Human Reproduction Vol.21, No.8 pp. 2047–2051, 2006
doi:10.1093/humrep/del064
Advance Access publication May 31, 2006.
Single-sperm analysis for haplotype construction
of de-novo paternal mutations: application to PGD
for neurofibromatosis type 1
G.Altarescu1, B.Brooks2, Y.Kaplan1, T.Eldar-Geva2, E.J.Margalioth2, E.Levy-Lahad1,3
and P.Renbaum1,4
1
Medical Genetics Unit, 2IVF Unit, Zohar PGD Laboratory Shaare Zedek Medical Center and 3Hebrew University-Hadassah Medical
School, Jerusalem, Israel
4
To whom correspondence should be addressed at: Medical Genetics Unit, Shaare Zedek Medical Center, Jerusalem 91031, Israel.
E-mail: [email protected]
BACKGROUND: Neurofibromatosis type 1 (NF1) is an autosomal dominant disorder caused by mutations in the
neurofibromin gene. Approximately, 50% of cases are caused by de-novo mutations. Even when the NF1 mutation is
known, accuracy of PGD is highly enhanced by simultaneous analysis of linked markers. In a childless couple
referred to PGD, the male carried a de-novo mutation, precluding the possibility of typing relatives to establish the
mutation-associated haplotype. We developed a single-sperm haplotype analysis strategy to establish the haplotype
linked to the NF1 mutation. METHODS: Spermatozoa from freshly ejaculated semen were used as a substrate for
multiplex PCR on single sperm. RESULTS: In addition to the NF1 mutation, six informative polymorphic markers
flanking the NF1 gene (D17S1294, D17S1849, D17S841, D17S975, NF1TG2 and NF1AC5) were linked to individual
alleles in single sperm from the affected male. CONCLUSIONS: Single-sperm analysis established the haplotypes of
both mutant and wild-type NF1 alleles and enabled the implementation of a PGD protocol using polymorphic marker
analysis. This method is generally applicable to PGD for any disease in which the haplotype of paternal mutations
cannot be determined by typing relatives.
Key words: single-sperm PCR/NF1/PGD/polymorphic markers
Introduction
Neurofibromatosis type 1 (NF1, von Recklinghausen syndrome) has an estimated birth incidence of about 1 in 3500
individuals worldwide and is caused by autosomal dominant
mutations in the neurofibromin gene (Hart, 2005; Ward and
Gutmann, 2005). NF1 is traditionally regarded as a neurocutaneous disorder (phakomatosis), but disease manifestations
occur in many other organ systems as well (Rose, 2004; Hart,
2005; Korf, 2005; Ward and Gutmann, 2005). The clinical picture varies from mild to severe, even in patients carrying the
same NF1 mutation, and there is significant intrafamilial variability (Carey and Viskochil, 1999; Viskochil, 2002). Although
the only consistent features of this disorder are cafe-au-lait
spots and fibromatous skin tumours, several other organs are
often found to be affected with NF1-related tumours (Rose,
2004; Page and Franklin, 2004). There is also a predisposition
towards malignant transformation of the benign tumours associated with NF1 in an estimated 3–15% of cases (Hart, 2005;
Hasiotou et al., 2005; Kim et al., 2005; Takazawa et al., 2005).
The neurofibromin gene, NF1, is located on the long arm of
chromosome 17. It is a particularly large gene, extending over
359 kb and including 58 exons. Over 500 different mutations
have been identified in the NF1 gene, the majority of which are
family specific (Carroll and Stonecypher, 2005). Many types of
NF1 mutations have been reported, including large and small
deletions and insertions, missense, non-sense, splicing mutations
and chromosomal aberrations (Carroll and Stonecypher, 2004).
Although protein-truncation testing (PTT) and sequencing can
detect up to 95% of mutations in clinically diagnosed NF1
patients, diagnostic criteria remain clinical, and mutation analysis
is not commonly performed owing to its high cost. In families
with more than one affected relative, linkage analysis provides a
much simpler and cheaper means of molecular diagnosis. However, over 50% of NF1 mutations occur de novo, precluding conventional linkage analysis in over half of NF1 cases.
