Download Identification of the five most common cystic fibrosis mutations in

Molecular Human Reproduction vol.2 no.3 pp. 203-207, 1996
Identification of the five most common cystic fibrosis mutations in
single cells using a rapid and specific differential amplification
system
Graeme Scobie13, Bridget Woodroffe2, Simon Fishel2 and Noor Kalsheker1
n
Clinical Chemistry, Dept. of Clinical Laboratory Sciences, Queen's Medical Centre, University Hospital, Nottingham NG7
2UH, and 2NURTURE, Nottingham University, Queen's Medical Centre, Nottingham, UK
^o whom correspondence should be addressed
We describe a rapid and specific differential amplification system which can detect five of the most common
cystic fibrosis mutations from a single cell. In the first round of the polymerase chain reaction (PCR), regions
of exons 4, 10 and 11 of the cystic fibrosis transmembrane conductance regulator (CFTR) gene containing the
mutations AF508, G551D, R553X, G542X and 621 + 1 O T were co-amplified in a single multiplex PCR. To
identify potential contamination, we included external amplification primers for the polymorphic human
tyrosine hydroxylase (HUMTH01) locus as a fingerprint for the sample. In the second round of PCR, detection
of any of the five mutations was achieved using the amplification refractory mutation system (ARMS) in two
separate reactions, each containing nested amplification primers for either wild type or mutant sequence. A
separate second round PCR for the fingerprinting was performed with nested HUMTH01 primers. Using this
procedure we have successfully and accurately detected five cystic fibrosis mutations in 30 single cells with
a failed amplification rate of 7% and a contamination rate of 4.6% and that PCR failure or possible
contamination will also be easily detected. This procedure allows detection of the five most common point
mutations and small deletions responsible for cystic fibrosis from a single cell in <8 h which could be
applicable to preimplantation diagnosis in human embryos.
Key words: cystic fibrosis/preimplantation diagnosis/lymphocytes
Introduction
Cystic fibrosis (CF) is one of the most common autosomal
recessive disorders inherited in man which results in a steady
decline of health until death in the second or third decade of
life. The fact that CF is extremely heterogeneous at the genetic
level, with more than 400 mutations identified to date (Mercier
et al, 1994), makes early genetic screening for the disease
extremely difficult and time-consuming. The most common
mutation, the AF508, results from a deletion of a codon
corresponding to the amino acid phenylalanine at position 508
of the CFTR gene (Riordan et al, 1989). Although this
mutation accounts for ~50-80% of cases (CF Genetic Analysis
Consortium, 1990), depending on the population being studied,
in the UK around 10% of the remainder present with one or
two of the four other most common mutations, these being
the G551D Gly-Asp, R553X Arg-Stop (Cutting et al, 1990),
the G542X Gly-Stop (Kerem et al, 1990) and the G-T splice
mutation 621 + 1 O T (the first intronic base in the splice
donor site flanking the 3' end of exon 4) (Zielenski et al,
1991). With increasing interest and demand from parents to
have offspring free of disease such as CF, screening of embryos
prior to implantation in the uterus requires a rapid, sensitive
and accurate procedure. Four out of five of the common CF
mutations are single base pair changes. Only the AF508
mutation (a 3 base pair deletion) has been successfully detected
in single blastomeres from human embryos prior to implantation (Handyside et al, 1992). In order to be able to offer a
© European Society for Human Reproduction and Embryology
comprehensive screening procedure for CF, the ability to detect
as many of the other common mutations in a single rapid test
would obviously be beneficial when both parents are not
carriers of the AF508 mutation.
