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 β2-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: Primer Hä up: 5′-GCTCTGTCTCCAGGTTCACACTC-3′ Hä anchor: 5′-LC Red 640-CTGGTCATCCACGTAGCCCAAAGCTTCAA-phosphorylated Hä mut sensor: 5′-CGGCGACTCTCATCATCATAGAACACGAACA-carboxyfluorescein Primer Hä down: 5′-CCCTCTCCACATACCCTTGC-3′ 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: Denaturation: 95 °C for 1 s in the first cycle; Denaturation: 95 °C for 0 s in further cycles; Annealing: 60 °C for 10 s; Elongation: 72 °C for 10 s, with a transition rate of 20 °C/s between temperature plateaus for a total of 45 cycles. 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 C→G 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. LightCycler-PCR melting curve analysis. Red curve, results of a DNA sample heterozygous for the H63D mutation but wild type at AA position 65. Pink curve, results of a DNA sample heterozygous for the H63D mutation and heterozygous for S65C. Blue curve, result obtained without DNA. 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 (C→G 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 C→G 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 C→G 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). Results of genotyping AA 63 of HFE by PCR-ONH and by LightCycler PCR with FRET hybridization probes targeted at HFE-63D.1 Dotted lines separate PCR analyses performed on different dates. oligo, oligonucleotide; P, patient; ND, not detected. 95% confidence interval for heterozygosity (H63, 63D) by PCR-ONH, 0.77–0.98 (absorbance ratio, mean ± 2 SD); 95% confidence interval for presence of H63 by LightCycler PCR, 62.49–63.91 [Tm, (°C), mean ± 2 SD]; 95% confidence interval for presence of 63D by LightCycler PCR, 68.89–69.93 [Tm (°C), mean ± 2 SD]. Results obtained with patient 11 (P11), later identified to carry S65C in addition to H63D, are in bold. ? indicates that the assignment was questionable because the absorbance ratio was outside the confidence interval for simple heterozygosity H63, 63D. XXX indicates that the LightCycler gave a melting temperature lower than the lower confidence limit for H63, and therefore, the allele could not be the H63 wild type. Results of genotyping AA 63 of HFE by PCR-ONH and by LightCycler PCR with FRET hybridization probes targeted at HFE-63D.1 Dotted lines separate PCR analyses performed on different dates. oligo, oligonucleotide; P, patient; ND, not detected. 95% confidence interval for heterozygosity (H63, 63D) by PCR-ONH, 0.77–0.98 (absorbance ratio, mean ± 2 SD); 95% confidence interval for presence of H63 by LightCycler PCR, 62.49–63.91 [Tm, (°C), mean ± 2 SD]; 95% confidence interval for presence of 63D by LightCycler PCR, 68.89–69.93 [Tm (°C), mean ± 2 SD]. Results obtained with patient 11 (P11), later identified to carry S65C in addition to H63D, are in bold. ? indicates that the assignment was questionable because the absorbance ratio was outside the confidence interval for simple heterozygosity H63, 63D. XXX indicates that the LightCycler gave a melting temperature lower than the lower confidence limit for H63, and therefore, the allele could not be the H63 wild type. 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.
No takes yet. Share an insight, caveat, or question.
Bollhalder et al. (1999) studied this question.
Synapse has enriched 4 closely related papers on similar clinical questions. Consider them for comparative context: