridization, and melting curve analysis.
Our results show that the FRET hybridization strategy is accurate
[33,34] and is a good alternative for PGD.
Fig. 5. Design of hybridization probes for real-time FRET detection of the c58.A>T Despite the preclinical validation results suggesting otherwise,
and c-78.A>G mutations of the HBB gene.
there is compelling evidence that significant ADO occurred during
the PGD cycle analysis. First, none of the eight embryos was het-
erozygous at either the c.-78 or c.52 mutation site. Based on allele
melting curve analysis was successfully applied to b-thalassemia. segregation and inheritance probabilities, the odds of an embryo
The illustrated technique can be used to analyze single-cell sam- being heterozygous at one or both mutation sites is 0.75, whereas
ples for a known mutation. the corresponding probability that an embryo is not heterozygous
Genotyping by FRET hybridization probes is generally consid- at any mutation site is 0.25. Thus, the joint probability that none of
ered as more reliable. The FRET hybridization probes can deter- the eight embryos would be heterozygous at any mutation site is
mine single-base variations by changes in Tm with high 0.25 to the power of 8, or 0.000015, a highly unlikely probability.
sensitivity and resolution. Moreover, FRET hybridization probes Second, embryo 1 showed only the mutant G allele at the c.-78
76 Preimplantation diagnosis of b-thalassemia / C.-C. Hung et al. / Anal. Biochem. 400 (2010) 69–77
Fig. 6. Human HBB gene sequence (NG_000007), primer sequences, probe positions, and the mutation sites. The first-round primers are labeled in boldface letters and orange
font, the mutation sites are labeled in boldface letters and red font, the sensor probes are marked with arrows and red font, and the anchor probes are marked with arrows and
blue font. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)
mutation site, whereas embryo 6 showed only the mutant T allele to 95% with ADO rates of 5 to 15%, and are within the guidelines set
at the c.52 mutation site. Because the couple carry different muta- by the Preimplantation Genetic Diagnosis International Society
tions, it is impossible for any embryo to inherit a homozygous mu- (PGDIS) [35].
tant genotype at either mutation site. An apparently homozygous In comparing the genetic results by FRET hybridization and pri-
mutant genotype is observed only when ADO of the normal allele mer extension, embryos 3, 4, 5, and 7 were healthy (either wild
occurs in an embryo that is heterozygous or compound heterozy- types or carriers). Embryos 1 and 6 were diagnosed with b-thalas-
gous for that mutation. The observation of apparently homozygous semia major and were rejected for transfer. Embryos 2 and 8 failed
mutant genotypes in two of the eight embryos is, therefore, com- primer extension minisequencing and could not be diagnosed.
pelling evidence of ADO of the normal allele at the respective Reanalysis of these two embryos also failed during the second
mutation sites. Third, embryos 3, 4, 5, and 7 were apparently and third FRET hybridizations (Table 1). This indicated that the
homozygous normal at both the c.-78 and c.52 mutation sites, two embryos were noninformative. After the couple underwent ge-
and embryos 5 and 7 were transferred, resulting in a singleton netic counseling, the woman’s uterus was implanted with embryos
pregnancy. Subsequent prenatal diagnosis, however, showed that 5 and 7 at the blastocyst stage. Prenatal diagnostic chorionic villus
the fetus was not homozygous normal but in fact a carrier of the sampling was performed and showed a normal fetus carrying the
maternal c.-78 mutation. This is further evidence that ADO, this maternal mutation (c.-78A>G). A normal infant boy was delivered.
time of the mutant allele at the c.-78 mutation site, had occurred We have reported the successful PGD and delivery of a healthy
in at least one of the two transferred embryos. The results as boy to a couple at risk for transmitting b-thalassemia. Using real-
shown in Table 1 strongly suggest that most of the ADO in this case time PCR with FRET hybridization for PGD, the entire genotyping
report occurred during the first PCR round. We think that this high procedure can be accomplished easily in a single working day. This
ADO is related in some way to the fact that we did not have a neu- rapid and accurate assay is highly desirable for clinical use in PGD.
