| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1
Institute of Clinical Chemistry, University Hospital Zurich, Rämistrasse 100, CH-8091 Zürich, Switzerland;
2
Laboratoire de Génétique Moléculaire, INSERM CRI 96 07, CHU-UBO, 46 Rue Félix Le Dantec, Brest, France;
3
TIB MOLBIOL, Tempelhofer Weg 11-12, D-10829 Berlin, Germany;
a author for
correspondence: fax 41-1-255-4590, e-mail fma{at}ikc.unizh.ch
Hereditary hemochromatosis is a common autosomal recessive disorder of iron metabolism, which in its homozygous form occurs in Caucasian populations with a prevalence of 0.2–0.5% (1)(2). Characteristic of the disease is the excessive accumulation of dietary iron and a progressive rise in iron stores. This may lead to serious clinical consequences, including cirrhosis, cardiac failure, diabetes, arthritis, and hepatocellular carcinoma. Treatment involves removal of the iron burden by regular venesection and leads to a normal life expectancy if implemented before the development of cirrhosis (3). Thus, early detection and treatment are critically important. The recent identification of a hemochromatosis gene (HFE, initially termed HLA-H) by Feder et al. (4) allows early genetic diagnosis and greatly simplifies the screening of a family once affected individuals have been identified. The HFE gene protein product is structurally similar to MHC class I-type molecules and interacts with ß2-microglobulin and the transferrin receptor to limit iron absorption (5)(6). Three disease-associated mutations have been detected in the HFE gene. Most individuals with hemochromatosis (80–100%, depending on the population studied) are homozygous for the mutation C282Y. In addition, a small number of compound heterozygotes for C282Y and a second mutation, H63D, may develop clinical iron overload (7). Recently, we demonstrated that a third mutation, S65C, is enriched in hereditary hemochromatosis patients who have a mild form of the disease and who have no mutations at C282 or H63 (8). Although mutations in the HFE gene thus account for most cases of hereditary hemochromatosis, it is clinically important that a minority of hereditary iron overload syndromes are not associated with mutations in the HFE gene. Absence of such mutations thus should not be interpreted to mean that the patient in question does not have hereditary iron overload (1)(2).
Several assays for the identification of the C282Y and H63D point mutations and one assay for the detection of the S65C mutation have been described, but they are labor-intensive and costly because post-PCR processing by restriction enzyme digestion and gel electrophoresis analysis is necessary (9). The LightCycler (Roche Molecular Biochemicals) is a combined microliter volume thermal cycler and fluorometer suitable for fast real-time fluorescence PCR. In addition, it is designed for mutation detection by melting point analysis with fluorescent hybridization probes using the fluorescence energy transfer (FRET) principle (10). A LightCycler procedure to detect the C282Y mutation by the use of by real-time fluorescence PCR has recently been developed (11), obviating the need for re-opening of PCR reaction tubes and post-PCR processing for this mutation.
We set out to construct a LightCycler PCR protocol for the simultaneous
detection of the H63D and S65C mutations of the HFE gene,
based on the principle of mutation detection by melting point analysis
with FRET hybridization probes. Because the mutation causing the S65C
variant [A
T at nucleotide (nt) 193] is located only six
nucleotides away from the mutation underlying H63D (nucleotides A193T
and C187G), we used a "sensor" probe covering both nucleotide
positions/both sites. Specifically, the sensor probe was designed
complementary to the 63D mutant and the S65 wild type, whereas the
anchor probe was constructed to cover an adjacent invariant gene
segment. Thus, the 63D mutant allele is expected to yield the highest
melting temperature (Tm), the H63 wild
type a lower Tm, and the 65C mutant an
even lower Tm. LightCycler PCR
mutation analysis was performed as follows: Blood was obtained in 5-mL
EDTA Vacutainer Tubes. DNA was isolated by a standard rapid lysis
technique using the Qiagen QIAamp Blood kit, and LightCycler PCR
was performed with the following oligonucleotides:
These four oligonucleotides were added to the LightCycler DNA Master Hybridization Probes mixture (Roche Molecular Biochemicals) in the following final concentrations: primer Hä up, 0.5 µmol/L; Hä anchor, 0.2 µmol/L; Hä mut sensor, 0.2 µmol/L; primer Hä down, 0.5 µmol/L. Proband or control DNA was added to a final concentration of 5 pg/20 µL. The total volume in the LightCycler capillary was 20 µL.
PCR was performed with the following temperature-time protocol:
Thereafter, melting curve analysis of amplicons was performed,
starting from 45 °C and proceeding until 85 °C at a linear rate
of 0.2 °C/s. In this assay, the H63D mutant allele is expected from
theoretical calculations to yield a Tm
of 69 °C, whereas the fluorescent probe is expected to dissociate
from an H63 wild-type allele at a significantly lower temperature
because of the single-nucleotide mismatch CG at nt 187 of
HFE cDNA. Furthermore, an allele wild type at AA
position 63 (H63) but mutated at AA position 65 (65C) would contain two
nucleotide mismatches with the fluorescent probe complementary to 63D
and would be expected to melt at an even lower temperature.
In a first series, DNA samples from 12 unselected patients referred to us for HFE gene analysis were tested with this LightCycler protocol. As the reference method, they were also typed for the presence of the H63D mutation by a commercially available kit based on PCR amplification followed by oligonucleotide hybridization (ONH) to microtiter well-immobilized oligonucleotides specific for either the wild type (H63) or the mutation (63D). In this assay, biotin residues are incorporated into amplicons during PCR, which later serve to reveal oligonucleotide-bound amplicons by a streptavidin-enzyme conjugate and colorimetric detection (Vienna Laboratory). The absorbances found after hybridization to both single probes are recorded, and an absorbance ratio between 0.5 and 2.0 is recommended by the manufacturer to identify heterozygous H63D carriers, whereas absorbance ratios >2 are indicative of a homozygous state, either H63H63 or 63D63D. DNA of one patient exhibiting a homozygous H63 wild type (patient P1) and from another patient exhibiting the heterozygous genotype H63,63D (patient P2) as established by the PCR-ONH assay were carried further as controls in all further determinations with both the PCR-ONH and the LightCycler PCR procedure ("WT control" and "HET control"). This was necessary because the internal genotype controls contained in the kit were too short for the primer combination chosen for the LightCycler PCR.
LightCycler PCR yielded a Tm for the
control mutant 63D allele of 69.4 ± 0.26 °C (mean ± SD;
n = 8 for subsequent results) and a 6.2 °C lower
Tm of 63.2 ± 0.35 °C for the
control wild-type H63 allele. The CV for the
Tm of the mutant 63D allele was lower
(0.37%) than that of the wild-type H63 allele (0.56%), in agreement
with the theoretical expectation. The relatively large
Tm difference allowed for a clear
distinction between the H63 and 63D alleles (Fig. 1
). The PCR-ONH assay also clearly distinguished the control
mutant 63D allele and the control wild-type H63 allele, with an
absorbance ratio of 0.87 ± 0.05 for the HET control
DNA and an absorbance ratio >4.2 for the WT control DNA. Notably, the
calculated 95% confidence interval for the absorbance ratio actually
obtained with HET control DNA, 0.77–0.98, was much narrower than the
confidence interval suggested by the manufacturer, 0.5–2.0.
|
Of the other 10 consecutive patients analyzed, unambiguous and
identical genotype assignment by both methods was observed in 9 samples
(Table 1
). However, LightCycler melting curve analysis of the DNA of
patient 11 consistently yielded one peak corresponding to an H63D
allele (Tm of 69.4 °C) and a
distinct second peak with a Tm of
59.3–59.4 °C, or 3.8–3.9 °C lower than the
Tm of the H63 wild-type allele (Fig. 1
) and thus clearly not contained within the
Tm confidence interval for either H63
or 63D. Because the lower Tm of the
H63D probe-H63 product is caused by one mismatch (CG at nt 187 of
HFE cDNA), we suspected that the lower
Tm observed in this patient was
attributable to the presence of a second nucleotide mismatch in the
stretch of the HFE gene spanned by the 63D-targeted
fluorescent hybridization probes, i.e., the presence of a third known
mutation in the HFE gene, A193T, which encodes the S65C
amino acid exchange. Interestingly, the PCR-ONH assay also yielded a
subtly abnormal result in this patient: the observed absorbance ratio
was 0.54 and would lead to a diagnosis of heterozygosity for H63D if
interpreted according to the manufacturer’s instructions. If, however,
the narrower confidence interval that we determined ourselves
(0.77–0.98) was used, the abnormal binding of the patient’s
"wild-type" allele to the wild-type oligonucleotide was obvious and
also pointed to a base exchange near, but distinct from, the CG
mutation at nt 187 of HFE cDNA. Therefore, the LightCycler
PCR product of this patient was sequenced on an ABI Prism model 310,
using either the 3' PCR primer or the 5' PCR primer in a standard
Perkin-Elmer protocol, which identified the A193T mutation in addition
to the CG mutation at nt 187 (H63D) in this patient. The patient was
a 59-year-old man with a medical history of hepatomegaly. Apart from
the S65C mutation, he carried only a heterozygous H63D mutation but no
mutation at C282. In a second series, DNA from another 15 patients we
had typed earlier for the H63D and S65C mutations by PCR plus
HinfI restriction fragment length polymorphism and
sequencing (8) was subjected to LightCycler PCR. Of these 15
samples, 10 carried the S65C mutation either alone or in conjunction
with the H63D mutation. In all 15 samples, the LightCycler PCR
procedure unambiguously yielded the correct genotype assignment (data
not shown). The presence of a polymorphic site in intron 4 of
HFE recently has been shown to prevent amplification of some
C282 wild-type alleles if primers are chosen to cover this site,
causing overestimation of the frequency of C282Y homozygotes
(12). Indeed, in the PCR reaction described here, the down
primer is localized on a polymorphic T/C at nt +4 in intron 2 of
HFE (~70% T and 30% C). However, the presence of a TT,
TC, or CC genotype at this site did not influence the efficiency of the
PCR reaction and did not affect correct H63D and S65C genotype
assignment (data not shown).
|
Thus, we have established a LightCycler PCR assay for the simultaneous detection of the H63D and S65C mutations in the HFE gene. When one fluorescent probe is targeted specifically to the H63D mutation, the H63D and S65C mutations cause easily recognizable Tm curve shifts in opposite directions from the wild-type situation. This principle may be applicable to other situations in which mutations within close proximity are to be detected (13)(14). In contrast to other multiplex formats, the procedure presented here is based solely on differences in Tm shifts and utilizes only one fluorescence readout channel, allowing additional multiplexing by incorporation of additional fluorophores. Specifically, together with an appropriate protocol for LightCycler detection of C282Y (11), the analysis of all currently recognized HFE mutations by LightCycler PCR and fluorescent Tm analysis is thus possible in a single reaction vial provided that C282Y and H63D/S65C are monitored by discernible fluorophores.
The analysis of samples using hybridization probes and melting curve analysis is like a fingerprint for a specific polymorphism, but in addition, it also clearly shows the presence of neighboring polymorphisms, which are not detectable with common techniques such as restriction fragment length polymorphisms. The procedure yields results superior to any solid-phase-based format that uses either microtiter plates or the new chip technology, where difficulties have been reported because of reactions/results between "yes" and "no" (15); it also yields better results when compared with DNA sequencing, which will not always find heterozygosity. In cases where multiple different mutations may occur in the vicinity of a "target" mutation (14), we suggest confirming any mutation with the appropriate mutation-specific probe.
References
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  J. Cheng, Y. Zhang, and Q. Li Real-time PCR genotyping using displacing probes Nucleic Acids Res., April 15, 2004; 32(7): e61 - e61. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Siebers How Accurate Are References in Clinical Chemistry? Clin. Chem., March 1, 2001; 47(3): 606 - 607. [Full Text] [PDF]
|
|
|
|
|
|
E. Schutz, N. von Ahsen, and M. Oellerich Genotyping of Eight Thiopurine Methyltransferase Mutations: Three-Color Multiplexing, ""Two-Color/Shared"" Anchor, and Fluorescence-quenching Hybridization Probe Assays Based on Thermodynamic Nearest-Neighbor Probe Design Clin. Chem., November 1, 2000; 46(11): 1728 - 1737. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. G. Donohoe, M. Laaksonen, K. Pulkki, T. Ronnemaa, and V. Kairisto Rapid Single-Tube Screening of the C282Y Hemochromatosis Mutation by Real-Time Multiplex Allele-specific PCR without Fluorescent Probes Clin. Chem., October 1, 2000; 46(10): 1540 - 1547. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
