|
|
||||||||
Articles |
1
The John F. Kennedy Institute, Gl. Landevej 7, 2600 Glostrup, Denmark.
2
Division of Clinical Chemistry and Biochemistry,
University Childrens Hospital, Steinwiesstrasse 75, 8032 Zurich,
Switzerland.
a Author for correspondence. Fax 45-4343-1130; e-mail flg{at}kennedy.dk
| Abstract |
|---|
|
|
|---|
Methods: We developed a method based on PCR in combination with denaturing gradient gel electrophoresis (DGGE) that rapidly scans the six coding sequences and all splice sites of the PTPS gene (PTS) for mutations. This method was used to examine the status of the PTS gene in control samples with known PTS mutations and in five patients with PTPS deficiency.
Results: Two features of the PTS gene posed
particular problems in relation to DGGE analysis: the very high GC
content of exon 1, and a 15-bp poly(dT) stretch in the acceptor splice
site of intron 1. Both problems were solved by special design of
amplification primers. PCR and DGGE conditions were adjusted to allow
simultaneous analysis of all six regions of the PTS
gene. Using this one-step approach, all control mutations were readily
resolved. Among the five PTPS patients, four mutations were identified,
including IVS1-3C
G, IVS2-7T
A, V57del, and V97M (289G
A). The
IVS1-3C
G mutation was shown by reverse transcription-PCR analysis to
produce multiple splice variants.
Conclusions: We have established a fast and reliable screening method for detection of mutations and small deletions/insertions in the PTS gene. This method should be useful for rapid diagnosis of PTPS deficiency in newborns with HPA.
| Introduction |
|---|
|
|
|---|
The synthesis of BH4 is carried out by three
enzymes, GTP cyclohydrolase I (EC 3.5.4.16), 6-pyruvoyltetrahydropterin
synthase (PTPS; EC 4.6.1.10), and sepiapterin reductase (EC 1.1.1.153).
Through the activity of, for example, PAH, BH4 is
oxidized to quinonoid dihydrobiopterin and is recycled to
BH4 by pterin-4
-carbinolamine
dehydratase (EC 4.2.1.96) and dihydropteridine reductase (EC
1.6.99.7) (2). Disorders of the
BH4 system generally are characterized by HPA and
neurological symptoms attributable to lack of the neurotransmitters
dopamine and serotonin. Deficiency of BH4 can be
caused by defects in either the synthesis or the recycling of
BH4. According to the International Database of
Tetrahydrobiopterin Deficiencies, more than one-half (58%) of
all BH4 deficiencies are caused by defects in the
PTPS enzyme (4).
Diagnostic procedures to differentiate between PAH and BH4 deficiencies include measurement of urinary pterins or oral loading tests with BH4(5)(6). Prenatal diagnosis of PTPS deficiency at 1524 weeks gestation has been made on the basis of pterin concentrations in amniotic fluid and enzymatic activity in fetal erythrocytes (7).
Clinically, PTPS deficiency is a heterogeneous disease with large variations in residual PTPS activity in erythrocytes, in concentrations of neurotransmitter metabolites (especially in cerebrospinal fluid), in the degree of HPA, and in neurological symptoms (typically hypotonia of the trunk and hypertonia of the extremities) (2). The "severe" or "typical" form of PTPS deficiency is characterized by lowered concentrations of neurotransmitter metabolites in cerebrospinal fluid and severe neurological symptoms. This form requires treatment with BH4, neurotransmitters, and their precursors such as L-3,4-dihydroxyphenylalanine, carbidopa, and 5-hydroxytryptophan (8)(9). Less severely affected patients are assigned to a heterogeneous group encompassing mild/peripheral or "atypical" forms of PTPS deficiency. The disease course in this group varies from cases with transient neonatal HPA (10) to cases where a mild form proceeds into a severe form with changes in cerebrospinal fluid neurotransmitter metabolites (11).
The gene encoding the PTPS enzyme (PTS) has been cloned and found to span ~8 kb and consist of six exons (12)(13). The gene maps to chromosome 11q22.3-q23.3 (14) and encodes the 145 amino acids of each subunit of the homohexameric enzyme (13)(15).
A method for fast identification of mutations causing PTPS deficiency would be a substantial improvement for diagnostic and therapeutic purposes, both for early determination of treatment requirements and for prenatal diagnosis. Here, we present an approach based on PCR and denaturing gradient gel electrophoresis (DGGE) to rapidly and simultaneously scan all six exons and flanking intronic sequences of PTS for mutations. We have evaluated the efficacy of this method by analyzing DNA with known PTS mutations, and we have applied the method for the detection of mutations in patients clinically diagnosed as having PTPS deficiency.
| Materials and Methods |
|---|
|
|
|---|
|
dna extraction
Genomic DNA was extracted from 10 mL of EDTA-anticoagulated blood
by standard procedures (17). DNA samples with previously
characterized mutations in the PTS gene were kindly provided
by Drs. Beat Thöny [mutations R16C (46C
T), T67M (200C
T),
P87L (260C
T) and D136V (407A
T)] and Harvey L. Levy [mutation
N47D (139A
G)], and were used for optimizing the method.
primer design
Optimal primers for amplification of the six exons of the
PTS gene with surrounding intron sequences were selected
using the OLIGO 4.0 Program from National Biosciences (18).
Subsequently, computer simulations of the melting behaviors of the six
PCR products were carried out using MELT87 (19) to enable
the addition of appropriate GC clamps to one of the primers in each
primer pair, thus yielding optimal melting behavior for each exon. All
primers are listed in Table 2
.
|
pcr amplification and dgge analysis
PCR amplification was carried out in 25-µL reactions, using a
commercial buffer and Taq polymerase (DNA Taq
polymerase; Qiagen). All reactions contained 1.5 mmol/L
MgCl2, 200 µmol/L dNTPs (Pharmacia Biotech), 20
pmol of each primer, 2 U of Taq polymerase, 200 ng of DNA,
and a loading buffer consisting of 0.08 g/L cresol red (Aldrich) and
120 g/L sucrose. Amplification of exon 1 required the addition of 75
mL/L dimethyl sulfoxide (Sigma). Amplification was carried out in a
Perkin-Elmer Cetus Thermocycler (PE 9600) with the following cycling
parameters: 5 min at 94 °C; 40 rounds of 94 °C (30 s),
5558 °C (30 s), and 72 °C (30 s); and 7 min at 72 °C. To
complete the formation of heteroduplexes, the reaction was
followed by 10 min at 99 °C, a 10-min ramp to 65 °C, 50 min at
65 °C, a 10-min ramp to 37 °C, and 50 min at 37 °C.
Gels for DGGE were made between glass plates of 20 x 18 cm with 1-mm spacers and consisted of 6% polyacrylamide (19:1 acrylamide:bisacrylamide, by weight; Life Technologies) with a linear denaturing gradient ranging from 0% to 90% denaturant (100% denaturant is 7.0 mol/L urea and 400 mL/L formamide, SigmaUltra; Sigma). PCR products (25-µL samples) were loaded onto the gel and subjected to electrophoresis at 160 V for 4.5 h in a PROTEAN II slab electrophoresis cell (Bio-Rad) submerged in 1x Tris-acetate-EDTA buffer kept at 56 °C. The gels were stained with ethidium bromide and photographed under ultraviolet transillumination. PCR products showing aberrant migration patterns were subjected to direct sequencing.
sequencing
PCR products for sequencing were produced in 100-µL reactions
using 2 U of Taq polymerase (Qiagen) with 200 ng of DNA, 20
pmol of each primer, and 50 µmol/L dNTPs. Primers for sequencing were
identical to primers for DGGE but were without GC clamps. Three
microliters of the PCR product was sequenced with the Thermo
SequenaseTM Cycle Sequencing Kit (Amersham Life Science), according to
the manufacturers instructions. Sequencing products were run in
sequencing gels (6% polyacrylamide and 7 mol/L urea), which were
subsequently dried and inspected using PhosphorImaging Technology
(Molecular Dynamics).