tralization step after blastomere lysis to ensure that the final PCR It is imperative for blastocyst biopsy because the time interval per-
pH conditions are near optimal. mitted for molecular diagnosis is stringent so as to transfer a fresh
Given the demonstrated occurrence of ADO in three embryos embryo. Real-time PCR with FRET hybridization can be successfully
(two involving a normal allele and one involving a mutant allele), performed for PGD of other monogenic diseases. It deserves further
the question is whether any ADO occurred in any of the other five investigation in the future.
embryos. Unfortunately, we will not be able to know this because
the PGD strategy adopted here does not allow for detection of ADO
of either or both mutant alleles when an embryo has an apparently Acknowledgments
homozygous normal genotype at both mutation sites. Nested
linked PCR of both mutation sites would have been the best way We are very grateful to the couple who participated in this re-
to prevent serious misdiagnosis of an affected embryo as being a search. We thank the National Science Council of Taiwan (Grants
carrier or normal embryo. Otherwise, it should have included NSC 95-2314-B-002-280-MY3 and NSC 96-2628-B-002-063-MY3)
flanking linked markers to mitigate this deficiency. The danger for financial support.
with ADO is the failure to detect the mutant alleles when both
are present, which would then result in a serious misdiagnosis.
References
Aside from lacking a means to detect ADO of the mutant alleles,
the assay also does not incorporate a means of monitoring for the [1] T.H. Huisman, M.F. Carver, The b- and d-thalassemia repository (nineth ed.,
presence of exogenous DNA contamination in the blastomere reac- part I), Hemoglobin 22 (1998) 169–195.
tions. We performed the 24 secondary PCR blank reactions, and the [2] R.V. Lebo, M.C. Cheung, B.D. Bruce, V.M. Riccardi, F.T. Kao, Y.W. Kan, Mapping
parathyroid hormone, beta-globin, insulin, and LDH-A genes within the human
results were blank. However, these serve a general purpose of chromosome 11 short arm by spot blotting sorted chromosomes, Hum. Genet.
monitoring overall DNA contamination, but there is no monitoring 69 (1985) 316–320.
of DNA contamination within the tubes containing the blasto- [3] R.C. Hardison, D.H. Chui, B. Giardine, C. Riemer, G.P. Patrinos, N. Anagnou, W.
Miller, H. Wajcman, HbVar: a relational database of human hemoglobin
meres. This is another deficiency of the assay design. variants and thalassemia mutations at the globin gene server, Hum. Mutat. 19
The amplification efficiency of real-time PCR with FRET hybrid- (2002) 225–233.
ization on single lymphocytes was 94 to 98% with an ADO rate of 6 [4] U. Bhardwaj, Y.H. Zhang, F. Lorey, L.L. McCabe, E.R. McCabe, Molecular genetic
confirmatory testing from newborn screening samples for the common
to 10%. The ADO rates for PCR were comparable to those in previ- African–American, Asian Indian, Southeast Asian, and Chinese b-thalassemia
ous PGD works, which reported PCR amplification efficiencies of 85 mutations, Am. J. Hematol. 78 (2005) 249–255.
Preimplantation diagnosis of b-thalassemia / C.-C. Hung et al. / Anal. Biochem. 400 (2010) 69–77 77
[5] T.E. Cohen-Overbeek, W.C. Hop, M. den Ouden, L. Pijpers, M.G. Jahoda, J.W. outcome of 250 cycles of preimplantation genetic diagnosis for single gene
Wladimiroff, Spontaneous abortion rate and advanced maternal age: disorders, Hum. Reprod. 21 (2006) 670–684.
consequences for prenatal diagnosis, Lancet 336 (1990) 27–29. [22] A. Obradors, E. Fernandez, M. Rius, M. Oliver-Bonet, M. Martinez-Fresno, J.
[6] B. Chadefaux, D. Rabier, Y. Dumez, J.E. Oury, P. Kamoun, Eleventh week Benet, J. Navarro, Outcome of twin babies free of Von Hippel–Lindau disease
amniocentesis for prenatal diagnosis of metabolic diseases, Lancet 1 (1989) after a double-factor preimplantation genetic diagnosis: monogenetic
849. mutation analysis and comprehensive aneuploidy screening, Fertil. Steril. 91
[7] R.G. Edwards, P. Hollands, New advances in human embryology: implications (2009) 933.e1–933.e7.
of the preimplantation diagnosis of genetic disease, Hum. Reprod. 3 (1988) [23] S.U. Chen, Y.N. Su, M.Y. Fang, L.J. Chang, Y.Y. Tsai, L.T. Lin, C.N. Lee, Y.S. Yang,
549–556. PGD of b-thalassaemia and HLA haplotypes using OmniPlex whole genome
[8] C. Coutelle, C. Williams, A. Handyside, K. Hardy, R. Winston, R. Williamson, amplification, Reprod. Biomed. Online 17 (2008) 699–705.
Genetic analysis of DNA from single human oocytes: a model for [24] K. Silander, J. Saarela, Whole genome amplification with Phi29 DNA
preimplantation diagnosis of cystic fibrosis, Br. Med. J. 299 (1989) 22–24. polymerase to enable genetic or genomic analysis of samples of low DNA
[9] M. Adinolfi, P.E. Polani, Prenatal diagnosis of genetic disorders in yield, Methods Mol. Biol. 439 (2008) 1–18.
preimplantation embryos: invasive and non-invasive approaches, Hum. [25] K.E. Pierce, J.E. Rice, J.A. Sanchez, L.J. Wangh, Detection of cystic fibrosis alleles
Genet. 83 (1989) 16–19. from single cells using molecular beacons and a novel method of asymmetric
[10] K. Sermon, A. Van Steirteghem, I. Liebaers, Preimplantation genetic diagnosis, real-time PCR, Mol. Hum. Reprod. 9 (2003) 815–820.
Lancet 363 (2004) 1633–1641. [26] J.E. Rice, J.A. Sanchez, K.E. Pierce, L.J. Wangh, Real-time PCR with molecular
[11] J.D. Delhanty, Preimplantation diagnosis, Prenat. Diagn. 14 (1994) 1217–1227. beacons provides a highly accurate assay for detection of Tay–Sachs alleles in
[12] G. Kokkali, J. Traeger-Synodinos, C. Vrettou, D. Stavrou, G.M. Jones, D.S. Cram, single cells, Prenat. Diagn. 22 (2002) 1130–1134.
E. Makrakis, A.O. Trounson, E. Kanavakis, K. Pantos, Blastocyst biopsy versus [27] C. Vrettou, J. Traeger-Synodinos, M. Tzetis, G. Palmer, C. Sofocleous, E.
cleavage stage biopsy and blastocyst transfer for preimplantation genetic Kanavakis, Real-time PCR for single-cell genotyping in sickle cell and
diagnosis of b-thalassaemia: a pilot study, Hum. Reprod. 22 (2007) 1443– thalassemia syndromes as a rapid, accurate, reliable, and widely applicable
1449. protocol for preimplantation genetic diagnosis, Hum. Mutat. 23 (2004) 513–
[13] P. Burlet, N. Frydman, N. Gigarel, V. Kerbrat, G. Tachdjian, E. Feyereisen, J.P. 521.
Bonnefont, R. Frydman, A. Munnich, J. Steffann, Multiple displacement [28] C.C. Hung, Y.N. Su, C.Y. Lin, Y.F. Chang, C.H. Chang, W.F. Cheng, C.A. Chen, C.N.
amplification improves PGD for fragile X syndrome, Mol. Hum. Reprod. 12 Lee, W.L. Lin, Comparison of the mismatch-specific endonuclease method and
(2006) 647–652. denaturing high-performance liquid chromatography for the identification of
[14] S. Coskun, O. Alsmadi, Whole genome amplification from a single cell: a new HBB gene mutations, BMC Biotechnol. 8 (2008) 62.
era for preimplantation genetic diagnosis, Prenat. Diagn. 27 (2007) 297–302. [29] C.H. Ho, S.U. Chen, F.S. Peng, C.Y. Chang, Y.R. Lien, Y.S. Yang, Prospective
[15] P.J. Renwick, J. Trussler, E. Ostad-Saffari, H. Fassihi, C. Black, P. Braude, C.M. comparison of short and long GnRH agonist protocols using recombinant
Ogilvie, S. Abbs, Proof of principle and first cases using preimplantation gonadotrophins for IVF/ICSI treatments, Reprod. Biomed. Online 16 (2008)
genetic haplotyping: a paradigm shift for embryo diagnosis, Reprod. Biomed. 632–639.
Online 13 (2006) 110–119. [30] H.H. Li, U.B. Gyllensten, X.F. Cui, R.K. Saiki, H.A. Erlich, N. Arnheim,
[16] C. Spits, C. Le Caignec, M. De Rycke, L. Van Haute, A. Van Steirteghem, I. Amplification and analysis of DNA sequences in single human sperm and
Liebaers, K. Sermon, Optimization and evaluation of single-cell whole-genome diploid cells, Nature 335 (1988) 414–417.
multiple displacement amplification, Hum. Mutat. 27 (2006) 496–503. [31] C.C. Hung, C.N. Lee, C.H. Chang, Y.J. Jong, C.P. Chen, W.S. Hsieh, Y.N. Su, W.L. Lin,
[17] C. Gutierrez-Mateo, J.F. Sanchez-Garcia, J. Fischer, S. Tormasi, J. Cohen, S. Genotyping of the G1138A mutation of the FGFR3 gene in patients with
Munne, D. Wells, Preimplantation genetic diagnosis of single-gene disorders: achondroplasia using high-resolution melting analysis, Clin. Biochem. 41
experience with more than 200 cycles conducted by a reference laboratory in (2008) 162–166.
the United States, Fertil. Steril. 92 (2008) 1544–1556. [32] Y.N. Su, C.N. Lee, C.C. Hung, C.A. Chen, W.F. Cheng, P.N. Tsao, C.L. Yu, F.J. Hsieh,
[18] E. Unsal, Y. Aktas, O. Uner, I.A. Baltac, S. Ozcan, F. Turhan, V. Baltaci, Successful Rapid detection of beta-globin gene (HBB) mutations coupling heteroduplex
application of preimplantation genetic diagnosis for Leigh syndrome, Fertil. and primer-extension analysis by DHPLC, Hum. Mutat. 22 (2003) 326–336.
Steril. 90 (2008) 2017.e11–2017.e13. [33] C.A. Mein, B.J. Barratt, M.G. Dunn, T. Siegmund, A.N. Smith, L. Esposito, S.
[19] H. Van de Velde, I. Georgiou, M. De Rycke, R. Schots, K. Sermon, W. Lissens, P. Nutland, H.E. Stevens, A.J. Wilson, M.S. Phillips, N. Jarvis, S. Law, M. de Arruda,
Devroey, A. Van Steirteghem, I. Liebaers, Novel universal approach for J.A. Todd, Evaluation of single nucleotide polymorphism typing with invader
preimplantation genetic diagnosis of b-thalassaemia in combination with on PCR amplicons and its automation, Genome Res. 10 (2000) 330–343.
HLA matching of embryos, Hum. Reprod. 19 (2004) 700–708. [34] M.P. Reboul, L. Higueret, N. Biteau, A. Iron, Rapid and accurate detection of the
[20] C. Moutou, N. Gardes, J.C. Nicod, S. Viville, Strategies and outcomes of PGD of CFTR gene mutation 1811+1.6 kbA>G by real-time fluorescence resonance
familial adenomatous polyposis, Mol. Hum. Reprod. 13 (2007) 95–101. energy transfer PCR, Mol. Cell. Probes 19 (2005) 358–362.
[21] F. Fiorentino, A. Biricik, A. Nuccitelli, R. De Palma, S. Kahraman, M. Iacobelli, V. [35] Preimplantation Genetic Diagnosis International Society, Guidelines for good
Trengia, D. Caserta, M.A. Bonu, A. Borini, M. Baldi, Strategies and clinical practice in PGD, Reprod. Biomed. Online 9 (2004) 430–434.