PGD for NF1 has been reported by two different centres. In
the first instance, maternal mutations were tested by sequential
PCR of polar bodies 1 and 2 in conjunction with three linked
polymorphic markers in introns 19A, 27A and exon. Two unrelated couples underwent two cycles each resulting in a pregnancy in each couple (Verlinsky et al., 2002). In the second
report, blastomeres were analysed for the NF1 mutation in
© The Author 2006. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. All rights reserved.
For Permissions, please email: [email protected]
2047
G.Altarescu et al.
conjunction with two polymorphic markers (D17S1800 and
D17S841) resulting in an unaffected male (Spits et al., 2005).
Both prenatal and preimplantation diagnoses of NF1 rely on
previous determination of the specific mutation, or in familial
cases, the linked haplotype. In PGD, genetic analysis is performed on single cells: the first and second polar bodies or
embryonic blastomeres. Molecular testing is done by singlecell PCR, which is prone to allele dropout (ADO), i.e. failure
of amplification of one of the alleles. Failure of amplification
of the mutant allele can result in an erroneous diagnosis in
which a mutant embryo is mistaken for a wild-type embryo.
This critical problem can be minimized by co-amplifying linked,
polymorphic markers because it is unlikely that ADO will
occur in all reactions. In blastomeres, simultaneous testing of
linked markers reduces the rate of misdiagnosis from 20–30%
to 3% (Verlinsky and Kuliev, 2000). Analysis of two blastomeres from each embryo has also been used for this purpose
but may result in lower implantation and pregnancy rates
(Cohen, 2005).
To allow the use of linked markers for PGD in men with denovo mutations, we have developed a strategy to determine
haplotypes from single sperm. We report a couple with a denovo NF1 mutation in a mildly affected male, who presented to
our clinic for PGD. Linkage analysis was not possible because
this couple had neither affected relatives nor previous pregnancies. To establish linkage between the new NF1 mutation and
polymorphic markers, we isolated and analysed single sperm
(Huang et al., 1995; Zhang et al., 1995; Brohede et al., 2004),
using single-cell PCR, for both the new NF1 mutation and
NF1-linked markers.
Materials and methods
Single-sperm isolation, IVF and blastomere biopsy procedure
Spermatozoa were prepared from freshly ejaculated semen using discontinuous gradient centrifugation (Isolate, Irvine Scientific, Santa
Ana, CA, USA). Separated sperm were washed in culture medium
(Flusing, Medi-Cult, Copenhagen, Denmark), and small numbers of
sperm were pipetted into a 2.5-μl droplet of polyvinylpyrrolidone
(PVP, Medi-Cult) under mineral oil. Single spermatozoa were then
transferred to individual 0.5-μl droplets of P1 culture medium (Irvine
Scientific) under mineral oil. Droplets containing single-sperm cells
were aspirated using finely drawn Pasteur pipettes and were transferred to PCR tubes containing 5 μl of alkaline lysis buffer (Thornhill
et al., 2001).
IVF treatment was performed by pituitary down-regulation with
GnRH analogue, followed by controlled ovarian hyperstimulation
with rFSH (Merviel et al., 2005). Vaginal ultrasound-guided oocyte
retrieval was performed 34 h after HCG injection. Cumulus–oocyte–
complexes (COCs) were identified, washed and transferred to organ
culture dishes containing equilibrated culture medium (Medi-Cult)
and placed in an incubator with 5% CO2. Oocytes were denuded with
hyaluronidase (Sigma-Aldrich, USA), and ICSI was performed on
each mature egg (M2). Injected oocytes were transferred to P1
medium (Irvine Scientific) under mineral oil. Fertilized oocytes were
cultivated in P1 medium for a further 48 h. Blastomere biopsy was
performed using mechanical zona slitting (Verlinsky and Kuliev,
2005). Blastomeres were carefully transferred to separate 0.5-ml tubes
containing 5 μl of proteinase K lysis buffer (Thornhill et al., 2001),
and embryos were transferred to fresh culture medium for blastocyst
2048
development (Irvine Scientific). A sample of culture medium from
the same droplet that contained each biopsied blastomere was used
as a ‘no template control’ (NTC).
Molecular analysis
A couple presented to our clinic for PGD with NF1, where the mildly
affected male was known to carry a de-novo NF1-splicing mutation,
6365(-2)A>G. The couple was childless and had no previous pregnancies, the woman was 25 years old. The male was previously shown to
harbour a de-novo NF1 splice mutation upstream of exon 34, 6365
(-2)A>G. In addition to the new NF1 mutation, polymorphic markers
flanking the NF1 gene were amplified in a single multiplex PCR for
each individual sperm. The reactions contained 1 U SuperTaq
polymerase (JMR, Kent, UK), 0.32 mM dNTPs, 10% DMSO, 0.1 μM
each forward and reverse primers (Table I), in a buffer supplied by the
manufacturer containing 1.5 mM MgCl2 and was cycled using a
touchdown protocol (Verlinsky et al., 2004). Aliquots (1.5 μl) from
the multiplex reaction were used as templates for individual heminested PCR using a nested primer (0.28 μM) 5′ fluorescently labelled
with 6-FAM, HEX (MWG-Biotech AG, Ebersberg, Germany) or
NED (Applied Biosystems, CA, USA) with one outside primer (0.28
μM) for each marker (Table I). Each nested PCR was amplified with
0.6 U coloured Taq polymerase (SILEKS, Moscow, Russia) and contained 0.24 mM dNTPs, 6% DMSO, with the buffer described above,
for 30 cycles. Amplification products were diluted and run on an ABI
Prism 3100 Avant automated capillary sequencer and analysed using
GeneScan or Genotyper software (Applied Biosystems, Foster City,
CA, USA), relavant peaks were distinguished from stutter peaks
according to standard procedures (User bulletin, ABI Prism linkage
mapping set, November 1997), and peak amplitude thresholds were
set at 50 pts. Analysis of the 6365(-2)A>G NF1 mutation was performed using a hemi-nested primer designed with two mismatches
which create a BsrGI (MBI Fermentas, Vilna, Lituania) restriction site
on the wild-type sequence. Following PCR, the amplified product was
digested with BsrGI, electrophoresed on a 4% Nusieve gel (Cambrex,
Rockland, ME, USA) and visualized by staining with ethidium bromide.
Results
Nine highly polymorphic markers flanking the NF1 gene were
selected for haplotype analysis (Figure 1), and an assay to
identify the 6365(-2)A>G mutation was designed (Figure 2A).
The male was found to be heterozygous for six of the nine polymorphic markers (D17S1294, D17S1849, D17S841,
D17S975, NF-TG2 and NF-AC5, see legend to Figure 1), and
all the six of these informative markers were used in this case.
The proband, his wife and the proband’s parents were each
haplotyped with these markers, and this allowed the identification of the individual alleles of the proband. However, no allele
could be linked to the NF1 mutation because it was a new
mutation in the proband.
We used a multiplex PCR, followed by individual heminested reactions, to establish linkage between the NF1 de-novo
mutation and the polymorphic markers on 14 isolated single
sperm. Digestion with BsrGI for the 6365(-2)A>G mutation
identified eight mutant and five wild-type sperm (data for five
single sperm are summarized in Table II). Sperm 1, 2 and 3
carried the mutant allele, whereas sperm 4 and 6 were wild
type. No amplification was observed in sperm 5 in any reaction
(data not shown). Simultaneous analysis of the six informative
polymorphic markers from each of these sperm (Figure 3)
Single-sperm analysis for NF1 haplotypes in PGD
NF1 AC1
Table I. NF1-flanking polymorphic makers: primer sequences (5′→3′) for
hemi-nested PCR
D17S1294-FOUT
TGGCATGCAATTGTAGTCTC
D17S1849-FOUT
ATAGCGGTACCACCTTGG
D17S841-FOUT
TGGACTTTCTTACATGGCA
D17S975-FOUT
CCTTAGGTCTTTGGCATTTG
NF1TG2-FOUT
TCATGCTCTGCTGTTAAACC
NF1AC5-FOUT
AGTTTGGAGTCCATGGGATA
D17S1863-FOUTa
CACTTGGGCTGTAAGATGTTT
NF1AC1-FOUTa
TGGTCTGAAACTCCTGTCCT
D17S1307-FOUTa
GAGTCTTGTTCTTTCCTAGCC
D17S1249-FIN
GAGGTTGAGCCTGCAATAAG
D17S1849-FIN
GATCCTGGAGAATTCCCTGG
D17S841-FIN
TCAGAAGGTACTTGCCATCA
D17S975-FIN
AGTCATGGGACTTCTCGGCT
NF1TG2-FIN
TCCTGTTTTTAGTGAGTGCGTA
NF1AC5-FIN
AATGTAACAATTGTGGAACTGC
D17S1863-FINa
GTACCTTTGATTTCAATATGCTG
NF1AC1-FINa
TGTCCTCAAGTGATCGGC
D17S1307-FIN
GTAGGAGACCTGCTGCCTTT
D17S1249-R
TTCTTTCCTTACTAAGTTGAGAACG
D17S1849-R
TTGGGCAAGACCCAGTG
D17S841-R
AGGTTAGTAGTCTATGTCACAGCG
D17S975-R
GTGTGCAGTGTTGATTTTCTTT
NF1TG2-R
TCATGCCACTGCACTCTAG
NF1AC5-R
GCTGGGATTACAGGTGTGAG
D17S1863-Ra
CAATTCACTGACTGTGGTTTT
NF1AC1-Ra
CGATCAACTGACACATCCAT
D17S1307-Ra
AGGGCAGAGAAACCTAAGGA
D17S1849*
D17S975*
D17S841*
200Kb
D17S1863
D17S1294*
NF1 AC5*
NF1 mutation
6365(-2)A>G
NF1 TG2*
NF1 gene
D17S1307
Figure 1. Polymorphic markers on chromosome 17 in the region of
the NF1 gene. Asterisks (*) indicate markers which were informative
and used in this case. Markers in the region surrounding the NF1 gene
were obtained from the following databases: D17S841 from GeneAtlas, D17S1294, D17S1849, D17S975, D17S1863 and D17S1307
from GeneCards, and AC1, AC5 and TG2 were identified using the
UCSC genome browser (July 2003 assembly, NCBI Build 34).
Figure 2. Detection of the 6365(-2)A>G NF1 mutation in single
sperm. Amplification and digestion with BsrGI of the 6365(-2)A>G
NF1 mutation (BsrGI digests the wild-type alelle). A, genomic DNA
(M: male, F: female) and B, isolated single sperm; Ma, pUC19MspI
digest DNA size marker (MBI Fermentas).
73 single fibroblasts informative for four different NF1 markers were measured and ranged from 1 to 10%, whereas actual
ADO observed for 10 blastomeres in the PGD cases showed 0
to 3 ADO events per marker (Table III). Only in one nested
PCR was an amplified product found in the medium NTC
(blank) for that particular oocyte, disqualifying that reaction.
Each of the three PGD cycles resulted in the transfer of one to
two wild-type embryos (Table IV). A chemical pregnancy was
achieved in cycle number 2.
FIN, forward inside primer; FOUT, forward outside primer; R, reverse primer.
Markers that were not informative in this couple were not used in this case.
a
Discussion
allowed the mutant and wild-type alleles to be linked to individual haplotypes (Table II). Construction of the patient’s haplotype enabled complete haplotype determination in this family
(Figure 4).
The couple has since undergone three cycles of PGD using
all six informative polymorphic markers in addition to mutation testing. Before the first cycle, the PGD protocol was tested
using all the six markers and the familial mutation in a multiplex PCR in single fibroblasts from an unrelated individual.