cDNA ANALYSIS
RNA was isolated from ~2 x 106
EpsteinBarr virus-transformed lymphocytes using the PurescriptTM RNA
Isolation Kit (Gentra Systems). Reverse transcription was performed
using Superscript IITM (Life Technologies)
in a 20-µL reaction with 150 pmol of random hexamer primers. PCR was
carried out using the ExpandTM High Fidelity PCR System (Boehringer
Mannheim) in a 25-µL reaction containing 1.5 mmol/L
MgCl2, 200 µmol/L dNTPs, 20 pmol of each
primer, 2 U of Taq polymerase, and 2 µL of first-strand
cDNA. The amplification was carried out with the following cycling
conditions: 1 min at 94 °C; 40 rounds of 94 °C (30 s), 58 °C
(30 s) and 72 °C (30 s); followed by a step of 72 °C for 7 min.
Reverse transcription-PCR products were run on a 20% premade
polyacrylamide gel (Novex) in 1x Tris-borate-EDTA buffer at 200 V for
4 h, stained with ethidium bromide, and photographed under
ultraviolet light. Bands of interest were excised, and products were
eluted in 100 µL of distilled H2O overnight and
sequenced directly.
| Results |
|---|
|
|
|---|
primer design
Primers for amplification of individual PTS exons were
positioned in the adjacent intronic sequences, aiming at the following
criteria: (a) all PCR products should have a two-domain
melting profile with the sequence of interest contained within
the low-melting domain; (b) all PCR products
should be of similar size; and (c) all PCR products should
withstand prolonged electrophoresis in a denaturing gradient gel.
Generation of amplification primers gave rise to two problems: one
relating to the very high GC content of exon 1; the other relating to a
15-bp poly(dT) stretch in the splice acceptor site of intron 1.
pts exon 1 is gc rich and high melting
For analysis of PTS exon 1, a standard 40-bp GC clamp
was initially incorporated into the 3' end of the PCR product to
generate the two-domain melting profile. Despite these theoretical
precautions, DGGE analysis of the amplification product failed to
produce band focusing and mutation resolution (data not shown). Closer
examination of the thermodynamic properties of the amplified sequence
by use of the MELT87 algorithm showed that the temperature at which
strand dissociation of the GC-clamped amplification product becomes
significant was below the melting temperature of the lower-melting
domain. The low temperature of strand dissociation implies direct
transition of the amplification product from the fully helical state to
the fully single-stranded state, thereby bypassing the partially melted
state and preventing resolution of mutations by DGGE. To increase the
temperature of strand dissociation above the melting temperature of the
lower-melting domain, the length of the GC clamp was increased to 56 bp
(Fig. 1
). In accordance with the theoretical considerations, DGGE
analysis allowed band focusing and clear resolution of the control
mutation.
|
poly(dT) STRETCH IN INTRON 1
For analysis of PTS exon 2, primers were initially
positioned ~30 bp upstream and downstream of exon 2, respectively.
Despite the optimal melting properties of the PCR product generated
with these primers, as evaluated by computerized simulation, DGGE
analysis revealed at least four distinct bands for all wild-type
control samples (Fig. 2
A), and resolution of the control mutation was virtually
impossible (Fig. 2A
, lane 4). We speculated that this unwanted band
pattern might relate to a 15-bp poly(dT) stretch that is located
immediately upstream of exon 2 as part of the polypyrimidine tract of
the splice acceptor site of intron 1. Previous studies have shown that
microsatellite repeats and mononucleotide runs are susceptible to
insertions and deletions during amplification in vitro and in vivo
because of "slipped strand mispairing" or recombination mechanisms
(23)(24)(25). The generation of frameshift products and
heteroduplexes differing in melting temperature from that of the
wild-type sequence would explain the observed multiplicity of bands in
a denaturing gradient gel.
|
Because modifications of different components of the PCR, including the
thermostable DNA polymerase, have little effect on the fidelity of
microsatellite repeat synthesis (26), we sought to modify
the upstream primer to decrease the production of frameshift products
using standard Taq polymerase. The most successful approach
is delineated in Fig. 2B
. A primer encompassing the entire poly(dT)
stretch was designed with two mismatches that cause the substitution of
the 5th and 10th thymidines of the stretch with guanosine and cytosine,
respectively. The use of this primer together with an unmodified
downstream primer allowed successful amplification under standard PCR
conditions, generating a product that resolved as a single band in the
denaturing gradient gel for all wild-type control samples and as a
characteristic four-band pattern for the mutant control sample (Fig. 2B
, lane 4).
one-step analysis of pts
To establish one common protocol for PCR amplification of all six
exonic regions of PTS, the annealing temperature and the
concentrations of MgCl2, primers, and dimethyl
sulfoxide were varied. At an annealing temperature of 58 °C, all six
regions were amplified with sufficiently high yields to allow
subsequent DGGE analysis, although these conditions were suboptimal for
exon 2 (optimum annealing temperature, 55 °C).
For simultaneous scanning of all six PCR products in a single DGGE run,
denaturing gradient range, buffer temperature, and electrophoresis time
were varied. Optimal resolution of control mutations was obtained in a
090% denaturing gradient gel kept at a constant temperature of
56 °C and run at 160 V for 4.5 h (Fig. 3
). The electrophoresis time needed for simultaneous analysis of
the six PCR products is the time it takes for all products to reach the
positions in the gel where they undergo partial melting and
consequently are retarded. Using these standardized PCR and DGGE
conditions, all control mutations were readily resolved.
|
analysis of patients
Using the method outlined above, we characterized eight mutant
alleles in five individuals, including two siblings, of whom four had
been clinically diagnosed with PTPS deficiency and one was a putative
carrier (Table 1
).
In patient A, the mutations V57del and IVS2-7T
A were identified. A
survey of the literature suggested that this patient had been studied
previously by others (27). V57del is a deletion of the
nucleotides 178180, producing an in-frame deletion of residue 57
(valine). This mutation causes the hexameric structure of the PTPS
enzyme to destabilize and aggregate into higher molecular structures
(28). IVS2-7T
A causes skipping of exon 3 and the
introduction of a premature stop codon (13). The severe
phenotype of this patient corresponds well with the dramatic impact of
these mutations on the PTPS enzyme.
The sibling patients B and C were found to be homozygous for the
IVS1-3C
G mutation. This splice defect previously has been described
to cause a complete inactivation of the enzyme (29), which
is in concordance with the severe phenotype of these patients.
The IVS1-3C
G mutation was also identified on one of the chromosomes
of patient D, who belongs to the mild/atypical group of PTPS-deficient
patients with transient HPA. PCR/DGGE analysis did not reveal mutations
on the other chromosome. The PTPS activity and urinary neopterin and
biopterin concentrations of the patient suggested a defect in the
PTS gene on both chromosomes. Sequence analysis of all
coding regions of the PTS gene confirmed the absence of
additional mutations.
In patient E, one previously undescribed mutation, V97M (289G
A)
(7599G
A; NCBI GenBank accession no. L76259), was detected. No
mutations were identified on the other chromosome, which is in
concordance with the biochemical diagnosis of heterozygosity based on
normal neopterin and biopterin concentrations (Table 1
). Obligate
heterozygosity previously has been described as a cause of HPA
(30). The impact of V97M on the PTPS enzyme function remains
to be determined.
