The Spectra of Disease-Causing Mutations in the Ferroportin 1 (<i>SLC40A1</i>) Encoding Gene and Related Iron Overload Phenotypes (Hemochromatosis Type 4 and Ferroportin Disease)

Uguen, Kevin; Ka, Chandran; Collod-Béroud, Gwenaelle; Le Tertre, Marlène; Guellec, Julie; Férec, Claude; Béroud, Christophe; Callebaut, Isabelle; Le Gac, Gérald

doi:https://doi.org/10.1155/2023/5162256

Human Mutation

On this page

Abstract Introduction Conclusion Data Availability Conflicts of Interest Authors’ Contributions Acknowledgments Supplementary Materials References Copyright Related Articles

Review Article | Open Access

Volume 2023 | Article ID 5162256 | https://doi.org/10.1155/2023/5162256

The Spectra of Disease-Causing Mutations in the Ferroportin 1 (SLC40A1) Encoding Gene and Related Iron Overload Phenotypes (Hemochromatosis Type 4 and Ferroportin Disease)

Kevin Uguen,^1,2Chandran Ka,^1,2,3Gwenaelle Collod-Béroud,⁴Marlène Le Tertre,^1,2Julie Guellec,^1,2,5Claude Férec,^1,2,5Christophe Béroud,⁴Isabelle Callebaut,⁶and Gérald Le Gac^1,2,3

Academic Editor: Stephen P. Robertson

Received02 Dec 2022

Revised17 May 2023

Accepted26 May 2023

Published13 Jun 2023

Abstract

SLC40A1 is the sole iron export protein reported in mammals and is a key player in both cellular and systemic iron homeostasis. This unique iron exporter, which belongs to the major facilitator superfamily, is predominantly regulated by the hyposideremic hormone hepcidin. SLC40A1 dysfunction causes ferroportin disease, and autosomal dominant iron overload disorder characterized by cellular iron retention, principally in reticuloendothelial cells, correlating with high serum ferritin and low to normal transferrin saturation. Resistant to hepcidin, SLC40A1 mutations are rather associated with elevated plasma iron and parenchymal iron deposition, a condition that resembles HFE-related hemochromatosis and is associated with more clinical complications. With very few exceptions, only missense variations are reported at the SLC40A1 locus; this situation increasingly limits the establishment of pathogenicity. In this mutation update, we provide a comprehensive review of all the pathogenic or likely pathogenic variants, variants of unknown significance, and benign or likely benign SLC40A1 variants. The classification is essentially determined using functional, structural, segregation, and recurrence data. We furnish new information on genotype-phenotype correlations for loss-of-function, gain-of-function, and other SLC40A1 variants, confirming the existence of wide clinical heterogeneity and the potential for misdiagnosis. All information is recorded in a locus-specific online database.

1. Introduction

The solute carrier family 40 member 1 (SLC40A1), now preferably referred to as ferroportin 1 (FPN1), is the sole iron export protein reported in mammals and is a key player in both cellular and systemic iron homeostasis [1–3]. It is expressed in all types of cells that handle major iron flow, including macrophages, duodenal enterocytes, hepatocytes, erythrocytes, and placenta syncytiotrophoblasts [4–6]. SLC40A1 activity is predominantly regulated by the liver-derived peptide hepcidin which, depending on the cell type, induces internalization and degradation of SLC40A1 or blocks the SLC40A1-related iron transport mechanism (occlusive effect of hepcidin), thus decreasing iron delivery to plasma [5].

The SLC40A1 gene (LRG_837), which is located on chromosome 2 (2q32.2), is composed of eight coding exons that are spread on approximately 20 kb of genomic DNA and give rise to a 571-amino acid protein of 62.5 kDa (UniProtKB: Q9NP59). The reference transcript (ENST00000261024.6; NM_145585.5) contains an iron-responsive element (IRE) in its 5-untranslated region (UTR), which allows for posttranscriptional regulation via the iron-responsive element/iron regulatory protein (IRE/IRP) system as observed for other important iron metabolism genes [7]. Under low-iron conditions, IRPs bind to the 5-UTR IRE, resulting in inhibition of FPN1 translation and reduction of iron efflux from cells. It is noteworthy that duodenal epithelial and erythroid precursor cells express a shortened SLC40A1 transcript (in addition to the canonical one), which encodes the full-length protein but lacks the IRE and thus escapes translational repression [8].

SLC40A1 belongs to the major facilitator superfamily (MFS), which is the largest group of phylogenetically related secondary active transporters (commonly called “solute carriers,” SLCs). MFS proteins are present in cells across all life kingdoms, controlling the flow of a wide range of substrates (inorganic ions, metabolites, neurotransmitters, and drugs) over lipid bilayers [9, 10]. Despite low sequence similarities, MFS members share a common architecture consisting in twelve transmembrane (TM) helices organized into two structurally similar lobes, each including 6 TMs (N-lobe: TM1-TM6 and C-lobe: TM7-TM12). The two lobes interact together and orchestrate transitions between two extreme conformational states, outward-facing (OF) and inward-facing (IF), according to the alternating access model of membrane transport [11, 12]. During the structural transitions, the substrate sits at the approximate center of the transporter and is not accessible to either side of the membrane.

The existence of an adult, autosomal dominant primary iron overload, soon regarded as an atypical form of hemochromatosis (HC) and now also referred to as ferroportin disease (FD) [13], was based on the description of two SLC40A1 missense mutations, namely, p.Ala77Asp and p.Asn144His, in Italian and Dutch pedigrees [14, 15]. The study of few other families [16] and the achievement of in vitro assays [17, 18] then revealed the existence of two functional categories of SLC40A1 gene mutations underlying two distinct clinical entities. Loss-of-function (LOF) mutations, such as p.Ala77Asp, affect cellular localization of SLC40A1 and/or its iron export function. This results in cellular iron retention, principally in reticuloendothelial cells, and relative plasma iron deficiency. Laboratory tests show a significant increase in serum ferritin, which contrasts with a normal or slightly increased transferrin saturation. As the disease progresses, iron is deposited in parenchymal cells, normally accompanied by a rise in transferrin saturation. This iron overload phenotype, which has historically been associated with HC type 4A, is typical of FD. Gain-of-function (GOF) mutations, such as p.Asp144His, result in partial to complete resistance to hepcidin. This leads to excessive iron export to the bloodstream, increased transferrin saturation, and progressive iron accumulation in parenchymal cells (primarily hepatocytes). These biologic and histologic features mimic the natural history of HFE-related HC; hence, GOF SLC40A1 mutations are associated with HC type 4 (historically referred to as HC type 4B and nowadays also described as nonclassical FD) [19, 20].

The interpretation of missense variations and their association with disease phenotypes is one of the greatest challenges in medical genetics. This is particularly true for SLC40A1-related diseases, which are overwhelmingly associated with rare or private missense variations and are characterized by incomplete penetrance and variable expressivity, with several phenocopies that may confound clinical diagnosis [20]. Pedigree analyses have revealed the influence of age and sex in the progression of iron overload, with clinical phenotypes being mainly observed in adult males [21]. Phenocopy, defined as an environmentally induced, nonhereditary phenotype that mimics one produced by a gene [22], is a growing concern due to increasing descriptions of variants in isolated individuals with hyperferritinemia. In vitro assays are increasingly used to differentiate causal mutations from neutral variations identified in suspected FD or HC type 4 patients. Some experimental data are however questionable, with no obvious relationship with clinical observations [20]. Various in silico prediction tools have also been tested but have not yet been found to be very informative [20, 23], probably due to the nonconsideration of the molecular-level effect of missense variants in the context of SLC40A1 3D structures. The problem may be exacerbated by the existence of several distinct disease mechanisms, some of which combine loss and gain of FPN1 function and/or depend on specific groups of amino acids [24, 25].

In this study, we provide an overview of all SLC40A1 variants that were published before January 2022. SLC40A1 variants associated with relevant phenotypical data were manually curated and entered into a new UMD locus specific database at http://umd.be/SLC40A1/. We summarize the latest in vitro data and provide a structural rationalization of LOF and GOF mutations. We classify all the curated variants according to the American College of Medical Genetics (ACMG) and Association for Molecular Pathology (AMP) guidelines and finally discuss the biological effect and clinical relevance of pathogenic or likely pathogenic mutations and variants of unknown significance.

2. Variants

Each published variation was checked for accuracy and named, or curated, using the Human Genome Variation Society nomenclature (https://varnomen.hgvs.org/) and the canonical full-length reference transcript NM_014585.5 (in human genome GRCh37). Of the total 65 variants compiled in Table 1 (corresponding to 48 different amino acids; Figure 1), 40 were found in single patients or single families (62%), 21 were recurrent in up to four families (32%), and only four (p.Arg178Gln and p.Gly490Asp, p.Gly80Ser, and p.Val162del) were reported in nine, 12, and 20 families, respectively (6%).

(a)

(b)

Figure 1

Distribution of the 65 selected variants along the SLC40A1 gene and the secondary structure of the SLC40A1 protein. (a) The SLC40A1 sequence is represented as a horizontal bar, and each of the segments corresponds to individual exons (coding parts in blue and 5- and 3-untranslated regions in grey), in proportional size and numbered. Introns are represented as a short nonproportional line linking exons. Banners above or below the coding sequence indicate the positions of all variations included in Table 1. Variations that have been studied in vitro are presented in green (loss-of-function), red (gain-of-function), or blue (no functional impact). Missense variations with no functional evidence of deleteriousness or neutrality are presented in grey (for 10 substitutions that occur at the same 8 positions than other variations described as LoF or GoF) or orange (for 8 substitutions that occur at 8 positions where no well-recognized pathogenic variant has been seen to date). (b) The 48 amino acid positions known to be mutated in patients with various iron overload phenotypes reported on the secondary structure topology of the human SLC40A1; amino acids for which variations are associated with a functional defect are colored in green (LoF) or red (GoF), whereas those for which no substitution has been characterized functionally are colored in orange. The human FPN1 protein (UniProtKB: Q9NP59) is made of 571 amino acids.

It is worth mentioning that none of the heterozygous variants currently associated with SLC40A1-related iron overload phenotypes appear to result in a total absence of iron export. This is certainly due to the very important and unique function of the SLC40A1 protein, which is essential for placental iron transport and early embryonic development, dietary iron absorption, cellular iron homeostasis, and iron distribution between tissues [1–3]. The strength of selection acting on heterozygotes is so high (as also indicated by the pLI and LOEUF gnomAD-derived constraint scores: 0.99 and 0.29, respectively [26]) that the missense variation type is almost exclusive (98%). Apart from this type of variation, only the p.Val162 single-amino-acid in-frame deletion has been consistently associated with iron overload phenotypes. The p.Gly468Ser substitution can be considered separately, in that the underlying c.1402G>A transition disrupts the exon 7 donor site, resulting in the partial use of an exonic cryptic splice site and the generation of a truncated reading frame [27, 28], while the Gly468 to Ser amino acid change by itself has no obvious effect on the ability of SLC40A1 to export iron [27].

2.1. Functional Annotation of SLC40A1 Variants according to In Vitro Analyses

The available experimental information, taken from 28 publications and a variety of heterologous cell-based assays (Table 1), was manually curated and used for SLC40A1 variant classification into “loss-of-function” (LoF), “gain-of-function” (GoF), or “neutral” categories. Twenty-one (of 65) SLC40A1 variants that were proven to reduce cell surface expression and/or result in decreased iron export (impact on FPN1 biogenesis or its conformational changes within the lipid bilayer) were classified as LoF (colored in green in Figure 1(a)). Eighteen SLC40A1 variants that did not significantly alter cell surface expression and iron export, but were shown to confer total or partial resistance to hepcidin, were classified as GoF (colored in red in Figure 1(a)). Eight SLC40A1 variants that produced no obvious change from the wild-type protein were listed as “neutral” (colored in blue in Figure 1(a)). It is important to mention that the eight variants were also investigated for their possible effect on pre-mRNA splicing and that no abnormalities were found [29].

No functional data was available for 17 missense variants, while we decided to ignore those published for the p.Ser209Leu substitution because of a substantial discrepancy between the reported biological data in 7 patients, only suggesting FD [30, 31], and the assumption made by An P. and collaborators from in vitro studies of a gain-of-function mechanism (without evidence of altered iron export) [30].

Ten of the eighteen missense variants without functional data were at the same eight positions (Tyr64, Gly80, Arg88, Asn144, Asp157, Leu233, Cys326, and Arg329) than substitutions that were analyzed in vitro and recognized as either LoF or GoF mutations (colored in grey in Figure 1(a)). Five of the eight residues thus considered were found to be strictly conserved in all vertebrates, whereas the remaining three (Asn144, Cys326, and Leu233) diverged only in bacteria (Supplementary Figure 1). Some interesting differences were observed between amino acid pairs in terms of side chain atom composition, polarity, and volume by using the Grantham matrix [32, 33] (Supplementary Table 1), but they were not considered sufficient to predict the deleteriousness of the variants (also see Section 2.3).

The functional importance of Leu129, Val160, Arg179, Ser209, Ala350, Gly494, Val511, and Val531 (colored in orange in Figure 1(a)) is yet to be elucidated. The eight residues are relatively well conserved in vertebrates (Supplementary Figure 1), suggesting the existence of structural and/or functional constraints, but they occur in regions of the protein whose functional importance has not yet been specified. The cellular effects of changes at these positions thus remain elusive.

2.2. SLC40A1 Variant Spectrum and Structural Rationalization of LoF and GoF Mutations

As illustrated in Figure 1(a), the 65 variants investigated in this study have been identified throughout the SLC40A1 coding sequence, with the exception of exons 1 and 2. No hotspot can be clearly defined, although ten residues are found to be recurrently mutated: p.Tyr64, p.Gly80, p.Arg88, p.Asn144n, p.Asp157, p.Gly204, p.Leu233, p.Cys326, p.Arg489, and p.Gly490. Functionally characterized LoF (variants in green) or GoF (variants in red) has been reported at any of these 10 amino acid positions. Similarly, variants have been identified in various topological regions of the SLC40A1 protein, with the exception of the N-terminus and the transmembrane helices 1, 9, and 10 (Figure 1(b)).

Billesbølle et al. recently used cryoelectron microscopy to solve three-dimensional structures of human SLC40A1 (HsSLC40A1) in the outward-facing conformational state, which is regarded as the basal state [34]. The mapping of disease-causing mutations on the HsSLC40A1 3D structure with hepcidin (pdb 6WBV) leads to observe a clear separate distribution into functional subtypes (Figure 2).

(a)

(b)

Figure 2

Ribbon representation of the 3D structure of human SLC40A1 in complex with hepcidin (pdb 6WBV), highlighting the distribution of the loss-of-function mutations (a) and gain-of-function mutations (b). The hepcidin hyposideremic hormone is colored in magenta. Amino acids for which mutations are reported as LoF are colored in green, whereas those for which mutations are reported as GoF are colored in red. Banners indicate the considered amino acid changes and their positions within the protein. The positions of the two divalent metal-binding sites within the central cavity are delimited by blue ellipses. The protein was positioned in a lipid bilayer using the PPM web server [75].

The loss-of-function missense mutations are rather located at the cytoplasmic interface (Figure 2(a)) suggesting that this region may be important for the iron export function. This point was already made by Wallace et al. over 10 years ago [35], based on de novo 3D structure models. As a matter of fact, we have shown that the intracellular gate, which regulates access from the cytoplasm to the central substrate-binding pocket (or central cavity), is stabilized by noncovalent interactions (salt bridges and hydrogen bonds) between the TM2, TM3, TM4, and TM11 transmembrane helices, on the one hand, and the TM5 and TM10 transmembrane helices, on the other hand [25]. We identified 12 residues (Asp81, Asp84, Lys85, Arg88, Asp157, Asn174, Arg178, Asp473, Gln478, Arg489, Gly490, and Ile491) that are critical for protein stability and/or conformational changes during the iron transport cycle. We pointed out that a total of 13 missense variations affecting 7 (Asp84, Arg88, Asp157, Asn174, Arg178, and Gly490) of the 12 considered gating residues were reported in 77 patients with common features of FD. These missense variations could be responsible for two distinct loss-of-function mechanisms: protein misfolding and reduced expression on the cell surface, or failure to export iron [25, 36, 37].

The gain-of-function mutations are rather concentrated around the central cavity (Figure 2(b)), which actually contains two divalent metal-binding sites (one in each lobe) and the hepcidin-binding site [34]. This last one is centered on the Cys326 residue, which is strictly conserved in all vertebrates and may have evolved in conjunction with hepcidin to control systemic iron homeostasis [38]. At first glance, a functional dichotomy seemed to emerge between missense mutations located in the C-lobe, in particular those affecting positions Cys326, Asp504, and His507, which were associated with the highest degree of hepcidin resistance, and missense mutations located in the N-lobe, in particular those affecting the recurrently mutated Gly204 and Asn144 positions, which were found to induce partial resistance to hepcidin [39]. But things could actually be more complex and depend on the effect of mutations on hepcidin binding (on each lobe) and/or hepcidin-dependent ubiquitination (in the case of hepcidin-induced endocytosis and degradation of SLC40A1) following subtle conformational changes [24, 34]. The effects of different mutations may also vary depending on the cell type [24].

Given the proximity between amino acids participating in the binding of iron or hepcidin and the involvement of some gating residues in conformational changes also governing the iron export function and its regulation by hepcidin, one can anticipate the existence of a third class of SLC40A1 pathogenic variants with both loss- and gain-of-function features. Examples supported by in vitro studies exist in the literature, but they are still very limited [24, 40, 41].

2.3. Structural Annotation of SLC40A1 Missense Variants without Functional Data

To better predict effects of the SCL40A1 missense variants occurring at recurrently mutated positions, especially those that are more likely to be tolerated (substitution with chemical similar amino acids), we leveraged the 3D structure of human SLC40A1 (in its outward-open facing conformation) taking into account current knowledge about the iron export mechanism and its regulation by hepcidin.

As aforesaid, Arg88, Asp157, and Arg489 play a critical role in the architecture of the human SLC40A1 intracellular gate [25]. The loss of a salt bridge in the cytoplasmic network connecting these three charged amino acids can result in folding defects in TM helices TM3, TM4, or TM11 and their assembly or local instability of the SLC40A1 structure in its outward-open conformation. In this regard, it is very likely that the p.Arg88Thr, p.Arg88Ile, p.Asp157Asn, p.Asp157Ala, and p.Arg489Ser missense variants reduce ability of SLC40A1 to export iron out of cells.

As shown in Figure 3(a), some other 3D structure features are likely to contribute to the stability of the human SLC40A1 inner gate. First, a tight packing of the helices TM2 and TM11 is ensured by two glycine zipper motifs (formed on the one hand by TM2 Gly76 and Gly80 and on the other hand by TM11 Gly490 and Gly494), which face each other. The glycine-to-aspartic acid 494 substitution in such an environment is predicted to cause local structure instability, as the p.Gly490Asp substitution which appears to be particularly frequent in patients originating from the Reunion Island and has been associated with a significant reduction of the SLC40A1 cell surface expression [29]. Second, an additional bond (H bond) is formed between the side chain of Arg88 and the oxygen main-chain atom of Leu233, further stabilizing the Arg489-Asp157-Arg88 interaction network. The Leu hydrophobic amino acid, which emerges at the end of a short helix that is located immediately after TM6, also makes CH-𝛑 contacts with Trp158 within the hydrophobic core of the gate. The leucine-to-proline 233 substitution, which has also been proven to reduce the cell surface expression of SLC40A1 [29], is likely to affect the interaction with Trp158 but also, via a modification of the local conformation of the short helix, lead to the loss of the H bond with Arg88. The p.Leu233Val missense mutation might have a more moderate effect, impairing the interaction with Trp158 without necessarily affecting the conformation of the small helix and the H bond with Arg88.

(a)

(b)

It has previously been established that the thiol group in the side chain of Cys326 in the SLC40A1 C-lobe is mandatory for hepcidin binding [42], and it has recently been demonstrated that the sulfur atom of this particular amino acid interacts with the main-chain oxygen atoms (O and OXT) of hepcidin Thr25 (3.28 and 3.78 Å; Figure 3(b)) [34]. As the functionally characterized and well-defined p.Cys326Ser and p.Cys326Tyr GoF mutants (Table 1), it must therefore be assumed that the p.Cys326Phe missense variant hampers the posttranslational regulation of SLC40A1 by hepcidin.

As shown in Figure 3(b), Tyr64 and Asn144 make part of the hepcidin-binding pocket within the SLC40A1 N-lobe. The Asn144 side chain atoms interact with the hydroxyl atom of Tyr501 (3.4 Å and 3.7 Å with the OD1 and ND2 atoms of Asn144), which stacks with the imidazole side chain of hepcidin His3. Tyr64 is located less deeply in the pocket, interacting with hepcidin Phe4 and Ile6, and it has been suggested that the outward displacement of TM2 near this amino acid pun hepcidin binding plays a key role in SLC40A1 ubiquitination [34]. In support to this idea, the p.Tyr64Asn missense mutation has been consistently reported to reduce SLC40A1 ubiquitination and prevent its intracellular degradation, without altering hepcidin binding [24, 43]. It can be assumed that the p.Tyr64His substitution produces similar effects. Regarding the SLC40A1-Tyr501/hepcidin-His3 interaction, it is worthwhile to remember the results obtained by Aschemeyer et al. showing that the p.Tyr501Cys mutant is no longer able to interact with hepcidin, in a manner comparable to the p.Cys326Ser mutant [24]. Less obvious, perhaps, is the situation of the mutants described at the Asn144 position (p.Asn144His, p.Asn144Asp, and Asn144Thr), which have been investigated by different groups and have been most commonly reported as partial GoF variants with no formal demonstration of a loss of interaction between hepcidin and SLC40A1 (all references provided in Table 1). This is well exemplified by the p.Asn144Asp variant which has been alternatively described as a variant that prevents binding to hepcidin [24] and as a variant that does not (or in a very partial manner [43]), although both studies indicate that SLC40A1 resistance to hepcidin is incomplete.

2.4. Performance of In Silico Prediction Tools for the Classification of SLC40A1 Missense Variants

There are a number of software algorithms which can be used to predict the impact of novel missense variants on protein function. In the present study, we evaluated the performance and reliability of ten prediction tools commonly used in medical genetics and by different web-based classification tools, namely, SIFT [44], PolyPhen-2 [45], MutationTaster2 [46], GERP++ [47], MetaLR [48], MetaSVM [48], UMD-Predictor [49], REVEL [50], CADD [51, 52], and DANN [53]. Our observations are summarized in Supplementary Table 2 and Supplementary Figure 2.

As they stand (i.e., considering 48 pathogenic variants and only 8 substitutions recognized as benign), our observations are consistent with a variety of results from similar studies [54–57], each confirming that computational methods generate a low rate of false negative but a high rate of false positive. The combination of two predictive neural networks (CADD and REVEL), as tested for the first time in the present study for SLC40A1 variants (versus older studies where in silico prediction tools were tested separately [20, 23]), did not overcome this drawback. At best, it may help focus resources on variants that could warrant family studies and/or functional explorations.

3. The UMD-SLC40A1 Database

We have established a web-based, manually curated database that gives a complete overview of the SLC40A1 variants which have been published in the medical literature and have been associated with mild to severe iron overload phenotypes. This database has been developed to aid both clinicians and scientists. It has been constructed from the generic software called UMD (“Universal Mutation Database” [58]), which is today recognized by the Human Genome Organization (HUGO) and the Human Genome Variation Society (HGVS) as a reference tool to build the Locus Specific Databases (LSDB). All the software has been built with the 4^th dimensional language (4D). UMD-SLC40A1 is a dynamic database, in the meaning that it also includes various computerized tools for the analysis of variants. Project and online publishing was approved by the French supervisory authority (“Commission Nationale pour l’Informatique et les Libertés”; registration no. 908361) and the national ethics committee, Comité Consultatif pour le Traitement de l’Information en matière de Recherche dans le domaine de la Santé (no. 07.421), and registered under no. 91513. The database will be freely accessible online at http://umd.be/SLC40A1/.

UMD-SLC40A1 currently contains the 65 variants presented in this article, which have been split into 336 entries (one per patient). We excluded a few patients harboring the p.Arg88Gly, p.Cys326Ser, p.Asp181Val, p.Gly204Ser, p.His507Arg, or p.Ser209Leu missense mutation due to the lack of individual phenotypic information [30, 59–63]. A unique entry number was created for each patient, and a second number was generated for each family; relatives () were assigned a number based on the index case () and the family numbers. Phenotypic data had to include gender, age at diagnosis, transferrin saturation, and serum ferritin. Other clinical data were systematically recorded when available; these included laboratory findings (hemoglobin, hematocrit, red blood cells, mean corpuscular volume, C-reactive protein, γ-glutamyl transferase, aspartate transaminase, alanine transaminase, and so on), magnetic resonnance imaging and/or biopsy data (hepatic iron concentration, tissue-specific localization of iron) and clinical manifestation of iron overload (asthenia, arthralgia, skin pigmentation, type 1 or type 2 diabetes mellitus). Patterns of excessive alcohol consumption and the metabolic syndrome, which cause hyperferritinemia, with or without associated iron overload, were also documented. The standard PubMed identifier (PMID) of each publication where phenotypic data were retrieved was entered into the database and linked to the relevant entry numbers. Relatives with negative SLC40A1 genetic testing were not included in the database.

Each variant was classified using a process consistent with the guidelines of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology in one of the following five classes [64]: benign variant (BV, class 1), likely benign variant (LBV, class 2), variant of uncertain significance (VUS, class 3), likely pathogenic variant (LPV, class 4), and pathogenic variant (PV, class 5). We informed manually the criteria based on our literature review and the GnomAD population database. For familial segregation, we calculated the cosegregation score as described by Jarvik and Browning [65].

Overall, we have described each variant in accordance with its (i) positions in the SLC40A1 gene (c.) and the human SLC40A1 protein (p.), also considering the protein structure (TM: transmembrane helices and ECL or ICL: extra- or intracellular loops), (ii) reported functional effect (loss-of-function, gain-of-function, and neutral), (iii) ACMG/AMP classification, and (iv) relationship with clinical phenotypes (ferroportin disease, hemochromatosis type 4, and unexplained hyperferritinemia).

4. Clinical Relevance of SLC40A1 Variants and Genotype-Phenotype Correlations

Among the phenotypes associated with SLC40A1 pathogenic or likely pathogenic mutations, ferroportin disease appears to be the most widely represented, with 29 loss-of-function mutations identified in 227 patients (96 index cases and 131 relatives). Hemochromatosis type 4 is associated with 21 gain-of-function mutations, identified in a total of 81 patients (27 index cases and 54 relatives). Unexplained hyperferritinemia only concerns 35 patients (22 index cases and 13 relatives), 17 of which have a variant that remains classified as a variant of uncertain significance and 18 have a benign or likely benign variant. The clinical characteristics and outcomes of all 343 patients are shown in Supplementary Table 3.

FD, HC4, and UH probands were identified at similar ages (Figure 4(a) and Table 2), most often after age 40 (mean, 44.6, 48.7, and 51.7 years, respectively). However, they presented different biological profiles (Figures 4(b) and 4(c)). Serum ferritin concentrations were significantly higher in FD and HC4 patients (median, 2,258 μg/L interquartile range: (1,292; 4,356) and 2,195 μg/L (1,000; 4,326) vs. 1,008 μg/L (777; 1,733); ), while transferrin saturation levels were significantly higher in HC4 patients (mean, 83 vs. 37 and 45%; ). A fairly clear division also appeared between FD and HC4 probands when transferrin saturation and serum ferritin were matched together (Figure 4(d)).

(a)

(b)

(c)

(d)

Figure 4

Iron overload profiles of the 96 probands with ferroportin disease (FD), 27 probands with hemochromatosis type 4 (HC4), and 22 probands with unexplained hyperferritinemia (UH). Distributions of age (a), serum ferritin (b), and transferrin saturation (c); medians (white dot on the violin plots) and interquartile ranges (the black bars in the center of violins) are shown. values were calculated on Student’s -test; and . (d) Relationship between serum ferritin and transferrin saturation in FD (green), HC4 (red), and UH patients (orange).

FD is characterized by iron accumulation in macrophages, leading to increased production of ferritin, and it releases into plasma. This can occur quite early in life [13] and can continue until very high serum ferritin concentrations that do not necessarily correlate with body iron stores [19]. This is well illustrated in Figure 5(a), which shows a positive correlation between age and serum ferritin in patients with a LoF mutation (, ). Further data are provided in Figure 5(b), showing a clear distinction in the ferritin/age ratio between LoF mutation patients and both GoF mutation patients and ACMG classes 1-3 variant patients. At referral (i.e., diagnosis or familial screening), 75% of LoF mutation patients had a ferritin/age ; this value was reached by only 38% of GoF mutation patients and 20% of patients with an unrecognized pathogenic variant.

(a)

(b)

(c)

(d)

(e)

Figure 5

Iron overload profile of the 227 patients with ferroportin disease (FD), 81 patients with hemochromatosis type 4 (HC4), and 35 patients with unexplained hyperferritinemia (UH). (a) Relationship between serum ferritin and age. Linear regression lines with 95% confidence interval have been fitted to the values from FD (green) and HC (red) patients. Pearson’s correlation coefficients () are provided. (b) Box plots show median and range for the ferritin/age ratio. values were calculated on Student’s -test; . (c) Categorization of the FD (green), HC4 (red), and UH (orange) patients according to serum ferritin and transferrin saturation cutoffs. Dark colors represent the index cases, whereas light colors represent the relatives. Sizes of the circles are proportional to the number of patients. (d) Relationship between mean corpuscular volume and transferrin saturation. Linear regression lines with 95% confidence interval have been fitted to the values from FD (green) and HC (red) patients. Pearson’s correlation coefficients () are provided. (e) Frequency distribution of clinical signs observed in FD (green), HC4 (red), and UH (orange) patients.

HC type 4 is usually associated with parenchymal iron deposition similar to HFE-related hemochromatosis and other rare forms of the disease resulting from inadequate levels of serum hepcidin to control systemic iron homeostasis. An elevated transferrin saturation level is considered a sign of early disease, which always precedes an elevated serum ferritin. This is well illustrated in Figure 5(c), which shows a very clear separation of patient groups according to mutation classes. Interestingly, patients with GoF mutations located within the FPN1 C-lobe had higher transferrin saturation values. There was no clear association with age at diagnosis compared with the group of patients with GoF mutations located in the FPN1 N-lobe (Supplementary Figures 3A and 3B). Rather, this could be a consequence of more important hepcidin resistance effects, especially for mutations that affect critical residues (hepcidin and/or iron binding) in transmembrane helices 7 and 11. With a slightly larger data set than that constituted by Zhang et al. in 2018, we confirm another interesting observation, which is the positive correlation between transferrin saturation and size of the red blood cells in GoF mutation patients (Figure 5(d); , ). The cellular mechanism leading to this observation is not well understood, but it might be related to the specific function of SLC40A1 in erythroid cells to support hemoglobin production while avoiding iron toxicity and participating in the maintenance of systemic iron homeostasis [5].

Information on clinical manifestation in patients with disease-causing mutations is quite sparse in literature (Figure 5(e)). This is particularly true for patients with a LoF mutation, which may suggest that macrophages are relatively tolerant to iron accumulation. Over time, however, iron can escape into the bloodstream, resulting in increased transferrin saturation as shown in Supplementary Figures 3C and 3D and iron deposition in parenchymal cells. This makes more likely iron-induced oxidative stress and tissue injury [19]. In line with this idea, we observed that 10 of the 16 LoF mutation patients in whom liver fibrosis had been reported also had high transferrin saturation levels (≥ 60% in men and ≥ 50% in women), pointing that three transferrin saturation values were missing in the 16 considered patients. GoF mutation patients had the highest proportions of clinical signs. They also presented with higher aspartate (47 vs. 27 UI/L; ) and alanine transaminase (83 vs. 31 UI/L; ) values (Table 2), which is classically considered a sign of hepatocellular injury.

There is a general trend toward mild hyperferritinemia without evidence of plasma iron overload (transferrin saturation in females and <60% in males) in patients with variants classified in ACMG categories 1-3 (Figures 4(b)–4(d) and 5(c)), and rather unexpectedly, this trend is more evident in patients with a variant of uncertain significance (Supplementary Figures 3E and 3F). It should be mentioned here that two variants with questionable classification, namely, p.Leu129Pro and p.Ser209Leu, account for a total of three probands and nine relatives (Supplementary Table 4). Iron parameters in the three probands were rather suggestive of FD. However, none of them had a serum ferritin greater than 1000 μg/L in age at diagnosis between 43 and 59 years. Although the penetrance of the different variants associated with FD is thought to be incomplete, it can further be noted that none of the relatives with the p.Leu129Pro had serum ferritin concentrations clearly above the normal range (338 μg/L at most in a 73-year-old male). The p.Ser209Leu variant was studied in vitro, but contrary to what the phenotypes of the 7 patients of Asian origin reported in the literature suggest, it was associated with a GoF effect [30]. Also, it is not very rare in the general population (allele frequency in the Genome Aggregation Database ) and more particularly in the eastern region of Asia where its estimating allele frequency reaches 0.06%. New phenotypic data and improved functional analysis would allow to reclassify the two variants, most likely as nonpathogenic.

5. Conclusion and Future Prospects

With the increasing description of case reports in the literature, or small family structures with hyperferritinemia, phenotyping is emerging as an important source of uncertainty in the diagnosis of FD and HC4. In such a context, no physician prescribing a genetic test should forget that approximately 90% of patients with mildly elevated serum ferritin do not have iron overload [66, 67], and that magnetic resonance imaging of the abdomen is a very reliable noninvasive tool to confirm iron excess in the liver, spleen, and/or bone marrow and possibly categorize FD and HC4 [68]. On the other hand, no biologist should forget that benign, but rare, missense variants represent a nonnegligible part of human genetic diversity [26] and that this also applies to the SLC40A1 gene. Less than one-third of the 213 missense variants reported in GnomAD at a are actually suspected to cause iron overload. The “ferroportin score” very recently proposed by Landemaine et al. to increase the proficiency of genetic testing for FD [69] could conceivably also be used to better distinguish pathogenic loss-of-function mutations from neutral rare variants identified by chance in patients with a secondary (but perhaps not clearly defined) cause of hyperferritinemia; we could not test this score because of the piecemeal information available in the literature (only 9 patients for whom the 5 considered criteria could have been reliably used).

We have established a web-based database which gives a comprehensive update of the all uncommon SLC40A1 coding variants that have been described in literature and individually associated with biological and clinical findings. It will help investigators and clinicians to better distinguish causal mutations from rare polymorphisms. Our database will also allow inclusion of new cases, hoping to contribute something even more determinant in the progression of knowledge and understanding of the molecular bases of FD and HC4. There is considerable variation in the data currently available regarding the onset, progression, and type of clinical manifestations that a patient may develop. The cause of this variability is not fully understood and can only be partially explained by a lack of penetrance, the influence of nongenetic factors, and the description of two categories of causative mutations.

That almost all of the variants described at the SLC40A1 locus correspond to missense variants undeniably represents an important challenge in the quest for an optimal clinical interpretation, and state-of-the-art prediction tools do not really help. False-negative predictions are rare, and MutationTaster might be judged an appropriate approach for pathogenicity exclusion (at least), but specificity is a recurrent problem that mandates the use of additional data. Interestingly, we have noted that mechanisms associated with the pathogenesis of LoF and GoF missense variants pretty well correlate with particular regions of the SLC40A1 three-dimensional structure. Similar observations have been made in the context of many other genetic diseases [70]. The combination of structural and functional analyses is a promising way to further clarify the SLC40A1 biology and the molecular mechanism of downregulation by hepcidin and, by the way, improve predictions of pathogenicity.

Data Availability

All information about referenced variants is recorded in a locus-specific online database that is accessible at http://umd.be/SLC40A1/. All other data are available upon request.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Authors’ Contributions

K. Uguen and G. Le Gac designed the study. C. Ka, M. Le Tertre, J. Guellec, C. Ferec, and G. Collod-Beroud participated in the literature review and analysis of the data. I. Callebaut conducted the structural analyses. C. Beroud helped to create the UMD_SLC40A1 database. K. Uguen, I. Callebaut, and G. Le Gac wrote the manuscript. All authors contributed to the editing of the final manuscript. Isabelle Callebaut and Gérald Le Gac share colast authorship.

Acknowledgments

This work was supported by grants from the French Hospital Clinical Research Program (Programme Hospitalier de Recherche Clinique 2009, Brest University Hospital (UF0857)) and the Gaetan Saleun Association.

Supplementary Materials

Supplementary Figure S1: multiple species sequence alignment of ferroportin. Supplementary Figure S2: scatter plot visualisation of CADD and REVEL scores for pathogenic variants, neutral variants, and variants of unknown significance. Supplementary Figure S3: iron overload profiles of the patients with hemochromatosis type 4, ferroportin disease, or unexplained hyperferritinemia. Supplementary Table S1: Grantham scores for the 10 amino acid changes without functional data. Supplementary Table S2: comparison of in silico predictors in a subset of 56 SLC40A1 missense variants (provided as an Excel file). Supplementary Table S3: clinical and biological data of the 343 patients reported in the literature. Supplementary Table S4: clinical, biological data of the 17 patients with functionally unexplored variants. (Supplementary Materials)

References

G. J. Anderson and D. M. Frazer, “Current understanding of iron homeostasis^†‡,” The American Journal of Clinical Nutrition, vol. 106, Suppl 6, pp. 1559S–1566S, 2017.
View at: Publisher Site | Google Scholar
H. Drakesmith, E. Nemeth, and T. Ganz, “Ironing out ferroportin,” Cell Metabolism, vol. 22, no. 5, pp. 777–787, 2015.
View at: Publisher Site | Google Scholar
M. W. Hentze, M. U. Muckenthaler, B. Galy, and C. Camaschella, “Two to tango: regulation of mammalian iron metabolism,” Cell, vol. 142, no. 1, pp. 24–38, 2010.
View at: Publisher Site | Google Scholar
A. Donovan, C. A. Lima, J. L. Pinkus et al., “The iron exporter ferroportin/Slc40a1 is essential for iron homeostasis,” Cell Metabolism, vol. 1, no. 3, pp. 191–200, 2005.
View at: Publisher Site | Google Scholar
D.-L. Zhang, M. C. Ghosh, H. Ollivierre, Y. Li, and T. A. Rouault, “Ferroportin deficiency in erythroid cells causes serum iron deficiency and promotes hemolysis due to oxidative stress,” Blood, vol. 132, no. 19, pp. 2078–2087, 2018.
View at: Publisher Site | Google Scholar
D.-L. Zhang, J. Wu, B. N. Shah et al., “Erythrocytic ferroportin reduces intracellular iron accumulation, hemolysis, and malaria risk,” Science, vol. 359, no. 6383, pp. 1520–1523, 2018.
View at: Publisher Site | Google Scholar
M. U. Muckenthaler, B. Galy, and M. W. Hentze, “Systemic iron homeostasis and the iron-responsive element/iron-regulatory protein (IRE/IRP) regulatory network,” Annual Review of Nutrition, vol. 28, no. 1, pp. 197–213, 2008.
View at: Publisher Site | Google Scholar
D.-L. Zhang, R. M. Hughes, H. Ollivierre-Wilson, M. C. Ghosh, and T. A. Rouault, “A ferroportin transcript that lacks an iron-responsive element enables duodenal and erythroid precursor cells to evade translational repression,” Cell Metabolism, vol. 9, no. 5, pp. 461–473, 2009.
View at: Publisher Site | Google Scholar
E. Perland, S. Bagchi, A. Klaesson, and R. Fredriksson, “Characteristics of 29 novel atypical solute carriers of major facilitator superfamily type: evolutionary conservation, predicted structure and neuronal co-expression,” Open Biology, vol. 7, no. 9, article 170142, 2017.
View at: Publisher Site | Google Scholar
N. Yan, “Structural biology of the major facilitator superfamily transporters,” Annual Review of Biophysics, vol. 44, no. 1, pp. 257–283, 2015.
View at: Publisher Site | Google Scholar
D. Drew, R. A. North, K. Nagarathinam, and M. Tanabe, “Structures and general transport mechanisms by the major facilitator superfamily (MFS),” Chemical Reviews, vol. 121, no. 9, pp. 5289–5335, 2021.
View at: Publisher Site | Google Scholar
E. M. Quistgaard, C. Löw, F. Guettou, and P. Nordlund, “Understanding transport by the major facilitator superfamily (MFS): structures pave the way,” Nature Reviews Molecular Cell Biology, vol. 17, no. 2, pp. 123–132, 2016.
View at: Publisher Site | Google Scholar
A. Pietrangelo, “Ferroportin disease: pathogenesis, diagnosis and treatment,” Haematologica, vol. 102, no. 12, pp. 1972–1984, 2017.
View at: Publisher Site | Google Scholar
G. Montosi, A. Donovan, A. Totaro et al., “Autosomal-dominant hemochrom-atosis is associated with a mutation in the ferroportin (SLC11A3) gene,” The Journal of Clinical Investigation, vol. 108, no. 4, pp. 619–623, 2001.
View at: Publisher Site | Google Scholar
O. T. Njajou, N. Vaessen, M. Joosse et al., “A mutation in SLC11A3 is associated with autosomal dominant hemochromatosis,” Nature Genetics, vol. 28, no. 3, pp. 213-214, 2001.
View at: Publisher Site | Google Scholar
G. Le Gac and C. Férec, “The molecular genetics of haemochromatosis,” European Journal of Human Genetics, vol. 13, no. 11, pp. 1172–1185, 2005.
View at: Publisher Site | Google Scholar
H. Drakesmith, L. M. Schimanski, E. Ormerod et al., “Resistance to hepcidin is conferred by hemochromatosis-associated mutations of ferroportin,” Blood, vol. 106, no. 3, pp. 1092–1097, 2005.
View at: Publisher Site | Google Scholar
L. M. Schimanski, H. Drakesmith, A. T. Merryweather-Clarke et al., “In vitro functional analysis of human ferroportin (FPN) and hemochromatosis-associated FPN mutations,” Blood, vol. 105, no. 10, pp. 4096–4102, 2005.
View at: Publisher Site | Google Scholar
A. Piperno, S. Pelucchi, and R. Mariani, “Inherited iron overload disorders,” Translational gastroenterology and hepatology, vol. 5, p. 25, 2020.
View at: Publisher Site | Google Scholar
L. T. Vlasveld, R. Janssen, E. Bardou-Jacquet et al., “Twenty years of ferroportin disease: a review or an update of published clinical, biochemical, molecular, and functional features,” Pharmaceuticals, vol. 12, no. 3, p. 132, 2019.
View at: Publisher Site | Google Scholar
C. Le Lan, A. Mosser, M. Ropert et al., “Sex and acquired cofactors determine phenotypes of ferroportin disease,” Gastroenterology, vol. 140, no. 4, pp. 1199–1207.e2, 2011.
View at: Publisher Site | Google Scholar
C. Klein, R. Chuang, C. Marras, and A. E. Lang, “The curious case of phenocopies in families with genetic Parkinson’s disease,” Movement Disorders, vol. 26, no. 10, pp. 1793–1802, 2011.
View at: Publisher Site | Google Scholar
R. Mayr, A. R. Janecke, M. Schranz et al., “Ferroportin disease: a systematic meta-analysis of clinical and molecular findings,” Journal of Hepatology, vol. 53, no. 5, pp. 941–949, 2010.
View at: Publisher Site | Google Scholar