Detection of CF mutations by a differential amplification
procedure, the amplification refractory mutation system
(ARMS) (Feme et al, 1992) has the potential to detect five
of the most common mutations in the standard kit, and up to
12 mutations in the standard plus kit (Cellmark), but neither
has been adapted to detect these mutations from a single cell
as a diagnostic procedure. The ARMS procedure relies on the
introduction of a mismatch between a polmerase chain reaction
(PCR) primer, especially at the 3' end, and the template such
that Taq polymerase will not extend from it and a product is
not generated. This is in contrast to the normal PCR where
there is no mismatch at the 3' end. By using two PCR reactions
(tubes A and B), with each tube containing either the normal
or mutant primer for a specific mutation (Figure 1), the
presence or absence of the appropriate band will allow a
diagnosis to be made. Using this simple test, the mutations in
carrier parents can be identified prior to in-vitro fertilization
(IVF) treatment. One or two blastomeres from fertilized
embryos can then be rapidly screened for these mutations, and
those free of the disease can be used for implantation into
the mother.
Using this procedure, we have shown it is possible to
genotype a single cell for any of the five most common
mutations responsible for CF, which when used in parallel
203
Cystic fibrosis diagnosis in single cells
G.Scobie et al.
A
200 bp
B
11S
10S
4S
621 + I O T
G551D0U53X)
10AS
4AS
11 AS
R553X
G542X
QS51D
S21 + 10 >T
AFSOS
AFSOS
L
G54ZX
Figure 2. Schematic diagram of external CFTR primers for exons
4, 10 and 11 and the relative positions of the cystic fibrosis
mutations 621 + 1 O T , AF508, G542X, R553D and G551X.
Amu tabc
A
B
AmpUOcatiM p r t e t n
621+1G>T
normal
AFSOS
Dormal
G551D(R553X)
mount
C342X
mutant
621+1G>T
AFSOS
G591D(R5S3X)
GS41X
mutant
mutant
normaj
Figure 1. Pattern of polymerase chain reaction products amplified
in tubes A and B from normal DNA using the ARMS standard
cystic fibrosis kit (top). Amplification primers contained in each
individual tube (bottom).
with IVF treatment and blastomere biopsy, would ensure the
implantation of a non-affected embryo. Previous comparisons
of several different methods of detecting CF mutations have
shown the ARMS procedure to be the most reliable and
sensitive compared to PCR/restriction enzyme digest/polyacrylamide gel electrophoresis and dot blot hybridization
when used as a potential screening method (Miedzybrodzka
etal., 1994).
As a precaution against the risk of extraneous DNA contamination, as well as using restriction enzymes to decontaminate
the PCR mastermixes, we incorporated the human tyrosine
hydroxylase polymorphic locus (HUMTH01) as a simple
fingerprint to detect any contamination that may be present in
the sample under investigation (Puers et al., 1993). Extraneous
contamination is a major problem and has been reported by
most other groups involved in this work (Holding et al., 1993;
Ray et al, 1993). Other researchers have shown that in nondecontaminated PCR mixes, as many as 70% of blanks have
amplified the appropriate fragment where no DNA was added
(Liu et al., 1992). This shows the necessity for strict decontamination protocols and extra procedures to detect any extraneous
contamination that may be present.
Materials and methods
Contamination control
All plastic consumables were UV-irradiated for at least 1 h to remove
any potential DNA contamination. Single cell isolation was set up in
an isolated room and primary PCRs were carried out in a sterile
vertical flow tissue culture hood (type II). Secondary PCRs were set
up in a separate room with a different set of pipettes and filter tips
(Varawalla et a!., 1991; Holding et al., 1993; Monk et al., 1993).
204
CF cell lines
Four lymphoblastoid cell lines with genotypes AF5O8/AF5O8, G542X/
G542X, G551D/R553X and 621+ l>G/AF508 were purchased from
Coriell Cell Repositories (Camden, NJ, USA) and one cell line
with genotype 621 + lG>T/normal was purchased from European
Collection of Animal Cell Cultures (Porton Down, UK). All cell lines
were maintained in Roswell Park Memorial Institute (RPMI) 1640
medium supplemented with 10% fetal calf serum, 2 mM L-glutamine,
100IU penicillin/streptomycin and 2 Hg/ml fungizone. Single lymphocytes were isolated using an Olympus CK5 micromanipulator and
40 u.m microneedle. Individual cells were washed twice in fresh
minimal essential medium (MEM) before being transferred to a 0.5 ml
PCR tube containing 10 u.1 sterile saline and stored at -20°C until
required.
Polymerase chain reaction
Figure 2 shows a schematic diagram of exons 4, 10 and 11 of the
CFTR gene with the positions of the five most common mutations
and the design of the amplification primers used in the first round of
PCR. Table I lists the PCR primers used in all primary and secondary
PCRs including the CF ARMS primers contained in the diagnostic
kit and their relevant concentrations.
Primary PCR
Dual amplification with either CFTR exon 4, 10 or 11 and the external
HUMTH01 primers was performed for individual CF mutation
analysis (or CF mutations in the same exon). Multiplexing of all
three CF exons and HUMTH01 was also carried out for the detection
of compound heterozygotes. All PCR primers were used at a concentration of 0.4 U.M per 50 u.1 reaction. Each primary PCR reaction
consisted of 10 mM Tris-HCl pH 9.0, 50 mM KC1, 2.5 mM MgCl2,
1% Triton X-100,0.4 U.M each primer, 0.2 mM dNTPs and 2.5 IU Taq
polymerase (Promega, Southampton, UK). All PCR mastermixes were
decontaminated with the restriction enzymes Mboll (CF) and Styl
(HUMTH01) for at least 3 h at 37°C. Primary mastermix (39 ul) was
added to the 10 u.1 saline/single cell and overlayed with 30 ul mineral
oil. Each tube was heated to 95°C for 5 min then chilled on ice. Taq
polymerase (1 u.1 of 2.5 IU) was carefully added through the oil and
35 cycles of 94°C for 1 min, 55°C for 30 s and 72°C for 30 s with
a final extension at 72°C for 10 min performed on a thermal cycler
(Biometra, Maidstone, Kent, UK).
Secondary PCR
I u.1 of each primary PCR reaction was transferred to tubes A and B
of the CF ARMS kit and to one tube containing the nested HUMTH0I
primers. Secondary PCRs were carried out for 22 cycles as previous
but with the annealing temperature increased to 60°C. Detection of
the ARMS products was by electrophoresis on a 3% agarose gel, and
the polymorphic fingerprinting on a 6% acrylamide gel.
Cystic fibrosis diagnosis in single cells
G.Scobie at al.
Table I. Sequences of pnmary and secondary polymerase chain reaction (PCR) primers including ARMS kit primers and their relevant uM concentrations
(compared to a standard concentration of 0.86 nM per reaction)
CFTR Exon
Sequence
Size (bp)
Primary PCR primers
Exon 4
Exon 10
Exon II
HUMTH01
5'5'5'5'5'5'5'5'-
CAAGTCTTATTTCAAAGTACCAAG
CAGCTCACTACCTAATTTATGACA
AAGTGAATCCTGAGCGTGATTTGATAATGA
CACAGTAGCTTACCCATAGAGGAAACATAA
TATTTAATGATCATTCATGACATTT
TAAAGCAATAGAGAAATGTCTGTA
ATTCAAAGGGTATCTGGGCTCTGG
GTGGGCTGAAAAGCTCCCGATTAT
Mutation and sequence
479
380
425
179-203
Size (bp)
Tube
Concentration
160
157
A/B
A
B
1.0
1.0
2.0
A/B
2.0
Secondary PCR primers
AF508
C: 5'- GACTTCACTTCTAATGATGATTATGGGAGA
N: 5'- GTATCTATATTCATCATAGGAAACACCAC
M: 5'- GTATCTATATTCATCATAGGAAACACCATT
Exon II
C: 5'- TAAAATTTCAGCAATGTTGTTTTTGACC
G551D/R553X
N: 5'- GCTAAAGAAATTCTTGCTCGTTGCC
M: 5'- AGCTAAAGAAATTCTTGCTCGTTGCT
285
286
I
A
2.0
1.0
G542X
N: 5'- ACTCAGTGTGATTCCACCTTCTAC
M: 5'- CACTCAGTGTGATTCCACCTTCTCA
256
257
B
A
1.0
1.0
621 + 1 O T
C: 5'- TCACATATGGTATGACCCTCTATATAAACT
N: 5'- TGCCATGGGGCCTGTGCAAGGAAGTATTCC
M: 5'- TGCCATGGGGCCTGTGCAAGGAAGTATTCA
380
380
A/B
A
B
0.5
0.5
0.5
HUMTHOI
5'- TGATTCCCATTGGCCTGTTCCTCC
5'- TGGCCCACACAGTCCCCTGTACAC
123-147
C = common primer, N = normal primer, M = mutant primer.
1 2
Results
Our results from 30 single cells containing CF mutations show
that primary amplification of either single CFTR exons or
multiplexing all 3 CFTR exons (exons 4, 10 and 11) in a
single reaction and subsequent detection of homozygous or
compound heterozygous mutations in a secondary PCR can
be readily achieved within 6-8 h from a single cell.
Figure 3 shows the detection of individual AF508,
G551D, R553X and G542X, 621 + 1 O T mutations using
the CF ARMS kit after primary amplification of individual
CFTR exons containing these mutations, and the corresponding
fingerprint from the same cells (Figure 4). Our initial results
using single exon 10 amplification and detection of the
homozygous AF508 mutation without fingerprinting showed a
slightly higher contamination rate (normal AF508 signal) than
reported by Ray et al. (1993) and Holding et al. (1993). By
adopting strict protocols including separate rooms for primary
and secondary PCR set-up, the overall contamination rate
was 4.6% (unpublished observation). The number of failed
amplifications was 7% which was determined by the absence
of PCR products in both tubes A and B. Some of the failed
PCRs were probably due to failed transfer of the single cell
to the primary PCR tube. One drawback of using single exon
3 4 5
6 7 8 9
10 0 X
621+1G>T
G551D/R553X
G542X
AF508
Figure 3. Detection of the five individual cystic fibrosis mutations
in single cells by dual amplification of the corresponding CFTR
exon (exons 4, 10 or II) and genotyping using the ARMS kit.
Tubes A and B for a normal genotype (lanes I and 2), then tubes
A and B for the mutations: AF5O8/AF5O8 (lanes 3 and 4),
621 + lOT/normal (lanes 5 and 6), G551D/R553X (lanes 7 and
8), and G542X/G542X (lanes 9 and 10).
amplification is that in the case of the homozygous AF508
mutation, the normal ARMS A tube appears blank which would
be unacceptable due to the possibility of failed amplification in
that tube.
205
Cystic fibrosis diagnosis in single cells
G.Scobie et al.
00
110
Figure 4. DNA fingerprint from individual single cells containing
the previous five cystic fibrosis mutations. Lanes 1-5 are cells
containing the mutations AF508/AF508, 621 + lG>T/normal,
G551D/R553X, G542X/G542X and 621 + 1OT/AF508
respectively. Actual fingerprints are the lower bands in the size
range 123-147 bp.
12
3
4 5
6 7 8 9 10 11 12
621+1G>T
G551D/R553X
G542X
AF508
Figure 5. Detection of the same five mutations using a multiplex
containing external primers for exons 4, 10 and 11, and subsequent
genotyping using the ARMS kit Tubes A and B for a normal
genotype (lanes 1 and 2), then tubes A and B for the mutations
AF5O8/AF5O8 (lanes 3 and 4), 621+ lG>17normal (lanes 5 and 6),
G551D/R553X (lanes 7 and 8), G542X/G542X (lanes 9 and 10)
and 621 + 1OT/AF508 (lanes 11 and 12).
To avoid the possibility of human error we included amplification primers for all three exons (and also the HUMTH01
locus primers) in the initial primary PCR (Figure 5). This
multiplex assures a product is always present in both ARMS
tubes. We found that there was no difference in the quality of
results by amplifying two exons instead of three which would
still achieve a PCR product in both tubes. Using this multiplex,
all cells diagnosed gave the correct ARMS pattern as expected.
However we did find that in some instances the intensities of
bands in the same sample did show some variation. There are
two possible explanations for this phenomenon: the hybridization rate of the ARMS primers to target sequence and the rate
at which the bases at the 3' end of the ARMS primer form a
suitable substrate for Tag polymerase (Wu et al, 1989). In
206
addition, the relative concentrations of each ARMS primer in
the multiplex is different to that used in single ARMS reactions,
and competition for reactants in the later stages of the PCR
can affect the total yield of products (Feme et al, 1992).
However there is a high degree of sensitivity and specificity
with the ARMS multiplex for cystic fibrosis (see below).
Under the conditions used here the mutant and normal ARMS
primers for the G551D mutation also detects the R553X
mutation such that the G551D/R553X compound heterozygote
is indistinguishable from a G551D homozygote (confirmed by
sequencing). This is most likely due to the normal G551D
ARMS primer, which is also destabilized at position-2 (Table
I), incurring a mismatch at position-6 caused by the R553X
C-T mutation, such that a PCR product is not observed with
the B tube (wild type) and a diagnosis of homozygous G551D
would be inferred. Since both parents would be typed for the
CF mutations prior to preimplantation diagnosis, it would be
possible to distinguish between homozygous G551D and
compound G551D/R553X heterozygotes.
Discussion
Due to the rapid progress in the identification of disease
causing mutations, it is apparent that a large percentage of
diseases are caused by heterogeneous single point mutations
which can be difficult to detect and may take several days
to yield a result. In the case of preimplantation diagnosis of
CF, four out of the five most common mutations are due
to single base changes and would require a diagnosis within
8-10 h for it to be useful. The ARMS procedure used here
has this advantage in that it is capable of detecting the
621 + 1 O T , AF508, G551D, R553X and the G542X mutations within 6-8 h from a single cell. Multiplex PCR has also
been used to identify the AF508 and W1282X mutations in
the Jewish population from oocytes and single blastomeres
(Avner et al., 1994). However this procedure requires further
analysis including heteroduplex formation and restriction
enzyme digest to obtain a result, whereas the ARMS (which
can also detect the W1282X in the standard plus kit) does not.
However, the ARMS procedure has its own problems in
that it is not applicable to all mutations, and several ARMS
primers may have to be tried and optimized before they can
be used successfully. The Cellmark CF ARMS kit used here
has used different concentrations of some of the primers to
optimize the product concentrations (Table I). In addition,
depending on the annealing temperature or the type of PCR
block in use, occasional non-specific bands are observed due
to the mutant PCR primer binding to the wild type allele
(Figure 4, lane 3). This can easily be overcome by a small
increase in the annealing temperature. It should be possible to
design and test the ARMS assay for preimplantation diagnosis
of other single gene disorders and we have shown this to be
the case in other disorders such as a-1 antitrypsin deficiency
(unpublished observation). The specificity and sensitivity of
the ARMS kit have both been previously reported to be 100%
(631 tests) (Ferrie et al, 1992). When these techniques are
applied to single cells we have assumed that the specificity
will still be the same. However, the risk of contamination by
Cystic fibrosis diagnosis in single cells
extraneous DNA for example is a real problem and potentially
will give rise to false results.
For preimplantation diagnosis, the parents would be accessible for blood sampling and we therefore included a basic
DNA fingerprint to exclude extraneous DNA contamination.
This has proved useful in detecting contamination but it too
may have its problems in interpretation since it is necessary
to resolve a 2-4 bp difference in fragment size on a standard
polyacrylamide gel. However, it allows potential contamination
to be detected readily and offers a greater degree of certainty
about the origin of the DNA being amplified. We thus
demonstrate that it is possible to detect five common CF
mutations in a single cell rapidly and accurately.
G.Scobie et al.
Varawalla, N.Y.. Dokras. A.. Old. J.M. et al. (1991) An approach to
preimplantation diagnosis of pVthalassaemia. Prenatal Diagn., 11, 775-785.
Wu. D.Y.. Vgozzoli, L.. Pal. B.K. and Wallace, R.B. (1989) Allele specific
enzymatic amplification of P-globin genomic DNA for diagnosis of sickle
cell anaemia. Proc. Natl. Acad. Sci. USA. 86, 2757-2760.
Zielenski. J., Bozon, D.. Kercm, B.S. et al. (1991) Identification of mutations
in exons I through 8 of the cystic fibrosis conductance transmembrane
regulatory gene. Genomics, 10, 229-235.
Received on August 3, 1995; accepted on December 8, 1995
Acknowledgements
We thank the University Hospital Trustees for financial support of
this project and Cellmark for providing the CF ARMS kits.
References
Avner, R., Laufer, N., Safran, A. et al. (1994) Preimplantation diagnosis of
cystic fibrosis by simultaneous detection of the W1282X and AF508
mutations. Hum. Reprod., 9, 1676-1680.
CF Genetic Analysis Consortium (1990) Worldwide survey of the AF508
mutation - report from the CF Genetic Analysis Consortium. Am. J. Hum
Genet., 47, 354-359.
Cutting, G.R., Kasch, L.M., Rosenstein, BJ. et al. (1990) A cluster of cystic
fibrosis mutations in the first nucleotide-binding fold of the cystic fibrosis
conductance regulatory protein. Nature, 346, 366-369.
Feme, R.M.. Schwartz, M.J., Robertson, N.H. et al. (1992) Development.
multiplexing, and application of ARMS tests for common mutations in the
CFTR gene. Am. J. Hum. Genet., 51, 251-262
Handyside, A.H., Lesko, M.S.. Tarin, JJ. et al. (1992) Birth of a normal girl
after in vitro fertilization and preimplantation diagnostic testing for cystic
fibrosis. N. Eng. J. Med., 327, 905-909.
Holding, C, Bentley, D., Roberts, R. et al. (1993) Development and validation
of laboratory procedures for preimplantation diagnosis of Duchenne
muscular dystrophy. J. Med. Genet., 30, 903-909.
Kerem, B.S., Zielenski, J., Markiewicz, D. et al. (1990) Identification of
mutations in regions corresponding to the two putative nucleotide (ATP)binding folds of the cystic fibrosis gene. Proc. Natl. Acad. Sci. USA, 87,
8447-8451.
Liu, J., Lissens, W., Devroey, P. et al. (1992) Efficiency and accuracy of the
polymerase-chain-reaction assay for cystic fibrosis allele AF508 in single
cells. Lancet, 339, 1190-1192.
Mercier, B., Lissens, W., Novelli, G. et al. (1994) A cluster of cystic fibrosis
mutations in exon 17b of the CFTR gene: a site for rare mutations. J. Med.
Genet., 31, 731-734.
Miedzybrodzka. Z.H., Yin, Z.. Kelly, K. and Haiters, N. (1994) Evaluation of
laboratory methods for cystic fibrosis carrier screening: reliability,
sensitivity, specificity, and costs. J. Med. Genet., 31, 545-550.
Monk, M.. Kenealy. M. and Mohadjerani, S. (1993) Detection of both the
normal and mutant alleles in single cells of individuals heterozygous for
the sickle cell mutation. Prenatal Diagn., 13, 45-53.
Puers, G. Hammond, H.A., Jin, L. et al. (1993) Identification of repeat
sequence heterogeneity at the polymorphic short tandem repeat locus
HUMTHOI[AATG]n and reassignment of alleles in population analysis by
using a locus-specific allelic ladder. Am. J. Hum. Genet., S3, 953-958.
Ray, P., Handyside. A., Harper, J. et al. (1993) Pregnancies following
preimplantation diagnosis of cystic fibrosis (AF508) and Lesch-Nyhan
syndrome: assessment of the risks of misdiagnosis. In Preimplantation
Diagnosis of Inherited Disease. Campus Workshop, 23-24 September.
ESHRE.
Riordan, J.R., Rommens. J.M., Kerem, B.S. et al. (1989) Identification of the
cystic fibrosis gene: cloning and characterization of complementary DNA.
Science, 245, 1066-1073.
207