Amplification was observed in all reactions. Rates of ADO in
An individual affected with NF1 has a 50% chance of giving
birth to an affected child in each pregnancy. Couples can elect
to have prenatal diagnosis for NF1 by either chorionic villus
sampling (CVS) or amniocentesis. Couples at risk can also
undergo PGD, pre-empting the need for pregnancy termination
in the case of an affected embryo. PGD is an especially attractive option for NF1, not only because of the high recurrence
risk (50%) but also because disease severity is unpredictable.
This may lead to ethical dilemmas in terminating a pregnancy
which could result in a minimally affected child.
2049
G.Altarescu et al.
Table II. Results of single-sperm analysis for an NF1 mutation and linked markers
Marker
Genomic DNA
Sperm 1
Sperm 2
Sperm 3
Sperm 4
Sperm 6
BsrGI digest
D17S1849
TG2
D17S841
D17S1294
D17S975
AC5
G/A
232/234a
137/135
176/174
186/182
242/246
272/266
G (mutant)
No result
137
176
186
242
272
G (mutant)
232
137
176
186
242
272
G (mutant)
232
137
176
186
242
272
A (wild type)
234
135
174
182
246
No result
A (wild type)
234
135
174
182
246
266
Mutant haplotypes are shown in bold-face type, wild-type haplotypes are shown in italics. No amplification products were obtained for
sperm 5 in any reaction.
a
Numbers represent marker allele length in base pairs.
Table III. Results of three cycles of PGD for NF1 in this study
COCs
M2
Fertilized
Embryos developed to six cells
Blastomeres diagnosed
Wild-type embryos
Mutant embryos
Embryos transferred
HCG positive (rising), 2 weeks
Cycle 1
Cycle 2
Cycle 3
9
8
7
5
2
1
1
1
No
18
16
13
6
3
2
1
2
Yes
29
22
19
10
5
3
2
2
No
COCs, Cumulus–oocyte–complexes; NF1, neurofibromatosis type 1.
Figure 3. Haplotype determination of single sperm. Capillary electrophoretograms of amplified fluorescently labelled linked polymorphic markers using paternal genomic DNA and isolated single sperm.
Sperm 1 harbours the mutant allele, whereas sperm 6 harbours the
wild-type allele (Figure 2B). Asterisks (*) indicate mutant alleles, and
paternal alleles are listed under each marker.
Table IV. ADO for six markers and the familial mutation in the NF1 gene
in single fibroblasts and blastomeres
Marker
D17S1294
D17S841
D17S975
TG2
D17S1849
AC5
NF1 6365(-2)A>G
a
ADO
Fibroblastsa
Blastomeresb
6%
1%
9%
10%
Not informative
Not informative
Not informative
1
3
2
1
2
1
0
Of 73 single fibroblasts from an unrelated individual.
Of 10 blastomeres from three cycles of PGD.
b
Figure 4. Haplotype analysis of the family on the basis of haplotypes
determined from sperm analysis and genomic DNA. The haplotypes
of the couple and the paternal parents are presented as determined
from GeneScan analysis (data not shown).
We report using single-sperm analysis for haplotype determination using a multiplex, single-cell PCR for PGD.
Although in theory PGD for this couple could be accomplished
using mutation analysis alone, this would be accompanied by
2050
an unacceptably high error rate due to ADO. Even the low
ADO rates observed here are significantly high for families
electing to undergo PGD to avoid pregnancy termination. A
second possibility would be to biopsy an additional blastomere
for confirmation; however, this may give rise to a reduced
pregnancy rate (Cohen, 2005). By looking at ADO in fibroblasts, we found that while certain markers produced only 1%
ADO, others gave up to 10%. Blastomere analysis yielded
ADO rates of 0–3 events per 10 reactions; however, it should be
noted that only 10 blastomeres were fully analysed (Table III).
ADO is the leading cause of misdiagnosis in PGD (Goossens et al.,
2003; Piyamongkol et al., 2003; Deng et al., 2005) and can be
particularly misleading when diagnosing autosomal dominant
disorders such as NF1, where ADO of the mutant allele yields
a result identical to that of a homozygote wild-type embryo.
Identification of ADO by simultaneous amplification of multiple,
Single-sperm analysis for NF1 haplotypes in PGD
informative and polymorphic markers, allows for the disqualification of these reactions and is essential to achieve maximal
diagnostic accuracy.
NF1 is particularly prone to de-novo mutations perhaps
because of its large size of almost 360 kb. De novo mutations
present a challenge for PGD because of the difficulty in establishing the mutant haplotype when there are no additional
affected relatives. By determining the haplotype of individual
sperm, we were able to set phase and establish the haplotypes
of an affected male with a new NF1 mutation. A similar strategy was recently reported for PGD of Currarino syndrome,
although in that case, haplotypes could also have been confirmed on the basis of linkage in the family (Verlinsky et al.,
2005). We used six polymorphic markers which span almost 5
Mb, a large region known to contain a number of recombination hotspots (HapMap Phase I data, release 16a). Recombination is regularly observed in both polar bodies and blastomeres
in PGD, although we have not yet observed any recombination
events in the NF1 region in these family members or any sperm
or blastomere in this family. Usage of markers on both sides of
the gene, and awareness of the proximities of each marker, in
addition to recombination frequencies in the region allowed
accurate identification of the mutant and wild-type alleles. This
method is generally applicable for establishing the haplotype of
any paternal mutation when linkage cannot be performed
because of lack or unavailability of other affected relatives. For
PGD, this method enables the use of linked markers in these
cases, thus greatly enhancing diagnostic accuracy.
Acknowledgements
We thank Rabbi David and Mrs. Anita Fuld for their generous and
ongoing support and Ms. Hadassa Hartman for her valuable editorial
assistance. This project was approved by the IRB of the Shaare Zedek
Medical Center.
References
Brohede J, Arnheim N and Ellegren H (2004) Single-molecule analysis of the
hypermutable tetranucleotide repeat locus D21S1245 through sperm genotyping: a heterogeneous pattern of mutation but no clear male age effect.
Mol Biol Evol 21,58–64.
Carey JC and Viskochil DH (1999) Neurofibromatosis type 1: a model condition for the study of the molecular basis of variable expressivity in human
disorders. Am J Med Genet 89,7–13.
Carroll SL and Stonecypher MS (2004) Tumor suppressor mutations and
growth factor signaling in the pathogenesis of NF1-associated peripheral
nerve sheath tumors. I. The role of tumor suppressor mutations. J Neuropathol Exp Neurol 63,1115–1123.
Carroll SL and Stonecypher MS (2005) Tumor suppressor mutations and
growth factor signaling in the pathogenesis of NF1-associated peripheral
nerve sheath tumors. II. The role of dysregulated growth factor signaling. J
Neuropathol Exp Neurol 64,1–9.
Cohen J (2005) PGD and IVF. Oral presentation. Sixth International Symposium on Preimplanation Genetics. RBO 10 (Suppl. 2),1.
Deng J, Zhuang GL, Peng WL, Zhou CQ, Li J, Fang C, Li SP, Chen Y and
Tong DY (2005) Closely linked polymorphic marker: successful application
in preimplantation genetic diagnosis for beta-thalassemia. Zhonghua Yi Xue
Yi Chuan Xue Za Zhi 22,391–395.
Goossens V, Sermon K, Lissens W, De Rycke M, Saerens B, De Vos A,
Henderix P, Van D, Platteau P, Van Steirteghem A et al. (2003) Improving
clinical preimplantation genetic diagnosis for cystic fibrosis by duplex
PCR using two polymorphic markers or one polymorphic marker in combination with the detection of the DeltaF508 mutation. Mol Hum Reprod
9,559–567.
Hart L (2005) Primary care for patients with neurofibromatosis 1. Nurse Pract
30,38–43.
Hasiotou M, Danassi-Afentaki D, Stefanaki K, Sfakianos G, Prodromou N and
Moschovi M (2005) Early atypical malignant transformation of a plexiform
neurofibroma in a 4-year-old boy with Neurofibromatosis 1. Pediatr Blood
Cancer 45,76–77.
Huang MM, Erlich HA, Goodman MF and Arnheim N (1995) Analysis of
mutational changes at the HLA locus in single human sperm. Hum Mutat
6,303–310.
Kim JG, Sung WJ, Kim DH, Kim YH, Sohn SK and Lee KB (2005) Malignant
peripheral nerve sheath tumor in neurofibromatosis type I: unusual presentation
of intraabdominal or intrathoracic mass. Korean J Intern Med 20,100–104.
Korf BR (2005) The phakomatoses. Clin Dermatol 23,78–84.
Merviel P, Najas S, Campy H, Floret S and Brasseur F (2005) Use of GNRH
antagonists in reproductive medicine. Minerva Ginecol 57,29–43.
Page TZ and Franklin J (2004) Neurofibromatosis type 1. School Nurse News
21,22–24.
Piyamongkol W, Bermudez MG, Harper JC and Wells D (2003) Detailed
investigation of factors influencing amplification efficiency and allele dropout in single cell PCR: implications for preimplantation genetic diagnosis.
Mol Hum Reprod 9,411–420.
Rose VM (2004) Neurocutaneous syndromes. Mo Med 101,112–116.
Spits C, De Rycke M, Van Ranst N, Joris H, Verpoest W, Lissens W, Devroey P,
Van Steirteghem A, Liebaers I and Sermon K (2005) Preimplantation genetic
diagnosis for neurofibromatosis type 1. Mol Hum Reprod 11,381–387.
Takazawa Y, Sakurai S, Sakuma Y, Ikeda T, Yamaguchi J, Hashizume Y,
Yokoyama S, Motegi A and Fukayama M (2005) Gastrointestinal stromal
tumors of neurofibromatosis type I (von Recklinghausen’s disease). Am J
Surg Pathol 29,755–763.
Thornhill AR, McGrath JA, Eady RA, Braude PR and Handyside AH (2001) A
comparison of different lysis buffers to assess allele dropout from single
cells for preimplantation genetic diagnosis. Prenat Diagn 21,490–497.
Verlinsky Y and Kuliev A (2000). In An Atlas of Preimplantation Genetic
Diagnosis, Section 2, PGD Illustrated, 2nd edn. Parthenon Publishing
Group, p. 157.
Verlinsky Y and Kuliev A (2005) Micromanipulation and biopsy of polar bodies and blastomeres. In An Atlas of Preimplantation Genetic Diagnosis, 2nd
edn. Chapter 2, Taylor and Francis, Parthenon Publishing Group, pp. 15–22.
Verlinsky Y, Rechitsky S, Verlinsky O, Chistokhina A, Sharapova T,
Masciangelo C, Levy M, Kaplan B, Lederer K and Kuliev A (2002) Preimplantation diagnosis for neurofibromatosis. Reprod Biomed Online 4,218–222.
Verlinsky Y, Rechitsky S, Sharapova T, Morris R, Taranissi M and Kuliev A
(2004) Preimplantation HLA testing. JAMA 291,2079–2085.
Verlinsky Y, Rechitsky S, Schoolcraft W and Kuliev A (2005) Preimplantation
diagnosis for homeobox gene HLXB9 mutation causing Currarino syndrome. Am J Med Genet A 134,103–104.
Viskochil D (2002) Genetics of neurofibromatosis 1 and the NF1 gene. J Child
Neurol 17,562–570.
Ward BA and Gutmann DH (2005) Neurofibromatosis 1: from lab bench to
clinic. Pediatr Neurol 32,221–228.
Zhang L, Fischbeck KH and Arnheim N (1995) CAG repeat length variation in
sperm from a patient with Kennedy’s disease. Hum Mol Genet 4,303–305.
Submitted on October 2, 2005; resubmitted on December 29, 2005, February 7,
2006; accepted on February 10, 2006
2051