The IVS1-3C
G mutation identified in patients B, C, and D previously
has been reported to cause the deletion of four codons (nucleotides
8491) via a cryptic splice site in exon 2 (29). However,
electrophoresis of reverse transcription-PCR products generated with
primers located in PTS exons 1 and 2, respectively, showed a
multitude of bands in patients homozygous and heterozygous for this
mutation (data not shown). Sequence analysis of individual bands
revealed a large number of different splice variants in addition to the
previously reported 12-bp deletion. For example, one of the aberrantly
spliced transcripts contained an insertion of 4 bp (TGAG) between exons
1 and 2. This corresponds to the highly unusual usage of a dinucleotide
GG, at positions -5 and -6 of IVS1, as acceptor splice site instead
of the wild-type, unaffected AG dinucleotide, underscoring the
importance of a cytosine at position -3 for maintaining the function
of the normal splice site.
| Discussion |
|---|
|
|
|---|
The PCR/DGGE-based method presented in here fulfills several criteria
that make it suitable for diagnostic applications. First, it has the
potential to detect all possible alterations within the regions
analyzed, irrespective of site and type of mutation. The high
specificity was achieved through careful design of amplification
primers, producing GC-clamped PCR products with two-domain melting
profiles. Second, the method is designed to allow simultaneous
examination of all exons and exon-intron boundaries of PTS
by applying standardized PCR and DGGE conditions. Thus, the entire
procedure may be performed in a single day with minimum hands-on time.
All sequence variants detected by DGGE eventually must be identified by
sequence analysis to confirm a likely effect on enzyme function and to
exclude possible silent variants. We have screened >108
wild-type PTS alleles and found only 1 silent variant
(T
C at codon 135) at a frequency of 4%, suggesting that the
PTS gene is not polymorphic and that most sequence variants
detected by DGGE-based mutation scanning are likely to be disease
causing. To date, 31 mutations associated with PTPS deficiency are
registered in the Database of Mutations Causing Tetrahydrobiopterin
Deficiencies (31). Sixteen of these mutations were
reviewed by Thöny and Blau (32), two were described by
Liu et al. (33), two by Scherer-Oppliger et al.
(10), and the remaining have been submitted online.
We have used the present method to examine the PTS gene in
five cases clinically diagnosed as having PTPS deficiency. In all
patients with severe PTPS deficiency, two mutations were identified in
homoallelic or heteroallelic constellations. In the two remaining
cases, only one mutant allele could be characterized. In both cases,
the entire coding sequence and splice sites of PTS were
sequenced to confirm the negative results of the DGGE-based scanning.
Case E had relatively low PTPS activity in erythrocytes but had urinary
neopterin and biopterin concentrations within the reference range,
suggesting that this patient may be a heterozygote. The other case (D)
with only one identifiable mutation, had a more severe, albeit
transient course [described in detail by Güttler et al.
(6)], and both parents had enzyme activities and urine
pterin values within the reference range for heterozygotes (data
not shown). Sequence analysis of the parents revealed the father as the
carrier of the IVS1-3C
G mutation, whereas no mutation was detected
in the mother. The discrepancy between mutation findings and
biochemical phenotypes in the patient and her mother could possibly be
explained by the presence of a mutation outside the regions analyzed in
this study. Mutations located far into the intronic regions may have an
influence on the normal splicing mechanism. In addition, the expression
of an allele could be affected by a mutation in the promoter region.
The latter mechanism has been suggested in a case of severe PTPS
deficiency, where analysis of the cDNA indicated homozygosity for the
mutation T67M, whereas analysis of the genomic DNA revealed
heterozygosity for the same mutation (29). A reduced or
absent expression of the allele not carrying the T67M mutation could
explain this phenomenon. We also attempted analysis of cDNA to
determine whether the apparent heterozygosity in patient D was
associated with lowered expression of the other allele. However, this
analysis was inconclusive because the multitude of splice variants
caused by the IVS1-3C
G mutation made it impossible to determine the
degree of expression of the wild-type allele in this patient.
In conclusion, we have developed a fast and inexpensive method for the detection of small mutations/insertions/deletions in exons and exon-intron boundaries of the PTS gene. This method should be useful for diagnosis and prenatal diagnosis as well as for large-scale screening studies of PTPS deficiencies.
| Acknowledgments |
|---|
| Footnotes |
|---|
| References |
|---|
|
|
|---|
-carbinolamine dehydratase maps to 10q22. Genomics 1994;19:365-368.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |