A mutation in SLC11A3 is associated with autosomal dominant hemochromatosis
Omer T. Njajou1, Norbert Vaessen1, Marijke Joosse1, Bianca Berghuis1, Jeroen W.F. van Dongen1, Martijn H. Breuning2, Pieter J.L.M. Snijders1, Wim P.F. Rutten3, Lodewijk A. Sandkuijl1, Ben A. Oostra1, Cornelia M. van Duijn1
& Peter Heutink1
1 Genetic-Epidemiology Unit: Department of Epidemiology & Biostatistics and Department of Clinical Genetics, Erasmus University Rotterdam, The Netherlands
2 Department of Anthropogenetics Leiden University Medical Center, The Netherlands
3 Stichting Huisarts Laboratorium Breda, The Netherlands.
Hereditary hemochromatosis (HH) is a very common disorder characterized by iron overload and multi-organ damage. Several genes involved in iron metabolism have been implicated in the pathology of HH (refs. 1,
2,
3,
4). We report that a mutation in the gene encoding Solute Carrier family 11, member A3 (SLC11A3), also known as ferroportin, is associated with autosomal dominant hemochromatosis.
Families in which HH segregates as a dominant or pseudo-dominant trait have been reported5,
6,
7. We have carried out a genome-wide scan for linkage in a large Dutch family (n=96) in which hemochromatosis segregates as a dominant trait5 (Fig. 1a) that does not map to any known locus. Individuals were considered to be affected if they had serum ferritin levels exceeding 450 ng/ml and/or transferrin saturation exceeding 50%. The final diagnosis of hemochromatosis was made by a specialist, using results obtained by liver biopsy and/or magnetic resonance imaging (MRI). Subjects with high serum ferritin and/or transferrin saturation but without evidence of iron accumulation in liver biopsy or MRI (Web Table A) were considered to be possibly affected (indicated by a question mark (Fig. 1a). Clinical symptoms of affected individuals are similar to those of other HH patients and include joint pains, osteoarthitis, fatigue, cardimyopathies and endocrine disorders.
Figure 1.a, Pedigree of Dutch autosomal dominant hemochromatosis family with haplotypes for chromosome 2q markers (ordered from centromere to telomere).
Black symbols represent affected individuals. White symbols represent unaffected individuals. Question mark indicates individuals who may be affected. + indicates confirmed carriers of N114H mutation. Bars next to haplotypes indicate a shared risk haplotype. Only relevant family members are shown and the pedigree has been scrambled for privacy reasons. b, Mutation analysis. Control (left) and patient (right) sequences are shown, depicting the heterozygous mutation (M). Amino acid sequences are indicated above the corresponding codons.
We used a conservative model of linkage analysis (Web Note A) in which individuals who were possibly affected were regarded as affection status unknown. Two-point linkage analysis that assumes HH to be a rare autosomal dominant disorder with sex- and age-dependent penetrance yielded positive lod scores for several markers on chromosome 2 (Web Table B). We obtained the highest lod score with marker D2S389 (Zmax=3.01 at =0.0), tested additional markers from the region and constructed haplotypes (Fig. 1a). All clinically diagnosed patients share a common allele for markers D2S389 and D2S167. Recombination events were identified using D2S2273 on the centromeric end and D2S117 on the telomeric end; these defined a critical region of 9 cM on the sex average genetic map. This is the fourth locus identified for HH and so we propose that it be called HFE4.
We carried out a search for candidate genes in public databases and the literature. The best candidate was the recently described SLC11A3 (its aliases are IREG1, MTP1, and FPN1), which has been mapped to chromosome 2q (refs. 8,
9,
10,
11), encompasses 20 kb, comprises 8 exons, and has an iron-responsive element (IRE) in its 5' UTR (Web Fig. A). This IRE binds to iron-responsive proteins 1 and 2, indicating that expression of SCL11A3 is probably regulated by intracellular iron levels. SCL11A3 encodes a basolateral membrane protein of 571 amino acids with 9 or 10 transmembrane domains and is implicated in the movement of iron across the enterocytes to the circulation9,
10,
11,
12,
13.
Mutation analysis of all exons, intron−exon boundaries, and the 5' and 3' UTR (including the IRE) revealed a heterozygous AC transversion at position 734 (A734C) in exon 5 (Fig. 1b) in all clinically defined patients but not in 200 healthy controls (unaffected family members and other Dutch people), indicating that this base change is associated with the disease. The age of the two unaffected individuals who carry the disease allele is consistent with the possibility that they might yet develop HH (age of onset in males is up to 60 years and about 10 years later in females).
The mutation is predicted to effect an amino acid substitution of Asp by His (N144H). Linkage disequilibrium of this mutation with an as-yet unidentified causal mutation (in SLCHA3 or another gene) is possible but not likely for a number of reasons. First, because of its role in iron transport, SLC11A3 is a good candidate gene. Second, we sequenced the entire gene and none of the other sequence alterations were found to co-segregate with the disease. Moreover, the substituted asparagine is a highly conserved amino acid in vertebrates (Fig. 2); it is therefore probable that it is critical to SLC11A3 function. A search for homology of the regions directly adjacent to the mutant residue revealed several other divalent metal transporters that otherwise show little homology to SLC11A3 (Fig. 2), indicating that the mutation resides within a region of the protein that affects iron binding and/or transport. The mutation is predicted to reside in a region encoding a transmembrane domain by all protein structure programs used (Web Note A). The replacement of Asp (a neutral amino acid) with His (a polar residue) is predicted to reduce the hydrophobicity of the transmembrane domain, and may affect the structure (folding) of the protein.
Figure 2. Sequence alignment of region around the mutant residue in SLC11A3 (human) and other proteins.
The N144H amino acid substitution is indicated by an arrow. Predicted transmembrane domains are indicated by red boxes. MNTH, stimulated, highly selective manganese (II) transporter (a natural resistance-associated macrophage protein (NRAMP).
It is possible that mutant ferroportin in HFE4 individuals results in the transport of excess amounts of iron out of enterocytes and into the circulation—in which case, the AspHis mutation would be predicted to activate the protein would be an activating mutation. Iron levels in enterocytes would be expected to be depleted. Activity of the divalent metal transporter 1 (DMT1; encoded by Slc11a2) is controlled by intracellular iron levels, and low iron levels result in increased iron uptake by DMT1. The excessive uptake of iron from the lumen of the gut and transport of iron into the circulation would lead to an accumulation of iron in blood plasma and body tissues. Mutation of Slc11a2 effects anemia in mice14,
15. If mutation in SCL11A3 upregulates DMT1 activity, SCL11A3 may represent a target for the therapy of anemia in addition to that of hemochromotosis.
Acknowledgments We thank the patients and relatives from the Genetic Research in Isolated Populations, the local healthcare centers and the municipality for making this study possible, J. Janssens and local physicians T.W.C. Snieders, G. Droge, P. van Wouw, M. Kraanen, C. Van Broeckoven, W.H.A. Bettink, C.J.M. Beukers and L.J. Dronkers for their help during the study, and E. Wauters and J. Houwing-Duistermaat for technical assistance.
=0.0), tested additional markers from the region and constructed haplotypes (
C transversion at position 734 (A734C) in exon 5 (
