|
|
||||||||
Articles |
Department of Clinical Biochemistry, Western General Hospital, Edinburgh EH4 2XU, UK.
a Author for correspondence. Fax 44-131-537-1023.
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
The mutations all occur in arginine codons and result in an apo B-100
molecule that exhibits defective binding to the LDL receptor, leading
to impaired uptake of LDL into the cell and consequently,
hypercholesterolemia. The first to be described, and the most
characterized, is caused by a G
A transition at nucleotide 10 708
and results in the substitution of arginine by glutamine at codon 3500
(apo B R3500Q) (5). The other two, both recent
discoveries, are each caused by a C
T transition, one at nucleotide
10 800 and the other at nucleotide 10 707. These result,
respectively, in the substitution of arginine by cysteine at codon 3531
(apo B R3531C) (6) and arginine by tryptophan at codon
3500 (apo B R3500W) (7).
The estimated heterozygote frequency of FDB in the UK, based on the R3500Q mutation, is believed to be ~1 in 500 (8). However, the prevalence could possibly be greater with the discovery of these two other mutations.
Traditionally, studies of FDB have centered on the R3500Q mutation and use of detection methods specific for this mutation (9)(10)(11). Because at least two other mutations are known to cause FDB, it is more appropriate to use a method that would detect all such mutations in a single analysis. Three procedures that may be suitable for this are the molecular-scanning techniques of heteroduplex analysis (12)(13), single-strand conformation polymorphism (SSCP) analysis (14)(15), and denaturing gradient gel electrophoresis (DGGE) (16). In this study, we have compared the ability of these three techniques to detect the mutations in and around codon 3500 of the apo B gene that cause FDB in a cohort of 432 hypercholesterolemic individuals. The aim was to ascertain which, if any, was the most appropriate method to detect FDB in the diagnostic setting.
| Materials and Methods |
|---|
|
|
|---|
Samples.
Whole-blood samples for DNA analysis were
anticoagulated with potassium-EDTA (1.2 mg/mL).
DNA extraction.
This was performed by the method of Miller et
al. (18)
Heteroduplex analysis.
Heteroduplex analysis was performed
according to the method of Kotze et al. (19). Briefly, a
345-bp fragment that spanned codon 3500 of the apo B gene was first
amplified by the polymerase chain reaction (PCR), with the
oligonucleotide primers and reaction conditions of Tybjærg-Hansen et
al. (8). PCR product (5 µL) was then mixed with an equal
volume of gel-loading buffer (950 mL/L formamide, 0.5 g/L bromphenol
blue, 0.5 g/L xylene cyanol), and the complete mixture was loaded on
the gel. To maximize throughput, up to 5 sample loadings were applied
to each track of the gel, and each application was separated by a
15-min, 800-V electrophoresis step. Electrophoresis was then performed
at 300 V for 17 h overnight, on a low-cross-linking 40 x
21 x 0.1 cm polyacrylamide gel (10% polyacrylamide, 0.1%
N,N'-methylenebisacrylamide, containing 150 g/L
urea in 0.6x TBE (1x TBE = 89 mmol/L Tris, 89 mmol/L
H3BO4, 2 mmol/L EDTA). The homo- and
heteroduplex DNA bands were then stained with ethidium bromide and
viewed over ultraviolet light.
SSCP analysis.
A method for SSCP analysis was developed that
gave the clearest bandshift when used to analyze a sample known to be
heterozygous for apo B R3500Q. This required the systematic
optimization of all of the variables known to influence the efficiency
of SSCP detection, namely, PCR-product size, percentage acrylamide,
percentage cross-linker, temperature of electrophoresis, and
presence/absence of glycerol. These studies lead to the choice of the
following procedure and conditions.
DNA was first amplified by the PCR with the same primers and conditions
as described above (8) and the addition of 1 µCi of
[
-32P]dCTP (3000 Ci/mmol; Amersham International) to
the reaction mixture before PCR. The radiolabeled PCR product was then
digested with 10 U of PstI in a final volume of 10.5 µL at
37 °C for 1 h. One microliter of 10 g/L SDS was then added
followed by 10 µL of gel-loading buffer (950 mL/L formamide, 0.5 g/L
bromphenol blue, 0.5 g/L xylene cyanol). Samples were denatured at
99 °C for 2 min and quenched in an ice-water bath, and then 23
µL was applied to the gel. The use of a sharks-tooth comb allowed up
to 48 samples to be run per gel. Electrophoresis was performed at room
temperature on a 40 x 21 x 0.04 cm polyacrylamide gel
(7.5% polyacrylamide, 0.15%
N,N-methylenebisacrylamide in 50 mL/L glycerol,
and 1x TBE) for 17 h overnight at 450 V. To maintain as constant
a temperature as possible during the electrophoresis step, a cooling
fan was directed at the gel for the duration of electrophoresis. After
fixing for 20 min in methanol/acetic acid (100 mL/L methanol, 100 mL/L
acetic acid) solution, the gel was dried, and the bands were detected
after autoradiography for 2448 h at -70 °C.
DGGE.
This was performed with the D Gene DGGE system (Bio-Rad
Laboratories). A region spanning codons 34573553 of the apo B gene
was amplified by the PCR with the primers and conditions described by
Nissen et al. (20). The resulting products were then
separated on a 20 x 16 x 0.1 cm gel of 6% polyacrylamide
containing a linearly increasing gradient of 2060% denaturant (100%
denaturant = 7 mol/L urea, 400 mL/L formamide) in 1x TAE buffer
(40 mmol/L Tris acetate, 1 mmol/L EDTA, pH 8.0) (20).
Electrophoresis was performed at 150 V for 5 h in 1x TAE buffer
heated to 60 °C. The gels were stained in ethidium bromide, and the
homo- and heteroduplex bands were viewed over ultraviolet light.
Mutation-specific method for detection of apo B R3500Q.
This
was performed by PCR with mutagenic primers followed by digestion with
MspI according to the method of Motti et al.
(11).
Sequence analysis of the apo B gene.
Direct DNA sequencing was
performed with the Sequenase PCR product sequencing kit (Amersham
International). PCR product was generated with the primer pairs of
either Tybjærg-Hansen et al. (8) or Motti et al.
(11). PCR amplification mixture (7 µL) containing ~0.5
pmol of PCR product was incubated with 1 µL each of exonuclease I
(10.0 U) and shrimp alkaline phosphatase (2.0 U) for 15 min at
37 °C. These enzymes were then inactivated by heating for 15 min at
80 °C, and sequencing was performed in both directions with the
primers that generated the PCR product, according to the
manufacturer's instructions.
| Results |
|---|
|
|
|---|
Heteroduplex bands were observed in all 7 samples known to be
heterozygous for the R3500Q mutation and also in 4 of the 425 samples
of unknown genotype. A typical example of a heteroduplex gel is shown
in Fig. 1
.
|
SSCP analysis.
Under nondenaturing conditions, single-stranded
DNA has a folded structure, or conformation, that arises from
intramolecular interactions that are a consequence of its sequence.
Substitution of one nucleotide with another, within a piece of DNA,
results in a change in this conformation and consequently its
electrophoretic mobility through a polyacrylamide gel, which may be
observed as a bandshift.
A bandshift was observed after SSCP analysis of six of the seven known
R3500Q heterozygotes and one of the remaining four samples of unknown
genotype that also tested positive by heteroduplex analysis. An example
of an SSCP run showing the mobility shift of a positive sample
heterozygous for the R3500Q mutation is shown in Fig. 2
.
|
DGGE.
During electrophoresis, double-stranded DNA amplified by
the PCR migrates through a gel containing a gradient of denaturant
until it melts; i.e., the strands separate. At this point, no further
migration occurs. Because the melting behavior of double-stranded DNA
is a function of its base sequence, the presence of a mutation within a
fragment of DNA will alter the melting behavior of the molecule. This
in turn will alter the position where the fragment is halted when
electrophoresis is performed in a denaturing gradient gel. The usual
pattern obtained on analysis of DNA heterozygous for a point mutation
is four bands, consisting of two faster-moving homoduplex and two
slower-moving heteroduplex molecules.
With DGGE, all seven of the individuals heterozygous for apo B R3500Q
were identified by the presence of two homo- and two heteroduplex
bands. In addition, the same four samples, of unknown genotype, that
tested positive by the heteroduplex method also tested positive by
DGGE. Two of these four gave the identical four-band DGGE pattern as
the seven R3500Q heterozygotes; however, in two of the samples, the
homo- and heteroduplex bands were spaced much farther apart (Fig. 3
).
|
Mutation-specific method for the detection of apo B R3500Q.
With the MspI-based mutagenic primer assay
(11), two of the four samples of unknown genotype, both of
which screened positive by heteroduplex analysis and DGGE, were shown
to be R3500Q heterozygotes (only one of these two samples had been
detected by SSCP analysis).
Sequence analysis of the apo B gene.
The remaining two samples
of unknown genotype that were identified by both the heteroduplex and
DGGE methods were shown by direct DNA sequencing to contain a C
T
transition at nucleotide 10 800 in codon 3531; i.e., they were both
heterozygous for apo B R3531C. A detailed description of the
biochemical features associated with this mutation in these individuals
has been described elsewhere (22). Neither of these
samples was detected by SSCP analysis. An example of an SSCP gel
containing undetected R3500Q and R3531C heterozygotes is shown in Fig. 4
. The findings of the method comparison are summarized in
Table 1
for clarity.
|
|
| Discussion |
|---|
|
|
|---|
Nine samples heterozygous for apo B R3500Q (seven known and two
unknown) were detected by both heteroduplex analysis and DGGE. In
contrast, SSCP analysis identified only seven of these individuals. The
finding that SSCP analysis detected some, but not all, of the R3500Q
heterozygotes is difficult to explain, especially as all of the method
variables, such as temperature of analysis, gel characteristics, and
size of PCR product, were optimized to detect this mutation. Whereas we
did observe some day-to-day variability in the mobility of the single
strands between assays (e.g., compare Fig. 2
with Fig. 4
), these were
small, and the polymorphism in the positive controls was always
detected. Humphries et al. have suggested (23) that the
small regions of base-pairing responsible for the conformations of the
single strands and therefore the potential polymorphisms are likely to
have a melting temperature close to the average room temperature. They
further reported that during hot weather the migration of bands changes
considerably so that some SSCPs are no longer detected
(23). Therefore during hot weather, some of our SSCP gel
assays, which were supposedly performed at room temperature, may have
become too warm, resulting in nondetection of the polymorphism.
However, we think this explanation unlikely because the bandshift in
the R3500Q positive controls was always observed. The reason for this
apparent inconsistency in R3500Q detection by SSCP remains to be
determined.
Unlike heteroduplex analysis and DGGE, SSCP analysis failed to detect the two R3531C heterozygotes present in this cohort, probably because the method was performed under conditions optimized for the detection of apo B R3500Q. In practice, to maximize mutation detection, SSCP analysis is usually performed under at least two different sets of conditions, e.g., with or without glycerol (24), at 4 °C or at room temperature. Therefore had we followed this procedure, the R3531C heterozygotes probably would have been detected. This approach, however, is much more labor-intensive and therefore unsuitable for screening many samples for diagnostic purposes, where in practice a one-pass technique is preferable.
SSCP analysis was the most laborious of the three techniques to perform in practice. Its main disadvantages were the small size (<200 bp) of the PCR product required for maximum efficiency of mutation detection (25), which necessitated digestion with the restriction endonuclease PstI before analysis, the use of radioactivity, the requirement of gels to be dried before autoradiography, the time taken for autoradiography, usually 2448 h, and the detection of only 64% of the mutations identified by the other two techniques.
In contrast, DGGE was not only highly efficient and simple to perform, but it also appeared able to differentiate between the mutation types. The four-band pattern observed in the nine R3500Q heterozygotes was quite distinct from that observed in the two R3531C carriers. In the former, the bands were closely spaced whereas in the latter, they were further apart. Moreover, the patterns were identical in all individuals with the same genotype, a finding in agreement with Nissen et al. (20). Thus, by including a control R3500Q sample in each run of new samples of unknown genotype, any R3500Q heterozygotes should readily be identified by inspection of their four-band pattern. Samples that give rise to variant patterns may be subjected to further analysis to identify the nature of the nucleotide substitution. The two main disadvantages of DGGE were the requirement for specialized equipment and the finding that this method had the lowest throughput, 32 samples per electrophoretic run.
The heteroduplex method used in this study was the simplest of the three techniques to perform and was as efficient as DGGE in detecting mutations in this region of the apo B gene. Furthermore, by multiple sample loadings, up to 100 samples could be analyzed in each run, giving it the biggest throughput of all three techniques. It does have two minor drawbacks, however. Firstly, unlike DGGE, it cannot distinguish DNA homozygous for a mutation from wild-type DNA. Because only a handful of individuals homozygous for FDB have been described worldwide, this is unlikely to be a problem in practice (26)(27). Secondly, it is unable to distinguish between different mutations, so any samples that give rise to heteroduplexes require further analysis by another method. This should not prove too much of a problem given the relatively low incidence of FDB, even in hypercholesterolemic populations (17).
Because we did not perform sequence analysis on every sample studied, we cannot exclude the presence of other undetected mutations in this population. However, because DGGE is believed to be almost 100% effective in mutation detection (28), we believe it unlikely that there are any. In conclusion, we found both DGGE and heteroduplex analysis to be equally effective in detecting mutations in and around codon 3500 of the apo B gene. Furthermore, both were more effective and less laborious to perform than SSCP analysis. However, in view of its greater ease, throughput, and absence of a requirement for specialized equipment, we believe that heteroduplex analysis is the method of choice for detection of mutations in this region of the apo B gene that give rise to FDB.
| Acknowledgments |
|---|
| Footnotes |
|---|
| References |
|---|
|
|
|---|
The following articles in journals at HighWire Press have cited this article:
![]() |
N. Jacobsen, M. Fenger, J. Bentzen, S. L. Rasmussen, M. H. Jakobsen, J. Fenstholt, and J. Skouv Genotyping of the Apolipoprotein B R3500Q Mutation Using Immobilized Locked Nucleic Acid Capture Probes Clin. Chem., April 1, 2002; 48(4): 657 - 660. [Full Text] [PDF] |
||||
![]() |
E. Fisher, H. Scharnagl, M. M. Hoffmann, K. Kusterer, D. Wittmann, H. Wieland, W. Gross, and W. Marz Mutations in the Apolipoprotein (apo) B-100 Receptor-binding region: Detection of apo B-100 (Arg3500->Trp) Associated with Two New Haplotypes and Evidence That apo B-100 (Glu3405->Gln) Diminishes Receptor-mediated Uptake of LDL Clin. Chem., July 1, 1999; 45(7): 1026 - 1038. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |