| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Molecular Diagnostics and Genetics |
1 Laboratory of Haematology, 2
Department of Hepatology, 3
Department of Internal Medicine E, 4
Department of Dermatology, 5
Department of Internal Medicine A, CHU of Montpellier, Montpellier, France.
6 Department of Hepato-Gastroenterology, CHU Nîmes, Nîmes, France.
7 Department of Oncology and Haematology, Lille University Hospital, Lille, France.
8 Department of Endocrinology, CHU of Montpellier, Montpellier, France.
aAddress correspondence to this author at: Laboratoire d’Hématologie, CHU de Montpellier, Hôpital Saint Eloi, 34295 Montpellier CEDEX 5, France. Fax 33-467-33-70-36; e-mail p-martinez{at}chu-montpellier.fr.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Methods: We used a standard 96-well microplate with a single PCR condition in an adaptation of the SCAIP (single-condition amplification with internal primer) method to sequence the HFE (hemochromatosis), HAMP (hepcidin antimicrobial peptide), HFE2/HJV [hemochromatosis type 2 (juvenile)], SLC40A1 (ferroportin), and TFR2 (transferrin receptor 2) genes, and the 5' untranslated region of the FTL (ferritin, light polypeptide) gene. To further simplify the method, we adjusted PCR conditions to avoid the use of an internal primer and applied this single-condition amplification method to 38 selected, unrelated patients. We tailored the genetic investigation according to the clinical picture, with the patients falling into 2 groups. Group 1 consisted of patients with hyperferritinemia and high transferrin saturation (TS) (classic adult and juvenile HH forms, groups 1A and 1B, respectively), and group 2 consisted of patients with hyperferritinemia and low, typical, or slightly increased TS, with or without iron overload (groups 2A and 2B, respectively).
Results: With this strategy we identified single-gene and multigene abnormalities, including 6 previously undescribed abnormalities in HFE (c.794dupA), HFE2 (c.–89–4dupT), and SLC40A1 (c.262A>G, c.533G>A, c.1468G>A, and c.–59_–45del).
Conclusion: This method is a simple approach for investigating hereditary iron overload or HF and allows rapid evaluation of patients.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
First, other HFE genotypes, including the compound heterozygote encoding the Cys282Tyr and His63Asp mutations, have been associated with an HH phenotype, although individuals with such genotypes accumulate less iron than Cys282Tyr homozygotes(1)(2)(3). Other rare private HFE mutations responsible for HH have also been reported. Second, Cys282Tyr homozygotes have been demonstrated to lack full penetrance(4), and some individuals homozygous for the Cys282Tyr mutant remain asymptomatic throughout life. Investigators have also described several other genes that present a phenotype similar to that of HFE-related HH, indicating that hemochromatosis is a multigene disorder. Nowadays, HH is considered an autosomal-recessive inherited disorder with 2 clinical forms, adult and juvenile HH(5)(6), defined according to the age of onset. Juvenile hemochromatosis (JH) is due to a mutation in 1 of 2 genes: HFE2 [hemochromatosis type 2 (juvenile); also known as HJV and encoding hemojuvelin](7) and, more rarely, HAMP (hepcidin antimicrobial peptide)(8). Adult HH is mainly due to mutation in HFE, and TFR2 (transferrin receptor 2) mutations are rarely found(9). Associations between mutations of different genes leading to modified clinical presentations have been described for some patients. Such findings demonstrate that hemochromatosis can be a multigenic or at least a digenic disease. Indeed, HAMP(10), HFE2(11), or TFR2(12) mutations that have been associated with HFE genotypes have been described to aggravate the clinical presentation.
On the other hand, clinical pictures characterized by high ferritinemia have recently emerged [for review, see(5)]. Hereditary hyperferritinemia (HF) mostly has a dominant inheritance pattern and may or may not feature iron overload. No iron overload is associated with hereditary hyperferritinemia cataract syndrome (HHCS)(13)(14), which is due to mutations in the 5' untranslated region (UTR) of FTL (ferritin, light polypeptide). On the other hand, alterations in SLC40A1 (ferroportin) cause iron overload(15)(16); in its classic form, however, this disorder, termed ferroportin disease, has clinical and biological presentations that are considerably different from HH(17). A 2nd clinical class of ferroportin-related disorders that produces a phenotype very similar to classic HH has been described for a group of different mutations(18).
These new findings increase the diagnostic and prognostic interest in the precise identification of the molecular disorder in patients with HH not related to a classic HFE genotype or with suspected HF, a technically challenging task given that the total number of genes to test has now increased and corresponds to a total of approximately 90 amplicons (including both the sense and antisense strands). To address this obstacle, we set up in our laboratory a sequencing procedure based on the SCAIP method (single-condition amplification with internal primer) described by Flanigan et al.(19) in 2003. We used a simplified adaptation of this technique to facilitate the straightforward sequencing of most of the genes implicated in human diseases related to iron metabolism.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
phenotypic assessment
Patients enrolled in this study had a phenotypic diagnosis of HH or HF according to previously proposed classification criteria(5)(6). We screened for the usual HFE genotype (encoding the Cys282Tyr and His63Asp mutations) before sequencing the other genes (except in cases of HHCS). To target the genes of interest, we divided patients into 2 groups according to their clinical profile(5). Group 1 included patients with a classic HH phenotype, i.e., displaying at least a high transferrin saturation (TS) associated with hyperferritinemia and increased iron in the liver, as assessed by MRI or liver biopsy. Acquired causes of iron overload were excluded for each patient. This group was divided into 2 subgroups. Group 1A consisted of patients with adult HH forms without a classic HFE genotype (i.e., neither Cys282Tyr homozygotes nor Cys282Tyr/His63Asp compound heterozygotes). Group 1B included patients with either a JH phenotype or patients with an HFE Cys282Tyr/Cys282Tyr or Cys282Tyr/His63Asp genotype and early (age <25 years) or severe clinical expression of hemochromatosis. Group 2 consisted of individuals with hyperferritinemia and low, typical, or slightly increased TS for whom other potential causes of hyperferritinemia, such as inflammation, chronic anemia, cancers, or liver pathologies, could be excluded. Group 2 patients were divided into group 2A (hyperferritinemia with iron overload evidenced by liver biopsy or MRI), which essentially targeted the SLC40A1 gene, and group 2B (hyperferritinemia without iron overload and possibly with cataracts), which targeted suspected HHCS cases.
molecular studies
DNA-sequencing strategy.
We studied 3 sets of genes selected according the patient’s initial phenotypic presentation. We first targeted HFE and TFR2 genes in group 1A and preferentially screened HAMP and HFE2 genes in group 1B. In group 2, we first tested patients with unexplained hyperferritinemia without iron overload (group 2B) for mutations in the 5' UTR of the FTL gene. When iron overload was present (group 2A), we sequenced the SLC40A1 gene. In the studied patients, we had no suspicion of ceruloplasmin deficiency or other clinical forms of hereditary iron overload with neurologic symptoms; thus, genes such as those encoding ceruloplasmin or frataxin were excluded from our experiments. Patients with an ambiguous clinical presentation were screened for all 3 sets of genes.
electronic database information
GenBank sequence accession numbers for the SLC40A1, FTL, HFE, HFE2, HAMP, and TFR2 genes are NC_000002.10, NC_000019.8, NC_000006.10, NC_000001.9,NC_000019.8, and NC_000007.12, respectively. Mutations were named in accordance with the standard international nomenclature guidelines recommended by the Human Genome Variation Society (HGVS at http://www.hgvs.org/mutnomen/recs.html).
single-condition amplification method
We extracted DNA from patient’s peripheral blood leukocytes with standard procedures. Direct sequence analysis of all the targeted iron metabolism-related genes was based on the SCAIP method(19), which allows simultaneous amplification, purification, and sequence analysis of multiple DNA regions on a standardized microplate with a single PCR condition. We used a simplified version of this method, the single-condition amplification (SCA) method, by adjusting PCR conditions according to Roux et al.(20), thereby avoiding the need for an internal primer. We chose amplicons to cover the 39 exons, exon/intron boundaries, and most of the 5' and 3' UTRs of HFE, TFR2, HFE2, HAMP, and SLC40A1, as well as the 5' UTR of FTL. This set constituted 44 amplicons spanning 16.4 kb over a total of 64.8 kb of genomic DNA sequence. Primer sets were adapted from those described in previously published articles or were designed for this study (see Table 1 in the Data Supplement that accompanies the online version of this article at http://www.clinchem.org/content/vol53/issue12). Each amplicon was designed for an optimal size range of 200–500 bp to maintain uniform PCR conditions and tested for compatibility under temperature "stepdown" conditions(21), an adaptation of the "touchdown" PCR method(22). The appropriate amplification gradient used for each gene is given in Table 2 in the online Data Supplement.
dna sequencing
Sequencing reactions were carried out in a 10-µL volume containing 10–50 fmol PCR template, 1.5 µL 5x BigDye® Terminator v1.1 Sequencing Buffer (Applera), 5 pmol/µL sequencing primer, 1.5 µL ABI PRISM® BigDye Terminator v1.1 reaction mixture (Applera), and 4 µL sterile water. Sequencing products were then purified on MontageTM-SEQ96 microplates (Millipore) according to the manufacturer’s instructions. We added 15 µL of Hi-DiTM formamide (Applera) to each well as recommended(22) and electrophoresed the samples on an ABI PRISM 310 DNA analyzer (Applera). We then analyzed sequence-trace files with Sequencing Analysis Software (version 5.1.1; Applera) and used SeqScape® software (version 2.1.1; Applera) to compare the data with the GenBank reference sequence.
control of dna sequencing
DNA from an anonymous control individual with typical erythrocytes and nonpathologic values for iron variables was sequenced with the SCA method in parallel with a classic sequencing method for all of the studied genes(23). For internal positive controls, we used samples that our group or others had previously identified to possess mutations.
analyses of individual mutations and of parents or siblings
We confirmed an identified mutation using a novel PCR product either by restriction enzyme analysis when available or by another sequencing experiment. We carried out a segregation analysis of family members of patients with an identified mutation.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
application to the identification of novel mutations in genes related to iron metabolism
We studied 38 patients with this method, including 16 patients with a suspected diagnosis of HH and 22 patients with suspected HF. The patients with potentially causative mutations are described in Table 1
. In addition, we identified several previously known or undescribed polymorphisms during the course of the study (see Table 3 in the online Data Supplement) and found no mutation in the targeted genes for 29 of the patients.
|
|
In group 1, 1 patient had a new HFE mutation, 1 patient had a novel HFE2 mutation, and 2 patients had a new SLC40A1 mutation. The new HFE mutant was a frameshift mutation, c.794dupA, in HFE exon 4 of patient HG3572 and was in trans of the Cys282Tyr mutant (see Fig. 1 in the online Data Supplement). This patient was classified into group 1A (classic HH). The HFE2 mutation was found in patient HG2280, a 23-year-old woman referred for discrepancies between her HFE genotype and the clinical phenotype. Although she was homozygous for the Cys282Tyr mutation and had a classic adult form of HH, as did her mother, the patient had abnormally high values of iron-related variables for her age (group 1B). A heterozygous T insertion 5' to exon 2 (c.–89–4dupT) was identified in the HFE2 gene inherited from the father, who was a Cys282Tyr/His63Asp compound heterozygote and affected with an adult form of HH. This insertion was not present in the mother. The HFE, HAMP, and TFR2 genes were wild type. A final assignment of causative status for this mutation will require analysis of HFE2 transcripts. Patient HG3943, age 73 years, was referred for suspected classic HH because of a high ferritin concentration (1500 µg/L) and high TS (80%) (group 1A). The HFE genotype was simply a His63Asp heterozygote. An MRI examination of the liver revealed iron overload with a hepatic iron concentration of 240 µmol/g. The patient also had liver enzyme anomalies. The genes involved in HH, namely HFE, TFR2, HFE2, and HAMP, were wild type. An undescribed short deletion, c.–59_–45del, was identified in the 5' UTR of the SLC40A1 gene. The 2nd patient, HG3135, is described below.
Group 2 included 2 unrelated 58-year-old female patients who were referred for isolated hyperferritinemia without iron overload and with cataracts (Table 1
, group 2B). Both patients had a previously described mutation in the 5' UTR of the FTL gene. Three unrelated patients had novel heterozygous mutations in the SLC40A1 gene (Table 1
, group 2A).
Patient HG3181, a 38-year-old man, was referred for high hyperferritinemia with slightly increased TS and a family history of iron overload. He had a wild-type HFE gene, and the genes involved in HH were also wild type. Analysis of the SLC40A1 gene revealed a missense mutation (c.262A>G) in exon 3 leading to an Arg88Gly substitution. This modification was reported simultaneously by us and another group (Aguilar-Martinez et al., unpublished data; Jouanolle et al., personal communication, January 30, 2004). It is noteworthy that the patient had become anemic under a phlebotomy program, which therefore had to be stopped. A liver biopsy performed at another center, which was initially unavailable to us, recently confirmed the iron overload in mixed parenchymal and Kupffer cells typically seen in "ferroportin disease". The patient’s father, who had the same mutation, recently died of liver cancer.
Two apparently unrelated patients (HG3142 and HG3134) who were referred independently shared the same previously undescribed c.1468G>A nucleotide substitution in SLC40A1. This mutation produces a Gly490Ser substitution at the same position of another substitution, Gly490Asp, which has previously been described to lead to ferroportin disease(24). Patient HG3142, who at 24 years at the time of diagnosis was the youngest of both, had isolated hyperferritinemia with typical TS and no additional biochemical abnormality, including hepatic enzymes (group 2A). The 2nd patient (HG3134) was a 71-year-old woman who had high hyperferritinemia (2710 µg/L) associated with high TS (83%). She experienced chronic fatigue and a cataract and was heterozygous for the His63Asp substitution. An MRI evaluation demonstrated a heavy hepatic iron overload (350 µmol/g). The FTL gene was wild type. This patient’s phenotype placed her in group 1A. The apparent phenotypic discrepancy between these 2 patients with the same mutation is discussed below.
Finally, patient HG4421 had a personal and family history of hyperferritinemia. She had a typical TS and no common HFE mutation and was classified in group 2A. A new mutation in the SLC40A1 gene (c.533G>A) was in heterozygous condition (Table 1
).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Although the technical steps have been greatly simplified, reading this large number of sequences still remains a time-consuming step, even with the help of automated computer programs. The SCA is an interesting method for mutation analysis, but, like other sequencing-based technologies, its usefulness is limited for detecting gross rearrangements (deletions or duplications). Although this kind of lesion does not seem frequent in genes related to iron metabolism, specific testing of these defects could be performed in combination with the SCA method, as, for example, with multiplex amplifiable probe hybridization analysis of duplications(28), multiplex ligation-dependent probe amplification(29), or quantitative multiplex PCR of short fragments(30). We have planned to set up such methods to screen gross rearrangements of iron-related genes, especially in patients for whom SCA sequencing has identified no causative mutation. Indeed, by applying the SCA method to genes involved in iron metabolism, we were able to identify 9 potentially causative mutations among 38 patients with suspected HH or HF. The absence of detectable mutations in the remaining patients has not yet been explained. One possibility is that these patients have an unrecognized acquired condition. The correct phenotypic assessment of patients with increased ferritin concentrations or even with increased ferritin concentration and TS is often not complete before the patient is referred to the genetics laboratory. The establishment of rules could avoid unnecessary sequencing.
Unlike the use of such screening strategies as single-strand conformation polymorphism analysis, denaturing gradient gel electrophoresis, or denaturing HPLC, the SCA method permits immediate identification of a sequence variant. Known polymorphisms can be readily recognized so that rare mutations can be considered, depending on the position in the gene sequence and the mutation’s possible role in causing the phenotype. We used an extensive literature review and the criteria proposed by Cotton and Scriver(31) to ascertain the deleterious role of newly identified mutants (Table 2
). Apart from 2 patients with previously identified FTL mutations, the other patients had new undescribed variations in iron metabolism-related genes.
|
Our attempt to make phenotype-to-genotype relationships among the identified defects stimulated interesting lines of discussion. We can outline 3 primary situations. First, digenism for mutations in other iron-related genes has been described for patients heterozygous for HFE Cys282Tyr mutation and an iron overload compatible with a diagnosis of adult HH(9)(10)(11)(32); however, the HFE gene must initially be screened because additional private mutations can be found in these patients(32)(33)(34). This is illustrated by patient HG3572, who had a 2nd frameshift mutation in the other HFE allele (Tables 1
and 2
). Second, digenism involving HFE2, HAMP, or TFR2 mutations must be investigated in patients homozygous for the HFE Cys282Tyr substitution who have an abnormally severe or early clinical presentation, as was previously described. HFE2 seems to be the more frequently affected gene(9), and although additional studies are needed to ascertain the causative role of this mutation, this particular situation seems to be the case for patient HG2280 (Tables 1
and 2
).
Finally, some patients in our series with ferroportin mutations initially had diagnoses of classic HH (group 1A patients HG3134 and HG3943; Table 1
). The absence of mutations in HH genes led to the correct diagnosis. This finding is in keeping with the existence of 2 kinds of clinical profiles associated with different types of ferroportin mutations [for review, see(35)]. The 1st class of mutations leads to the originally described ferroportin disease, and the 2nd class displays a phenotype indistinguishable from HH. These 2 classes can be difficult to separate by only phenotype, however, especially in older patients (e.g., patients HG3134 and HG3943, ages 71 and 73 years, respectively), and functional assays are needed to ascertain the effect of the mutation.
The application of an easy-to-use, flexible sequencing method for the diagnosis of hereditary iron overload and HF has become a subject of great interest. The precise diagnosis of these disorders is clinically useful, especially from a therapeutic point of view. Phlebotomies, the main treatment for iron overload in nonanemic patients, are a well-tolerated and effective way to treat HH, but they can lead to anemia in class 1 ferroportin disease(17) and are contraindicated in HHCS. The availability of the causative mutation is also useful for genetic counseling of family members. This information offers the possibility of early diagnosis in relatives and may help prevent severe complications, as in JH(36).
|
Acknowledgments |
|---|
Financial disclosures: None declared.
Acknowledgments: We thank Hélène Igual, Sabrina Julien, and Anouk Bernard for technical assistance and Dr. Yves Lecorre and Hubert Andreani for patient referral. We are also grateful to Drs./Profs. Marc Ferriere, Christian Jorgensen, Amadou Konate, Pascal Perney, Marie Christine Picot, Jeanne Ramos, and Jean-François Rossi from the AOI 2004 Working Group, CHU Montpellier, for their constant support.
|
Footnotes |
|---|
2 Human genes: HFE, hemochromatosis; HFE2/HJV, hemochromatosis type 2 (juvenile); HAMP, hepcidin antimicrobial peptide; TFR2, transferrin receptor 2; FTL, ferritin, light polypeptide; SLC40A1, ferroportin.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  P. Aguilar-Martinez, S. Cunat, F. Becker, F. Blanc, M. Nourrit, P. Pouderoux, and J.-F. Schved Iron Overload in C282Y Heterozygotes: Identification of New Rare HFE Gene Mutants and a Step Strategy for Diagnosis. Blood (ASH Annual Meeting Abstracts), November 16, 2008; 112(11): 1859 - 1859. [Abstract]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