S. Aschemeyer, B. Qiao, D. Stefanova et al., “Structure-function analysis of ferroportin defines the binding site and an alternative mechanism of action of hepcidin,” Blood, vol. 131, no. 8, pp. 899–910, 2018.
View at: Publisher Site | Google Scholar
J. Guellec, A. Elbahnsi, M. Le Tertre et al., “Molecular model of the ferroportin intracellular gate and implications for the human iron transport cycle and hemochromatosis type 4A,” The FASEB Journal, vol. 33, no. 12, pp. 14625–14635, 2019.
View at: Publisher Site | Google Scholar
S. Gudmundsson, M. Singer-Berk, N. A. Watts et al., “Variant interpretation using population databases: lessons from gnomAD,” Human Mutation, vol. 43, no. 8, pp. 1012–1030, 2022.
View at: Publisher Site | Google Scholar
M. Le Tertre, C. Ka, L. Raud et al., “Splicing analysis of SLC40A1 missense variations and contribution to hemochromatosis type 4 phenotypes,” Blood Cells, Molecules & Diseases, vol. 87, article 102527, 2021.
View at: Publisher Site | Google Scholar
P. L. Lee, T. Gelbart, C. West, and J. C. Barton, “SLC40A1 c.1402G-->a results in aberrant splicing, ferroportin truncation after glycine 330, and an autosomal dominant hemochromatosis phenotype,” Acta Haematologica, vol. 118, pp. 237–241, 2008.
View at: Publisher Site | Google Scholar
I. Callebaut, R. Joubrel, S. Pissard et al., “Comprehensive functional annotation of 18 missense mutations found in suspected hemochromatosis type 4 patients,” Human Molecular Genetics, vol. 23, no. 17, pp. 4479–4490, 2014.
View at: Publisher Site | Google Scholar
P. An, L. Jiang, Y. Guan et al., “Identification of hereditary hemochromatosis pedigrees and a novel SLC40A1 mutation in Chinese population,” Blood Cells, Molecules & Diseases, vol. 63, pp. 34–36, 2017.
View at: Publisher Site | Google Scholar
Y. Wang, Y. Du, G. Liu et al., “Identification of novel mutations in HFE, HFE2, TfR2, and SLC40A1 genes in Chinese patients affected by hereditary hemochromatosis,” International Journal of Hematology, vol. 105, no. 4, pp. 521–525, 2017.
View at: Publisher Site | Google Scholar
R. Grantham, “Amino acid difference formula to help explain protein evolution,” Science, vol. 185, no. 4154, pp. 862–864, 1974.
View at: Publisher Site | Google Scholar
W. H. Li, C. I. Wu, and C. C. Luo, “Nonrandomness of point mutation as reflected in nucleotide substitutions in pseudogenes and its evolutionary implications,” Journal of Molecular Evolution, vol. 21, no. 1, pp. 58–71, 1984.
View at: Publisher Site | Google Scholar
C. B. Billesbølle, C. M. Azumaya, R. C. Kretsch et al., “Structure of hepcidin-bound ferroportin reveals iron homeostatic mechanisms,” Nature, vol. 586, no. 7831, pp. 807–811, 2020.
View at: Publisher Site | Google Scholar
D. F. Wallace, J. M. Harris, and V. N. Subramaniam, “Functional analysis and theoretical modeling of ferroportin reveals clustering of mutations according to phenotype,” American Journal of Physiology-Cell Physiology, vol. 298, no. 1, pp. C75–C84, 2010.
View at: Publisher Site | Google Scholar
C. Ka, J. Guellec, X. Pepermans et al., “The SLC40A1 R178Q mutation is a recurrent cause of hemochromatosis and is associated with a novel pathogenic mechanism,” Haematologica, vol. 103, no. 11, pp. 1796–1805, 2018.
View at: Publisher Site | Google Scholar
L. T. Vlasveld and D. W. Swinkels, “Loss-of-function ferroportin disease: novel mechanistic insights and unanswered questions,” Haematologica, vol. 103, no. 11, pp. 1753–1755, 2018.
View at: Publisher Site | Google Scholar
E. Nemeth and T. Ganz, “Hepcidin-ferroportin interaction controls systemic iron homeostasis,” IJMS, vol. 22, no. 12, p. 6493, 2021.
View at: Publisher Site | Google Scholar
M. Le Tertre, C. Ka, J. Guellec et al., “Deciphering the molecular basis of ferroportin resistance to hepcidin: structure/function analysis of rare SLC40A1 missense mutations found in suspected hemochromatosis type 4 patients,” Transfusion Clinique et Biologique, vol. 24, no. 4, pp. 462–467, 2017.
View at: Publisher Site | Google Scholar
G. Le Gac, C. Ka, R. Joubrel et al., “Structure-function analysis of the human ferroportin iron exporter (SLC40A1): effect of hemochromatosis type 4 disease mutations and identification of critical residues,” Human Mutation, vol. 34, no. 10, pp. 1371–1380, 2013.
View at: Publisher Site | Google Scholar
A. Viveiros, M. Panzer, N. Baumgartner et al., “Reduced iron export associated with hepcidin resistance can explain the iron overload spectrum in ferroportin disease,” Liver International, vol. 40, no. 8, pp. 1941–1951, 2020.
View at: Publisher Site | Google Scholar
G. Preza, A. Fernandes, R. J. Clark, D. J. Craik, T. Ganz, and E. Nemeth, “Structural aspects of hepcidin-ferroportin binding,” Blood, vol. 112, no. 11, pp. 119–119, 2008.
View at: Publisher Site | Google Scholar
A. Fernandes, G. C. Preza, Y. Phung et al., “The molecular basis of hepcidin-resistant hereditary hemochromatosis,” Blood, vol. 114, no. 2, pp. 437–443, 2009.
View at: Publisher Site | Google Scholar
P. C. Ng and S. Henikoff, “Predicting deleterious amino acid substitutions,” Genome Research, vol. 11, no. 5, pp. 863–874, 2001.
View at: Publisher Site | Google Scholar
I. A. Adzhubei, S. Schmidt, L. Peshkin et al., “A method and server for predicting damaging missense mutations,” Nature Methods, vol. 7, no. 4, pp. 248-249, 2010.
View at: Publisher Site | Google Scholar
J. M. Schwarz, D. N. Cooper, M. Schuelke, and D. Seelow, “MutationTaster2: mutation prediction for the deep-sequencing age,” Nature Methods, vol. 11, no. 4, pp. 361-362, 2014.
View at: Publisher Site | Google Scholar
E. V. Davydov, D. L. Goode, M. Sirota, G. M. Cooper, A. Sidow, and S. Batzoglou, “Identifying a high fraction of the human genome to be under selective constraint using GERP++,” PLoS Computational Biology, vol. 6, no. 12, article e1001025, 2010.
View at: Publisher Site | Google Scholar
C. Dong, P. Wei, X. Jian et al., “Comparison and integration of deleteriousness prediction methods for nonsynonymous SNVs in whole exome sequencing studies,” Human Molecular Genetics, vol. 24, no. 8, pp. 2125–2137, 2015.
View at: Publisher Site | Google Scholar
D. Salgado, J.-P. Desvignes, G. Rai et al., “UMD-predictor: a high-throughput sequencing compliant system for pathogenicity prediction of any human cDNA substitution,” Human Mutation, vol. 37, no. 5, pp. 439–446, 2016.
View at: Publisher Site | Google Scholar
N. M. Ioannidis, J. H. Rothstein, V. Pejaver et al., “REVEL: an ensemble method for predicting the pathogenicity of rare missense variants,” American Journal of Human Genetics, vol. 99, no. 4, pp. 877–885, 2016.
View at: Publisher Site | Google Scholar
M. Kircher, D. M. Witten, P. Jain, B. J. O’Roak, G. M. Cooper, and J. Shendure, “A general framework for estimating the relative pathogenicity of human genetic variants,” Nature Genetics, vol. 46, no. 3, pp. 310–315, 2014.
View at: Publisher Site | Google Scholar
P. Rentzsch, D. Witten, G. M. Cooper, J. Shendure, and M. Kircher, “CADD: predicting the deleteriousness of variants throughout the human genome,” Nucleic Acids Research, vol. 47, no. D1, pp. D886–D894, 2019.
View at: Publisher Site | Google Scholar
D. Quang, Y. Chen, and X. Xie, “DANN: a deep learning approach for annotating the pathogenicity of genetic variants,” Bioinformatics, vol. 31, no. 5, pp. 761–763, 2015.
View at: Publisher Site | Google Scholar
C. Ernst, E. Hahnen, C. Engel et al., “Performance of in silico prediction tools for the classification of rare BRCA1/2 missense variants in clinical diagnostics,” BMC Medical Genomics, vol. 11, no. 1, p. 35, 2018.
View at: Publisher Site | Google Scholar
I. U. S. Leong, A. Stuckey, D. Lai, J. R. Skinner, and D. R. Love, “Assessment of the predictive accuracy of five in silico prediction tools, alone or in combination, and two metaservers to classify long QT syndrome gene mutations,” BMC Medical Genetics, vol. 16, no. 1, p. 34, 2015.
View at: Publisher Site | Google Scholar
L. Meléndez-Aranda, A. R. Jaloma-Cruz, N. Pastor, and M. M. . J. Romero-Prado, “In silico analysis of missense mutations in exons 1-5 of the F9 gene that cause hemophilia B,” BMC Bioinformatics, vol. 20, no. 1, p. 363, 2019.
View at: Publisher Site | Google Scholar
L. A. Miosge, M. A. Field, Y. Sontani et al., “Comparison of predicted and actual consequences of missense mutations,” Proceedings of the National Academy of Sciences, vol. 112, no. 37, pp. E5189–E5198, 2015.
View at: Publisher Site | Google Scholar
C. Béroud, G. Collod-Béroud, C. Boileau, T. Soussi, and C. Junien, “UMD (Universal Mutation Database): a generic software to build and analyze locus-specific databases,” Human Mutation, vol. 15, no. 1, pp. 86–94, 2000.
View at: Publisher Site | Google Scholar
L. Détivaud, M.-L. Island, A.-M. Jouanolle et al., “Ferroportin diseases: functional studies, a link between genetic and clinical phenotype,” Human Mutation, vol. 34, no. 11, pp. 1529–1536, 2013.
View at: Publisher Site | Google Scholar
R. Praschberger, M. Schranz, W. J. H. Griffiths et al., “Impact of D181V and A69T on the function of ferroportin as an iron export pump and hepcidin receptor,” Biochimica et Biophysica Acta, vol. 1842, no. 9, pp. 1406–1412, 2014.
View at: Publisher Site | Google Scholar
P. C. J. L. Santos, R. D. Cançado, A. C. Pereira et al., “Hereditary hemochromatosis: mutations in genes involved in iron homeostasis in Brazilian patients,” Blood Cells, Molecules & Diseases, vol. 46, no. 4, pp. 302–307, 2011.
View at: Publisher Site | Google Scholar
R. L. Sham, P. D. Phatak, C. West, P. Lee, C. Andrews, and E. Beutler, “Autosomal dominant hereditary hemochromatosis associated with a novel ferroportin mutation and unique clinical features,” Blood Cells, Molecules & Diseases, vol. 34, no. 2, pp. 157–161, 2005.
View at: Publisher Site | Google Scholar
N. Yamakawa, K. Oe, N. Yukawa et al., “A novel phenotype of a hereditary hemochromatosis type 4 with ferroportin-1 mutation, presenting with juvenile cataracts,” Internal Medicine, vol. 55, no. 18, pp. 2697–2701, 2016.
View at: Publisher Site | Google Scholar
S. Richards, N. Aziz, S. Bale et al., “Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology,” Genetics in Medicine, vol. 17, no. 5, pp. 405–424, 2015.
View at: Publisher Site | Google Scholar
G. P. Jarvik and B. L. Browning, “Consideration of cosegregation in the pathogenicity classification of genomic variants,” American Journal of Human Genetics, vol. 98, no. 6, pp. 1077–1081, 2016.
View at: Publisher Site | Google Scholar
P. C. Adams and J. C. Barton, “A diagnostic approach to hyperferritinemia with a non-elevated transferrin saturation,” Journal of Hepatology, vol. 55, no. 2, pp. 453–458, 2011.
View at: Publisher Site | Google Scholar
K. Wong and P. C. Adams, “The diversity of liver diseases among outpatient referrals for elevated serum ferritin,” Canadian Journal of Gastroenterology, vol. 20, no. 7, pp. 467–470, 2006.
View at: Publisher Site | Google Scholar
A. Pietrangelo, E. Corradini, F. Ferrara et al., “Magnetic resonance imaging to identify classic and nonclassic forms of ferroportin disease,” Blood Cells, Molecules & Diseases, vol. 37, no. 3, pp. 192–196, 2006.
View at: Publisher Site | Google Scholar
A. Landemaine, H. Hamdi-Roze, S. Cunat et al., “A simple clinical score to promote and enhance ferroportin disease screening,” Journal of Hepatology, vol. 76, no. 3, pp. 568–576, 2022.
View at: Publisher Site | Google Scholar
M. Quinodoz, V. G. Peter, K. Cisarova et al., “Analysis of missense variants in the human genome reveals widespread gene- specific clustering and improves prediction of pathogenicity,” American Journal of Human Genetics, vol. 109, no. 3, pp. 457–470, 2022.
View at: Publisher Site | Google Scholar
L. Cemonesi, G. L. Forni, N. Soriani et al., “Genetic and clinical heterogeneity of ferroportin disease,” British Journal of Haematology, vol. 131, no. 5, pp. 663–670, 2005.
View at: Publisher Site | Google Scholar
S. Cunat, M. Giansily-Blaizot, M. Bismuth et al., “Global sequencing approach for characterizing the molecular background of hereditary iron disorders,” Clinical Chemistry, vol. 53, no. 12, pp. 2060–2069, 2007.
View at: Publisher Site | Google Scholar
L. Relvas, M. T. Claro, M. C. Bento, and M. L. Ribeiro, “Novel human pathological mutations. Gene symbol: SLC40A1. Disease: haemochromatosis, type 4,” Human Genetics, vol. 125, no. 3, p. 338, 2009.
View at: Google Scholar
A. Del-Castillo-Rueda, M.-I. Moreno-Carralero, N. Cuadrado-Grande et al., “Mutations in the HFE, TFR2, and SLC40A1 genes in patients with hemochromatosis,” Gene, vol. 508, no. 1, pp. 15–20, 2012.
View at: Publisher Site | Google Scholar
A. L. Lomize, S. C. Todd, and I. D. Pogozheva, “Spatial arrangement of proteins in planar and curved membranes by PPM 3.0,” Protein Science, vol. 31, no. 1, pp. 209–220, 2022.
View at: Publisher Site | Google Scholar

Copyright

Copyright © 2023 Kevin Uguen et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

PDF Download Citation

Download other formats

Order printed copies

Views

502

Downloads

273

Citations