Iron is an essential element for almost all organisms. In eukaryotes, it is mainly used in mitochondria for the biosynthesis of iron–sulfur clusters and haem group maturation. Iron is delivered into the mitochondrion by mitoferrins, members of the MCF (mitochondrial carrier family), through an unknown mechanism. In the present study, the yeast homologues of these proteins, Mrs3p (mitochondrial RNA splicing 3) and Mrs4p, were studied by inserting them into liposomes. In this context, they could transport Fe2+ across the proteoliposome membrane, as shown using the iron chelator bathophenanthroline. A series of amino acid-modifying reagents were screened for their effects on Mrs3p-mediated iron transport. The results of the present study suggest that carboxy and imidazole groups are essential for iron transport. This was confirmed by in vivo complementation assays, which demonstrated that three highly conserved histidine residues are important for Mrs3p function. These histidine residues are not conserved in other MCF members and thus they are likely to play a specific role in iron transport. A model describing how these residues help iron to transit smoothly across the carrier cavity is proposed and compared with the structural and biochemical data available for other carriers in this family.

INTRODUCTION

Iron is one of the most important metals required for the development of life in archaea, bacteria and eukaryotes. Owing to the electronic properties of iron, many proteins involved in several important biological functions, such as respiration, photosynthesis, nitrogen fixation or even oxygen transport, contain iron at their active site. Here, the metallic ion plays a role in electron transfer, as a Lewis acid, or has structural functions. Apart from in ferritin, transferrin, and non-haem mononuclear or binuclear iron-containing proteins, iron is present in two major metalloprotein classes differing on the basis of the structure of the iron cofactor. The first of these, iron-sulfur proteins, contain iron chelated by inorganic sulfur atoms or cysteine residues. These complexes can have various nuclearities, such as [2Fe–2S], [3Fe–4S] or [4Fe–4S] and are named iron–sulfur clusters [1]. The second class, haemoproteins, contain iron complexed at the centre of a porphyrin ring to form a haem group [2].

These inorganic cofactors are synthesized by specific protein machineries which are highly conserved throughout evolution. In eukaryotes, iron–sulfur proteins and haemoproteins are mainly biosynthesized in mitochondria. For example, iron–sulfur clusters are synthesized by members of the iron–sulfur cluster assembly machinery, which, for the most part, are mitochondrial proteins [3]. In addition, the last step in haem biosynthesis, when iron is inserted into protoporphyrin IX, is catalysed by ferrochelatase in the mitochondrial matrix [4]. To be available for these biosynthetic pathways, iron must be efficiently imported into the mitochondrion across both membranes to reach the matrix [5]. Porins in the outer mitochondrial membrane make it relatively porous to metabolites, but iron requires specific transport systems to cross the inner membrane.

Many specific carriers are found in the inner mitochondrial membrane, the largest group of these is the MCF (mitochondrial carrier family). This family is highly conserved in eukaryotes [6]. Members of the MCF exhibit a tripartite structure; each of the three parts organizes into two helices and consists of approximately 100 amino acids surrounding a conserved PX(D/E)X2(K/R) motif [7]. Thus members of the MCF have six transmembrane helices, H1–H6. The even- and odd-numbered helices are connected by two intermembrane space loops and by three matrix loops which include short D-helical stretches [8]. Members of the MCF transport various solutes between the intermembrane space and the matrix of mitochondria. Substrates of these carriers vary in size and structure, from protons to large molecules such as ATP and ADP. Yeast has been used to study different MCF members and roles have been established for many of them [9]. Saccharomyces cerevisiae possesses 35 members of the MCF, two of which, Mrs3p (mitochondrial RNA splicing 3) and Mrs4p, play a direct role in yeast mitochondrial iron homoeostasis under iron-limiting conditions [1014], by contributing to its transport across the inner membrane [15]. Genes encoding homologues of Mrs3p and Mrs4p have been identified in all higher eukaryotes, where they are generally known as mitoferrins. Their direct involvement in iron homoeostasis has been shown in zebrafish [16], Drosophila [17,18], Caenorhabditis [19] and murine fibroblasts [20]. Mitoferrins have also been linked to some human genetic diseases. For example, mitoferrin expression is up-regulated in yeast and mouse models of Friedreich's ataxia and in human cell culture models of Parkinson's disease. This expression pattern correlates with the mitochondrial iron accumulation and cytosolic iron deficiency observed in those diseases [2123].

Since the initial publications describing the roles of Mrs3p, Mrs4p and mitoferrins, most studies of these proteins have been based on genetic manipulation in yeast and higher eukaryotes. We found only one in vitro biochemical study in our literature search, where SMPs (submitochondrial particles) were used to assess iron transport across the inner membrane using a metal-sensitive fluorescent probe [15]. These authors found Mrs3p and Mrs4p to be responsible for most of the importing of Fe2+ into SMPs. This transport was driven by a high external Fe2+ concentration and was moderately stimulated by an acidic pH [15].

Despite these interesting results, the carrier itself and its intrinsic transport mechanism have not been studied further. Thus, although there is no doubt that mitoferrins transport iron, little is known about how the iron transits through the cavity of these carriers. To study this, we undertook to biochemically characterize both Mrs3p and Mrs4p to identify the residues involved in the iron transport mechanism. Using both biochemical and genetic approaches, we were able to show that three mitoferrin-specific conserved histidine residues are essential to iron translocation across the cavity of the carrier. On the basis of these results, we propose a model in which these three histidine residues could form a ladder of iron-binding ligands across the cavity.

MATERIALS AND METHODS

Chemicals and immunochemicals

Sarkosyl and Triton X-100 were purchased from Anatrace. Trypsin, BSA, N-ethylmaleimide, cyanate, 2-mercaptoethanol, mercuric acetate, iodine, carbodi-imide, EDC [1-ethyl-3-(3-dimethylaminopropyl)carbodi-imide], 2-hydroxy-5-nitrobenzyl bromide, Woodward's Reagent K (N-ethyl-5-phenylisoxazolium-3′-sulfonate), bathophenanthroline and methylmercuric iodide were all purchased from Sigma.

Bacterial and yeast strains

Two Escherichia coli strains were used in the present study, JM109 (Promega) and BL21 (DE3) C43 [24]. Bacteria were grown in LB medium (Difco) supplemented with 50 μg/ml kanamycin for selection as indicated. Bacteria were transformed using a standard CaCl2 method. W303 was used as the wild-type yeast strain. MRS3 and MRS4 were disrupted as described previously [11] to produce the W303 ∆mrs3mrs4 strain. Yeast strains were grown at 28°C or 37°C and transformed using the standard LiCl procedure. The compositions of all media used, YPL (lactate-containing rich medium) and YNB Glc W− (tryptophan-free minimal medium containing glucose; MP Biomedicals) have been described previously [25]. Iron-free media were prepared using iron-free YNB (ForMedium).

Cloning, site-directed mutagenesis and plasmid construction

S. cerevisiae genomic DNA was prepared from overnight cultures of the W303 yeast strain in YPD. The ORFs coding for Mrs3p (NCBI Reference Sequence NP_012402) and Mrs4p (NCBI Reference Sequence NP_012978) were PCR-amplified from this genomic DNA using VentR high fidelity polymerase (New England Biolabs) and specific primers (Table 1). For expression in bacteria, PCR fragments were subcloned into the pET28b vector (Novagen). For yeast expression, PCR fragments were subcloned into a centromeric plasmid derived from the plasmid pRS314. Wild-type and mutated genes were under the control of the MRS3 promoter and yeast ANC2 (ADP/ATP carrier isoform 2) terminator sequences. The resulting vector (pMD102) was used to transform the W303 ∆mrs3mrs4 strain. Site-directed mutagenesis of MRS3 was performed using the Transformer site-directed mutagenesis kit (Roche Applied Science) with the mutagenic primers listed in Table 1. The mutated MRS3 genes were subcloned into pMD102 and their capacity to complement ∆mrs3mrs4 deletion was assessed by plating on to solid YPL medium at 28°C and on solid iron-free YNB Glc W− medium at 37°C.

Table 1
Cloning and site directed mutagenesis strategy

pMD102 corresponds to the yeast expression vector pRS314 harbouring wild-type MRS3 or MRS4 genes or a mutated MRS3 genes under the control of the MRS3 promoter sequence and a yeast ANC2 terminator sequence. Restriction sites (a) and the modified codons (b) are underlined. Fw, forward; Rc, reverse complement; Rv, reverse.

(a) Genes and promoter cloning 
Gene (vector) Oligonucleotide Sequence (5′→3′) 
MRS3 (pET28b) MRS3-NcoI Fw AAGAAGCCATGGTAGAAAACTCGTCGA 
 MRS3-XhoI His-tagged Rv TGTAAGAACCTCGAGATACGTCATTAG 
 MRS3-XhoI Rv TGTAAGAACCTCGAGCTAATACGTCATTAG 
MRS4 (pET28b) MRS4-NcoI Fw CAGTTAATACCATGGATACTTCAGAAC 
 MRS4-XhoI His-tagged Rv GGGGAAAAACTCGAGATTTTTCATTAA 
 MRS4-XhoI Rv GGGGAAAAACTCGAGTCAATTTTTCATTAA 
MRS3 (pMD102) MRS3-EcoRI Fw TTTTTTGAATTCATGGTAGAAAACTCGTCGAG 
 MRS3-BamHI Rc TTTTTTGGATCCCTAATACGTCATTAGGAAATG 
MRS4 (pMD102) MRS4-EcoRI Fw TTTTTTGAATTCATGAATACTTCAGAACTGTC 
 MRS4-BamHI Rc TTTTTTGGATCCTCAATTTTTCATTAAAAAATCG 
MRS3 promoter (pMD102) MRS3-SalI Fw TTTTTTGTCGACATGTAGTGTAAATGTATCCG 
 MRS3-EcoRI Rc TTTTTTGAATTCAATATCTTCTTTTTTTATAAC 
(b) Mrs3p amino acid changes 
Mutation Oligonucleotide Sequence (5′→3′) 
H48A AHH-Fw CTGGTATAATGGAAGCTTCAGTGATGTTCCC 
 AHH-Rv GGGAACATCACTGAAGCTTCCATTATACCAG 
H105A HAH-Fw GTGCGGGACCTGCGGCTGCAGTGTATTTTGG 
 HAH-Rv CCAAAATACACTGCAGCCGCAGGTCCCGCAC 
H222A HHA-Fw CAACCCCCTCATAGCTTGTCTGTGTGGCAG 
 HHA-Rc CTGCCACACAGACAAGCTATGAGGGGGTTG 
(a) Genes and promoter cloning 
Gene (vector) Oligonucleotide Sequence (5′→3′) 
MRS3 (pET28b) MRS3-NcoI Fw AAGAAGCCATGGTAGAAAACTCGTCGA 
 MRS3-XhoI His-tagged Rv TGTAAGAACCTCGAGATACGTCATTAG 
 MRS3-XhoI Rv TGTAAGAACCTCGAGCTAATACGTCATTAG 
MRS4 (pET28b) MRS4-NcoI Fw CAGTTAATACCATGGATACTTCAGAAC 
 MRS4-XhoI His-tagged Rv GGGGAAAAACTCGAGATTTTTCATTAA 
 MRS4-XhoI Rv GGGGAAAAACTCGAGTCAATTTTTCATTAA 
MRS3 (pMD102) MRS3-EcoRI Fw TTTTTTGAATTCATGGTAGAAAACTCGTCGAG 
 MRS3-BamHI Rc TTTTTTGGATCCCTAATACGTCATTAGGAAATG 
MRS4 (pMD102) MRS4-EcoRI Fw TTTTTTGAATTCATGAATACTTCAGAACTGTC 
 MRS4-BamHI Rc TTTTTTGGATCCTCAATTTTTCATTAAAAAATCG 
MRS3 promoter (pMD102) MRS3-SalI Fw TTTTTTGTCGACATGTAGTGTAAATGTATCCG 
 MRS3-EcoRI Rc TTTTTTGAATTCAATATCTTCTTTTTTTATAAC 
(b) Mrs3p amino acid changes 
Mutation Oligonucleotide Sequence (5′→3′) 
H48A AHH-Fw CTGGTATAATGGAAGCTTCAGTGATGTTCCC 
 AHH-Rv GGGAACATCACTGAAGCTTCCATTATACCAG 
H105A HAH-Fw GTGCGGGACCTGCGGCTGCAGTGTATTTTGG 
 HAH-Rv CCAAAATACACTGCAGCCGCAGGTCCCGCAC 
H222A HHA-Fw CAACCCCCTCATAGCTTGTCTGTGTGGCAG 
 HHA-Rc CTGCCACACAGACAAGCTATGAGGGGGTTG 

Expression and purification of recombinant Mrs3p and Mrs4p

E. coli BL21 (DE3) C43 cells, transformed with the plasmids pET-MRS3 and pET-MRS4 or the cloning vector pET28b, were grown at 37°C in LB medium containing kanamycin (50 μg/ml), to a D600 value of approximately 0.7. Protein expression was then induced by adding IPTG (0.5 mM final concentration). The bacterial culture was incubated at 37°C for an additional 4 h before harvesting. Cells were pelleted and then washed with buffer [50 mM Tris/HCl (pH 8) and 200 mM NaCl], pelleted once again and frozen at −80°C until use. Proteins were harvested as follows. Cells obtained from a 1 litre culture were suspended in 15 ml of 10 mM Tris/HCl (pH 8.0) and disrupted using a French press. The non-soluble fraction was collected by centrifugation (5000 g for 30 min at 4°C). Pellets were resuspended in 2 ml of 10 mM Tris/HCl (pH 8.0) buffer, and inclusion bodies were purified on a discontinuous sucrose gradient as described in [26].

Proteoliposome reconstitution

To produce liposomes, 50 mg of asolectin (Woodside) was mixed with 1 ml of liposome buffer [50 mM Mes (pH 7.0) and 50 mM NaCl] and 100 mM bathophenanthroline was added. The mixture was sonicated on ice for 15 min at 10–12 W with a CV18 probe fitted on a Vibracell-72405 ultrasonicator (Bioblock Scientific). Purified inclusion bodies (40 μl) were solubilized with 1.8% (w/v) sarkozyl. The protein-containing solution was diluted 3-fold with water to reduce the sarkosyl concentration to 0.6%. The solution was centrifuged for 15 min (16000 g at 4°C) to remove the insoluble proteins. Then, 100 μl of sarkosyl-solubilized proteins (approximately 400 μg) were mixed with 400 μl of the liposome solution. After a 30 min incubation on ice, external bathophenanthroline and free detergent were removed from solutions of proteoliposomes by gel-filtration chromatography at 4°C using an Ultrogel AcA202 column (2.5 cm×5 cm) equilibrated in liposome buffer.

Iron transport assay

In a disposable 1 ml cuvette, 50–200 μl of proteoliposomes were mixed into 1 ml (final volume) liposome buffer to which FeSO4 was added extemporaneously (500 μM FeSO4 final concentration). After mixing rapidly, the kinetics of iron chelation by bathophenanthroline inside proteoliposomes was monitored at 535 nm (molar absorption coefficient 22350 M−1·cm−1) on a dual-beam spectrophotometer (Shimadzu UV-VIS 1605). Trypsin- (0.1 mg/ml) treated proteoliposomes were used as a negative control. For transport inhibition assays, proteoliposome solutions were pre-incubated with inhibitors (the concentrations tested are specified in the Results section) before assaying iron transport. The inhibitors were prepared as concentrated solutions in organic solvent or water and a small volume (a few microlitres) was added to the mixture of proteoliposomes. After a 30 min incubation on ice, inhibition reactions were quenched by diluting 1:5 with liposome buffer. Proteoliposomes were then desalted on a Zeba spin desalting column (Thermo Scientific) at 4°C to remove the excess reagent. Sample elution was monitored at 280 nm. Inhibitor-treated proteoliposomes were then used for the iron transport assay.

Isolation of mitochondria and preparation of mitochondrial membranes

Yeast cells grown on YPL medium were harvested during the late exponential phase (D600 near 5). Mitochondria were prepared as described previously [27] and resuspended in lysis buffer [10 mM Tris/HCl (pH 7.3), 0.1 mM Na2SO4 and 1 mM EDTA) containing 1 mM di-isopropyl fluorophosphate for lysis at a protein concentration of 10 mg/ml. The homogenate was centrifuged at 15000 g at 4°C for 30 min. The pelleted mitochondrial membranes were resuspended in Laemmli buffer for analysis by SDS/PAGE (12.5% gel) and Western blotting.

Mitochondrial iron content assay

Freshly prepared mitochondria (200 μl; 30 mg/ml total proteins) were washed in 0.6 M mannitol and 10 mM Tris/HCl (pH 7.4) and then lysed in 200 μl of NaOH (50 mM) for 30 min at room temperature (22–23°C). An aliquot (40 μg) of the lysate was transferred to a 1 ml tube and the complexed iron was released by adding acidic KMnO4 solution. The total iron content was determined by the ferrozine method [28] using a Shimadzu UV-VIS 1605 dual-beam spectrophotometer. For intersample comparison, the intramitochondrial iron content was normalized against the yeast ADP/ATP carrier Anc2p content, as described in [29].

SDS/PAGE, Western blotting and antibodies

Protein concentrations were determined using a BCA protein assay kit (Sigma) with BSA as a standard. For SDS/PAGE, samples were prepared as described in [27]. Antibodies directed against SDS-treated yeast Anc2p [25] were used at final dilutions of 1:1500 and 1:1000. Antibodies directed against His-tagged Mrs3p and Mrs4p were prepared from pure and GdmCl (guanidinium chloride)-treated carriers (Centre Lago). Briefly, inclusion bodies from a 1 litre culture of BL21 (DE3) C43 cells were solubilized in 3 ml of solubilization buffer [50 mM Tris/HCl (pH 8.0), 200 mM NaCl and 6 M GdmCl]. After centrifugation at 10000 g at 4°C for 30 min, His-tagged carriers were batch-purified by mixing with 1 ml of Ni-NTA (Ni2+-nitrilotriacetate) resin (Qiagen) for 1 h at 4°C. After extensive washing with solubilization buffer, proteins were eluted with 2 ml of the same buffer containing 0.5 mM imidazole. The imidazole-eluted carriers were then extensively dialysed against deionized water. This produced white precipitates which were then free-dried and used to immunize rabbits. Antibodies directed against GdmCl-treated Mrs3p or Mrs4p were both used at a 1:1500 dilution (Centre Lago). Antibody binding was revealed using horseradish peroxidase-coupled Protein A and an ECL system (Life Technologies).

RESULTS

Expression and solubilization of recombinant Mrs3p and Mrs4p proteins

Carriers (Mrs proteins) were expressed in E. coli Bl21 (DE3) C43 cells. After sequential centrifugation, both recombinant proteins were only present in lower-density pellets (Figure 1A, lanes 1 and 2) and not in the microsomal fractions containing the E. coli membranes (Figure 1A, lanes 3 and 4). Thus the proteins expressed are mainly found in inclusion bodies. None of our attempts to increase protein solubility in E. coli were successful; however, previous characterizations of different yeast members of the MCF indicated that recombinant proteins solubilized from inclusion bodies were active when reconstituted into liposomes [9]. Therefore we purified Mrs3p and Mrs4p inclusion bodies on a sucrose gradient and treated them with 1.8% (w/v) sarkosyl. Both recombinant proteins were efficiently solubilized (Figure 1B). These proteins were detected by their respective anti-(Mrs protein) serum at a molecular mass of approximately 33–35 kDa (Figure 1C, lanes 2 and 3), which corresponds to the expected size of the recombinant carriers (Figure 1B). Owing to the high degree of identity between the two yeast mitoferrins (76%), antibodies raised against Mrs3p also recognized Mrs4p (Figure 1C, lane 1) and vice versa (Figure 1C, lane 4). The identities of these proteins were confirmed by MS (results not shown).

Recombinant Mrs3p and Mrs4p expression, purification and immunolabelling

Figure 1
Recombinant Mrs3p and Mrs4p expression, purification and immunolabelling

(A) E. coli BL21 (DE3) C43 strain transformed with pET-MRS3 or pET-MRS4 was grown in LB medium, IPTG-induced for 2 h and disrupted. The crude fractions obtained were subjected to sequential 5000 g (5K, lanes 1 and 2) and 50000 g (50K, lanes 3 and 4) centrifugation and analysed by SDS/PAGE (12.5% gel) and Coomassie Blue staining. Approximately 10 μg of protein was loaded into each lane. Recombinant carriers are indicated by an arrow. (B) Mrs3p (lane 1) and Mrs4p (lane 2) inclusion bodies were purified on a discontinuous sucrose gradient, then solubilized with 1.8% (w/v) sarkosyl and analysed by SDS/PAGE with Coomassie Blue staining. Approximately 5 μg of proteins was loaded into each lane. Molecular masses are indicated in kDa. (C) Immunolabelling of the recombinant carriers with polyclonal antibodies raised against Mrs3p (lanes 1 and 2) and Mrs4p (lanes 3 and 4). Immune complexes were revealed using ECL.

Figure 1
Recombinant Mrs3p and Mrs4p expression, purification and immunolabelling

(A) E. coli BL21 (DE3) C43 strain transformed with pET-MRS3 or pET-MRS4 was grown in LB medium, IPTG-induced for 2 h and disrupted. The crude fractions obtained were subjected to sequential 5000 g (5K, lanes 1 and 2) and 50000 g (50K, lanes 3 and 4) centrifugation and analysed by SDS/PAGE (12.5% gel) and Coomassie Blue staining. Approximately 10 μg of protein was loaded into each lane. Recombinant carriers are indicated by an arrow. (B) Mrs3p (lane 1) and Mrs4p (lane 2) inclusion bodies were purified on a discontinuous sucrose gradient, then solubilized with 1.8% (w/v) sarkosyl and analysed by SDS/PAGE with Coomassie Blue staining. Approximately 5 μg of proteins was loaded into each lane. Molecular masses are indicated in kDa. (C) Immunolabelling of the recombinant carriers with polyclonal antibodies raised against Mrs3p (lanes 1 and 2) and Mrs4p (lanes 3 and 4). Immune complexes were revealed using ECL.

In vitro reconstitution of iron transport

Proteoliposomes containing the sarkosyl-solubilized carriers Mrs3p and Mrs4p were generated from fresh asolectin liposomes containing bathophenanthroline. Ferric iron is not soluble in water and therefore only the ferrous state is bioavailable. In biofluids, it is often chelated by small molecules. We tested the iron–citrate complex in our transport assay, but did not see any transport (results not shown). Thus it is tempting to consider that mitoferrins transport ferrous iron and not a complex of iron. Indeed, adding 500 μM FeSO4 to the proteoliposome solutions resulted in a kinetic increase in the D535 value due to the formation of the bathophenanthroline–iron complex. This complex formed in a Mrs3p- or Mrs4p-dependent manner (Figure 2A), resulting in pink-tinged solutions. Liposomes (without carriers) or proteoliposomes treated with trypsin did not show this significant increase in the D535 value, and these solutions only switched from clear to pink upon the addition of 0.1% Triton X-100 (Figure 2, insets). Proteoliposomes treated with trypsin were used as a control in all experiments and the resulting increase in D535 value was automatically subtracted from the samples kinetics shown (see the Material and methods section). Our results show unambiguously that the change in the D535 value corresponds specifically to the amount of iron uptake in the proteoliposomes encompassing Mrs3p or Mrs4p.

Recombinant Mrs3p and Mrs4p mediate iron transport across a liposomal membrane

Figure 2
Recombinant Mrs3p and Mrs4p mediate iron transport across a liposomal membrane

Mrs3p- and Mrs4p-containing proteoliposomes were generated from asolectin liposomes containing bathophenanthroline. FeSO4 (500 μM) was added to the proteoliposome solutions and resulted in a kinetic increase in the D535 value due to the formation of the bathophenanthroline–iron complex (μmol/g of protein incorporated in the proteoliposomes). Insets: kinetics of iron uptake in liposomes (1) or in proteoliposomes treated with trypsin (2) which were used as a negative control. (A) Iron transport kinetics at identical external and internal salt concentrations (50 mM). (B) Iron transport kinetics with external salt concentrations of 100 mM or 10 mM compared with an internal concentration of 50 mM.

Figure 2
Recombinant Mrs3p and Mrs4p mediate iron transport across a liposomal membrane

Mrs3p- and Mrs4p-containing proteoliposomes were generated from asolectin liposomes containing bathophenanthroline. FeSO4 (500 μM) was added to the proteoliposome solutions and resulted in a kinetic increase in the D535 value due to the formation of the bathophenanthroline–iron complex (μmol/g of protein incorporated in the proteoliposomes). Insets: kinetics of iron uptake in liposomes (1) or in proteoliposomes treated with trypsin (2) which were used as a negative control. (A) Iron transport kinetics at identical external and internal salt concentrations (50 mM). (B) Iron transport kinetics with external salt concentrations of 100 mM or 10 mM compared with an internal concentration of 50 mM.

In these first experiments, the inside and outside compartments of the proteoliposomes were similar, except that bathophenanthroline was present inside the proteoliposome and iron salts were added to the external buffer. No pH gradient was applied and the difference in Fe2+ concentration appears to be the main driving force for this transport, as reported previously [15]. To determine whether pH influences the Fe2+ flux across the membrane of proteoliposomes, experiments were performed where the external pH varied from 6.2 to 7.5. Over this pH range, no significant modifications to the kinetics of iron chelation by bathophenanthroline were observed (results not shown). Below pH 6, bathophenanthroline was progressively released from the proteoliposomes, probably due to a membrane fusion phenomenon. This phenomenon has been described to occur even under mildly acidic conditions [30]. Interestingly, when the salt concentration outside proteoliposomes was twice the internal concentration, iron chelation by bathophenanthroline was faster (Figure 2B). In contrast, when the external salt concentration was one-fifth the internal concentration, the kinetics of iron chelation were slower (Figure 2B). Similar results were obtained for a salt concentration of 100 mM inside the proteoliposomes and 20, 100 or 200 mM outside (results not shown). Thus it seems that the electrostatic pressure imposed on our system influences the initial rate of iron transport into proteoliposomes, but these results are difficult to interpret. The similar kinetic profiles obtained with Mrs3p and Mrs4p in all of these experiments suggested that they have comparable biochemical properties. Consequently, we focused only on Mrs3p in the remainder of the present study.

Effect of different amino acid-modifying reagents on Mrs3p-mediated iron transport in vitro

The main iron ligands in iron-binding proteins are histidine, cysteine, aspartate or glutamate, and tyrosine residues. To determine the identity of the residues directly involved in iron transport through Mrs3p, we screened a series of reagents reacting more or less specifically with particular amino acid functional groups, and assessed how much they interfere with Mrs3p-mediated iron transport. This approach requires the amino acids of interest to be accessible to the reagents. With 2-mercaptoethanol, methylmercuric iodide, N-ethylmaleimide, or cyanate (results not shown) no significant effect was noted, indicating that thiol or disulfide groups and lysine residues are not involved in iron transport, or are not accessible to the reagents. In contrast, Woodward's reagent K used at concentrations between 20 and 100 μM and EDC between 50 and 500 μM, were both potent inhibitors of Mrs3p activity (Supplementary Figure S1 at http://www.biochemj.org/bj/460/bj4600079add.htm). These two reagents react mainly with carboxy groups and, to a lesser extent, with the imidazole group of histidine residues. We therefore concluded that histidine, glutamate and/or aspartate residues may contribute to the iron transport mechanism of Mrs3p. DEPC (diethyl pyrocarbonate), iodine and mercuric acetate react mainly with histidine residues and, to a lesser extent, with tyrosine residues. All of these compounds significantly inhibited iron transport at concentrations between 5 and 50 μM, 25 and 100 μM and 0.5 and 10 μM respectively (Supplementary Figure S1). This further suggests a role for histidine residues, although tyrosine involvement cannot be excluded. N-bromosuccinimide reacts mainly with tryptophan and tyrosine and, to a lesser extent, with histidine residues. This molecule weakly inhibited iron transport by Mrs3p (Supplementary Figure S1). Since, no inhibition was found with 2-hydroxy-5-nitrobenzylbromide, which reacts preferentially with tryptophan residues (results not shown), we conclude that the weak effect of N-bromosuccinimide is probably due to interaction with histidine or tyrosine residues. Taken together, these results suggest that histidine and carboxylic residues are the most likely to be important in iron transport through Mrs3p.

Three histidine residues are conserved in mitoferrin sequences

Comparative analysis of primary amino acid sequences revealed that Mrs3p and Mrs4p share a higher percentage of histidine residues (2.9 and 2.6% respectively) than other yeast members of the MCF (an average of 1.3%). Mrs3p and Mrs4p have nine and eight histidine residues respectively in their primary sequence. Sequence comparisons of mitoferrins between the main biological models studied from yeast to human reveal three highly conserved histidine residues (Figure 3). The first two of these are strictly conserved in all of the mitoferrin amino acid sequences compared and are present in highly conserved regions of the cavity (Figure 3). The first is localized in helix H1, near the first MCF motif (PXDX2K), whereas the second is close to a highly conserved proline residue in helix H2, which is part of another signature sequence [PX2AX2X2YEX(2/3)K] found in all mitoferrin even-numbered helices (Figure 3, and Supplementary Figure S2 at http://www.biochemj.org/bj/460/bj4600079add.htm). The third histidine residue is conserved in 16 out of the 18 mitoferrin amino acid sequences analysed in the present study. In the other two sequences, it is replaced by a tyrosine residue, which is also a potential iron ligand. This third histidine/tyrosine residue is localized at the C-terminal extremity of helix H5, towards the intermembrane space. This region is less conserved than the two regions where the two other histidine residues are located. The three conserved histidines are His48, His105 and His222 in Mrs3p and His38, His95 and His212 in Mrs4p. Except for the histidine residue localized in helix H1, which is also found at the same position in both isoforms of the phosphate carrier [Pic1p (Pi carrier 1) and Pic2p], the other two are not conserved in any other members of the MCF identified in yeast (results not shown). Sequence comparisons between the yeast members of the MCF did not reveal any carboxylic residues characteristic of Mrs proteins (results not shown). Thus the strong conservation of these histidine residues combined with the results of our biochemical assays suggest that these residues could be involved in Fe2+ flux across the cavity of the mitoferrins. We therefore decided to further investigate their functional role.

Mitoferrin sequence alignments reveal conserved histidine residues

Figure 3
Mitoferrin sequence alignments reveal conserved histidine residues

Multiple sequence alignments for regions of the transmembrane helices H1, H2 and H5 from S. cerevisiae Mrs3p and Mrs4p with mitoferrins from Arabidopsis thaliana (AtMfrn1p and AtMfrn2p), Glycine max (Mfrnp), Aspergillus niger (AnMrsp), Homo sapiens (HsMfrn1p and HsMfrn2p), Mus musculus (MmMfrn1p and MmMfrn2p), Gallus gallus (GgMfrn1p and GgMfrn2p), Danio rerio (DrMfrn1p and DrMfrn2p), Caenorhabditis elegans (CeMfrnp), Drosophila melanogaster (DmMfrnp), Dictyostelium discoideum (DdMcfF) and Trypanosoma brucei (TbMCP17). Alignments were generated using ClustalW2 and analysed with Jalview v2.0.1 [43]. The canonical signature sequence [PXDX2K] of MCF members and the [PX2AX2X2YEX(2/3)K] signature identified in mitoferrin even-numbered helices are highlighted by a broken-line box in the alignments. *Highly conserved histidine residues which are boxed in the alignments.

Figure 3
Mitoferrin sequence alignments reveal conserved histidine residues

Multiple sequence alignments for regions of the transmembrane helices H1, H2 and H5 from S. cerevisiae Mrs3p and Mrs4p with mitoferrins from Arabidopsis thaliana (AtMfrn1p and AtMfrn2p), Glycine max (Mfrnp), Aspergillus niger (AnMrsp), Homo sapiens (HsMfrn1p and HsMfrn2p), Mus musculus (MmMfrn1p and MmMfrn2p), Gallus gallus (GgMfrn1p and GgMfrn2p), Danio rerio (DrMfrn1p and DrMfrn2p), Caenorhabditis elegans (CeMfrnp), Drosophila melanogaster (DmMfrnp), Dictyostelium discoideum (DdMcfF) and Trypanosoma brucei (TbMCP17). Alignments were generated using ClustalW2 and analysed with Jalview v2.0.1 [43]. The canonical signature sequence [PXDX2K] of MCF members and the [PX2AX2X2YEX(2/3)K] signature identified in mitoferrin even-numbered helices are highlighted by a broken-line box in the alignments. *Highly conserved histidine residues which are boxed in the alignments.

Role of conserved Mrs3p histidine residues in restoring the ∆mrs3mrs4 strain growth phenotype

Several groups have shown that deletion of both MRS3 and MRS4 results in iron-dependent phenotypes [1014]. We also found that the ∆mrs3mrs4 strain exhibits a growth defect in lactate-rich medium (YPL) at 28°C (Figure 4A). This phenotype was slightly exacerbated at 37°C and could be restored at both 28°C (Figure 4A) and at 37°C (result not shown) by adding iron to the culture medium. Thus deletion of MRS3 and MRS4 impairs yeast growth on a non-fermentable substrate in an iron-dependent manner, in agreement with the biochemical function of mitoferrins in mitochondria. When a centromeric plasmid expressing Mrs3p or Mrs4p was used to transform the ∆mrs3mrs4 strain, wild-type growth on YPL medium was restored (Figure 4B). Thus, as their similar biochemical properties would appear to suggest, Mrs3p can substitute for Mrs4p and vice versa. We next used Mrs3p to construct variant carriers where the three conserved histidine residues (His48, His105 and His222) were replaced by alanine residues. Mrs3p-[H48A] and Mrs3p-[H105A] were unable to restore the growth of ∆mrs3mrs4 in YPL medium at 28°C or 37°C, whereas Mrs3p-[H222A] complemented the phenotype at 28°C, but not 37°C (Figure 4B). In liquid YPL cultures at 28°C, the doubling time for the ∆mrs3mrs4 strain expressing Mrs3p was 2.5 h, with Mrs3p-[H222A] it was 3.5 h and with the empty vector it was 8 h (results not shown). This intermediate phenotype was lost at 37°C. As expected, the residual growth of Mrs3p-[H222A] was abolished when this mutation was combined with the H48A or H105A mutations (Figure 4B). Taken together, our results indicate that histidine residues 48, 105 and 222 in Mrs3p are important to its function.

Yeast strains harbouring His→Ala mutations grow poorly in iron-free media

Figure 4
Yeast strains harbouring His→Ala mutations grow poorly in iron-free media

(A) Strains W303 and W303 ∆mrs3mrs4 grown on different carbon sources. After an overnight culture at 28°C in synthetic iron-free glucose medium, the wild-type and ∆mrs3mrs4 strains were spotted (serial dilutions) on to solid glucose-rich medium or solid lactate-rich (YPL) medium supplemented with 100 μM FeSO4 where indicated. Plates were incubated at 28°C or 37°C for 3–4 days. (B) The strain ∆mrs3mrs4 transformed with empty pRS314 plasmid (top row) or pMD102 to express wild-type Mrs3p and Mrs4p (second and third rows) or Mrs3p forms in which conserved histidine residues have been replaced by alanine residues (fourth to eighth rows). The three positions were mutated in pairs (Mrs3p-[H48A,H222A], Mrs3p-[H105A,H222A]) or singly (Mrs3p-[H48A], Mrs3p-[H105A] or Mrs3p-[H222A]). Yeast transformants were cultured in complete tryptophan-free liquid minimal medium containing 2% glucose as carbon source (YNB Glc W−) and then transferred to iron-free YNB Glc W−medium. When cultures reached the exponential phase, cells were diluted and spotted on to YNB Glc W− or rich lactate-containing medium (YPL) plates. Plates were incubated for 3 days (YNB Glc W−) or for 3–4 days (YPL) at 28°C and 37°C.

Figure 4
Yeast strains harbouring His→Ala mutations grow poorly in iron-free media

(A) Strains W303 and W303 ∆mrs3mrs4 grown on different carbon sources. After an overnight culture at 28°C in synthetic iron-free glucose medium, the wild-type and ∆mrs3mrs4 strains were spotted (serial dilutions) on to solid glucose-rich medium or solid lactate-rich (YPL) medium supplemented with 100 μM FeSO4 where indicated. Plates were incubated at 28°C or 37°C for 3–4 days. (B) The strain ∆mrs3mrs4 transformed with empty pRS314 plasmid (top row) or pMD102 to express wild-type Mrs3p and Mrs4p (second and third rows) or Mrs3p forms in which conserved histidine residues have been replaced by alanine residues (fourth to eighth rows). The three positions were mutated in pairs (Mrs3p-[H48A,H222A], Mrs3p-[H105A,H222A]) or singly (Mrs3p-[H48A], Mrs3p-[H105A] or Mrs3p-[H222A]). Yeast transformants were cultured in complete tryptophan-free liquid minimal medium containing 2% glucose as carbon source (YNB Glc W−) and then transferred to iron-free YNB Glc W−medium. When cultures reached the exponential phase, cells were diluted and spotted on to YNB Glc W− or rich lactate-containing medium (YPL) plates. Plates were incubated for 3 days (YNB Glc W−) or for 3–4 days (YPL) at 28°C and 37°C.

Mutating histidine residues 48, 105 and 222 does not prevent Mrs3p translocation to mitochondria, but decreases iron import

The growth impairment observed on YPL in strains harbouring mutated Mrs3p forms with modified histidine residues could be related to lower Mrs3p levels in the mitochondrial inner membrane. To determine whether this was the case, we used Western blotting to evaluate the Mrs3p content in mitochondrial lysates. Mitochondrial protein loading was normalized based on levels of the yeast ADP/ATP carrier Anc2p, which is constitutively expressed when cells are grown in YPL medium and is independent of the MRS3 genetic background (Figure 5A). All strains were grown on YPL medium at 28°C or 37°C. Mrs3p was detectable in mitochondrial membranes from all transformants and similar overall amounts were present in wild-type and mutant strains (Figure 5A). Consequently, the growth decrease observed for the ∆mrs3mrs4 strain expressing the mutated Mrs3p on YPL medium at 28°C or at 37°C is due to impaired iron transport activity.

Biochemical characterization of variant Mrs3p

Figure 5
Biochemical characterization of variant Mrs3p

(A) Relative levels of mitochondrial wild-type and variant Mrs3p. Mitochondrial protein extracts (20 μg/lane) were prepared from control (lane 1) MRS3 (lane 2) and MRS3 variant strains (lanes 3–7). Strains were cultivated on non-fermentable carbon sources at 28°C (upper panels) or 37°C (lower panels). After SDS/PAGE (12.5% gel) and Western blotting, proteins were detected using polyclonal antibodies against Mrs3p (1:1500 final dilution). Yeast Anc2p, revealed using a polyclonal antibody against SDS-treated Anc2p (1:1500 final dilution), was used as a loading control. Immune complexes were detected with ECL. Identified polypeptides are indicated by arrows. (B) Iron content of fresh mitochondrial extracts. Mitochondria were lysed (200 μl; 30 mg/ml total protein) and aliquots of the lysates were transferred to 1 ml tubes where complexed iron was released by treatment with acidic KMnO4 solution. Total iron content was determined using the ferrozine method [28]. Results are means±S.D. for three representative experiments. The intramitochondrial iron content was normalized against protein levels based on the yeast ADP/ATP carrier Anc2p content as described in [29]. (C) Intraliposomal iron transport kinetics mediated by the recombinant carriers Mrs3p and Mrs3p-[H48A]. Iron transport assays were performed as described in the legend of Figure 2.

Figure 5
Biochemical characterization of variant Mrs3p

(A) Relative levels of mitochondrial wild-type and variant Mrs3p. Mitochondrial protein extracts (20 μg/lane) were prepared from control (lane 1) MRS3 (lane 2) and MRS3 variant strains (lanes 3–7). Strains were cultivated on non-fermentable carbon sources at 28°C (upper panels) or 37°C (lower panels). After SDS/PAGE (12.5% gel) and Western blotting, proteins were detected using polyclonal antibodies against Mrs3p (1:1500 final dilution). Yeast Anc2p, revealed using a polyclonal antibody against SDS-treated Anc2p (1:1500 final dilution), was used as a loading control. Immune complexes were detected with ECL. Identified polypeptides are indicated by arrows. (B) Iron content of fresh mitochondrial extracts. Mitochondria were lysed (200 μl; 30 mg/ml total protein) and aliquots of the lysates were transferred to 1 ml tubes where complexed iron was released by treatment with acidic KMnO4 solution. Total iron content was determined using the ferrozine method [28]. Results are means±S.D. for three representative experiments. The intramitochondrial iron content was normalized against protein levels based on the yeast ADP/ATP carrier Anc2p content as described in [29]. (C) Intraliposomal iron transport kinetics mediated by the recombinant carriers Mrs3p and Mrs3p-[H48A]. Iron transport assays were performed as described in the legend of Figure 2.

To confirm this, we used a colorimetric ferrozine-based assay [28] to quantify iron in mitochondrial extracts prepared from the ∆mrs3mrs4 strain transformed with plasmids expressing wild-type or mutated MRS3 genes. Protein content was normalized for yeast Anc2p by titration [29]. Cells containing Mrs3p contained 2–3-fold more mitochondrial iron than cells containing the empty vector (Figure 5B). At 28°C, Mrs3p-[H48A] or Mrs3p-[H105A] had no effect on the mitochondrial iron content, whereas Mrs3p-[H222A] expression resulted in an intermediate iron level. However, Mrs3p-[H222A] did not increase the mitochondrial iron content at 37°C (Figure 5B). Thus mutation of the conserved histidine residues clearly decreases mitochondrial iron content.

Finally, to directly demonstrate the importance of His48 in iron transport, we measured iron import by Mrs3p-[H48A] in an in vitro proteoliposome assay. Iron import into proteoliposomes was much slower with Mrs3p-[H48A] than with the wild-type carrier (Figure 5C). Although it is difficult to compare two proteoliposome solutions prepared independently from two different inclusion body fractions (wild-type or mutant), this result is consistent with strongly impaired iron transport in proteoliposomes containing Mrs3p-[H48A]. Taken together, all these data suggest that His48, His105 and, to a lesser extent, His222 are specifically required for Mrs3p-mediated mitochondrial iron transport.

DISCUSSION

In the present paper, we report the recombinant production of S. cerevisiae Mrs3p and Mrs4p in E. coli cells as inclusion bodies. Sarkosyl treatment made it possible to solubilize and isolate pure proteins, and a qualitative iron transport assay was developed using proteoliposomes. In this assay, iron transport was clearly dependent on Mrs3/4p, as pre-treatment of the reconstituted proteoliposomes with trypsin abolished iron transport. We used this assay to screen various reagents modifying specific amino acids to determine whether these modifications interfered with iron transport (Supplementary Figure S1). These experiments indicated the involvement of histidine residues in mitochondrial iron transport; carboxylic and tyrosine residues are also potential candidates. Interestingly, no inhibition was observed with reagents targeting thiol or disulfide groups despite the presence of conserved cysteine residues in Mrs3p and the fact that these groups are known to be important in a number of ways for the activity of many similar carriers. For example, the yeast phosphate carrier Pic1p, the yeast ADP/ATP carrier Anc2p and the mouse ornithine/citrulline carrier, all members of the MCF, are sensitive to several thiol-reactive compounds [3133].

Although either Mrs3p or Mrs4p in their wild-type form can complement the growth defect of the ∆mrs3mrs4 strain on non-fermentable medium, the histidine to alanine residue mutants at position 48, 105 (and to a lesser extent at position 222) clearly failed to complement the growth defect (Figure 4B). Mitochondrial protein levels were similar for these mutant proteins and wild-type protein (Figure 5A), but mitochondrial iron content was reduced with the mutant proteins (Figure 5B). This was further confirmed by using Mrs3p-[H48A] in our transport assay. The bathophenanthroline–iron complex formed more slowly when the mutant protein was present compared with the wild-type carrier (Figure 5C). Altogether, these data demonstrated that these histidine residues play a key role in mitochondrial iron transport mediated by Mrs3p and Mrs4p. His105/95 and His222/212 have no counterparts in any other members of the yeast MCF. In contrast, His48/38 is conserved in the yeast phosphate carriers Pic1p and Pic2p. Interestingly, Pic2p has been shown recently to be a mitochondrial copper transporter [34]. In the mitoferrin sequences, a fourth histidine residue is also conserved at 72%, mainly in higher eukaryotes. This residue is localized in helix H4, more precisely in the signature sequence [PX2AX2X2YEX(2/3)K] for even-numbered helices (Supplementary Figure S2). In yeast and fungi, this position is filled by an asparagine or glutamine residue respectively, whereas in the amoeba Dictyostelium discoideum and in the parasitic protist Trypanosoma brucei a tyrosine residue is present. Although absent from the yeast mitoferrins, this residue may also play a role in the transport mechanism of mitoferrins from other organisms.

In Nature, a number of residues are used by iron-containing proteins to ligate metal. Among these, histidine serves as a ligand in multiple classes of iron-containing proteins, including mononuclear and dinuclear iron enzymes, such as superoxide reductase or methane mono-oxygenase [35,36]. In the haemoprotein model myoglobin, outside the porphyrin ring the first axial ligand of iron is a histidine residue, whereas the second axial position binds oxygen. Even in the iron–sulfur class, which mainly relies on cysteine ligands, histidine can be present in the co-ordination sphere. This is the case for Rieske proteins where the [2Fe–2S] cluster has two cysteine and two histidine ligands. In rare cases, iron–sulfur clusters can be ligated by only one histidine and three cysteine residues. This pattern allows cluster transfer into an apoprotein, as observed with the mitoNEET protein (CDGSH iron-sulfur domain 1 protein) [37]. An obvious advantage of histidine over cysteine as a ligand is its potential for protonation. With a pKa slightly below the physiological pH (~6.8), histidine can easily vary from an unprotonated to a protonated state, thus modulating its affinity for the metal. Previous biochemical study of Mrs3p and Mrs4p showed the pH gradient to be somewhat implicated in iron transport [15]. This supports a role for histidine residues. However, in the present study the transport kinetics were not affected by pH. In contrast, there was a clear difference in the rate of iron transport due to changes in the electrostatic pressure across the proteoliposomal membrane (Figure 2). However, we have not found any explanation for this.

Using bovine and yeast ADP/ATP carriers as prototypical MCF members, we have shown that the closing and opening of the cavity on either side of the inner membrane could mimic the movement of a diaphragm [38]. The switch between the two different conformations in this model leads to substrate translocation between mitochondrial compartments [38]. In the absence of any structural data specific to Mrs3p and Mrs4p, we propose a model based on the structure of the bovine nucleotide carrier (Figure 6). In this structure, the three conserved histidines characterized in the present study [His48, His105 and His222 (Mrs3p numbering)] face the internal cavity of the carrier which is accessible from the intermembrane space (Figures 6A and 6B). The presumed spatial arrangement of these histidine residues, especially His48 and His105, is reminiscent of a spiral staircase. This arrangement could help the metal to glide down to the bottom of the cavity when the carrier closes on the intermembrane side, switching from one conformation to another (Figure 6C). A similar mechanism has been proposed for the yeast ADP/ATP carrier Anc2p where aromatic residues could help nucleotides to glide across the cavity [39]. According to our model, among the nine histidine residues in Mrs3p, six face the intermembrane space, only His80 faces the matrix, and His48 and His105 are inside the cavity. Even if they are not fully conserved, it is tempting to consider that His30, His35, His128, His129 and His309, situated in flexible loops toward the intermembrane space, could also play a role in iron capture and act together with or as a substitute for His222 (Figure 5). This could explain the growth phenotype observed at 28°C on YPL with Mrs3p-[H222A].

Hypothetical model of how histidine residues help Fe2+ to glide across the mitoferrin carrier cavity

Figure 6
Hypothetical model of how histidine residues help Fe2+ to glide across the mitoferrin carrier cavity

The three-dimensional model of Mrs3p was generated using the Protein Homology/analogY Recognition Engine (Phyre2) server in intensive modelling mode with PDB code 1OKC (bovine isoform I ADP/ATP carrier). The model presents a six-helix conformation which is open toward the intermembrane space and closed toward the matrix, as represented in grey by PyMOL v1.6. Highly conserved histidine residues present in mitoferrin sequences are represented as sticks, with carbon atoms in green and nitrogen atoms in blue. Non-conserved histidine residues are represented as sticks, with carbon atoms in grey and nitrogen atoms in blue. Iron is represented as a red sphere. (A) Upper view from the intermembrane space (ims). (B) Side view. Helix H4 was removed to allow better visualization of the conserved histidine residues. (C) Model showing diaphragm-like closing of the top of the cavity for MCF members (towards the intermembrane space) leading to substrate transition down the internal cavity, as described in [38]. The c-state corresponds to the carrier in the conformation open towards the intermembrane space and closed on the matrix side and vice versa for the m-state.

Figure 6
Hypothetical model of how histidine residues help Fe2+ to glide across the mitoferrin carrier cavity

The three-dimensional model of Mrs3p was generated using the Protein Homology/analogY Recognition Engine (Phyre2) server in intensive modelling mode with PDB code 1OKC (bovine isoform I ADP/ATP carrier). The model presents a six-helix conformation which is open toward the intermembrane space and closed toward the matrix, as represented in grey by PyMOL v1.6. Highly conserved histidine residues present in mitoferrin sequences are represented as sticks, with carbon atoms in green and nitrogen atoms in blue. Non-conserved histidine residues are represented as sticks, with carbon atoms in grey and nitrogen atoms in blue. Iron is represented as a red sphere. (A) Upper view from the intermembrane space (ims). (B) Side view. Helix H4 was removed to allow better visualization of the conserved histidine residues. (C) Model showing diaphragm-like closing of the top of the cavity for MCF members (towards the intermembrane space) leading to substrate transition down the internal cavity, as described in [38]. The c-state corresponds to the carrier in the conformation open towards the intermembrane space and closed on the matrix side and vice versa for the m-state.

Similar histidine residues have been shown to contribute to the function of the iron carrier DMT1 (divalent metal transporter 1). This 12-transmembrane-domain protein is a proton-coupled iron carrier involved in duodenal iron absorption and iron transport across the endosomal membrane. DEPC treatment of Cos-7 cell expressing DMT1 revealed the involvement of at least one histidine residue in iron transport, although it is more likely to play a role in regulating the pH as part of transport [40]. Until proven otherwise, this is also possible with the mitoferrins. In the case of mitoferrins, the histidine residues will not provide a full co-ordination sphere for Fe2+, and other amino acids are almost certainly involved in transporting iron across the cavity. Probable candidates are carboxylic residues which have been implicated in iron transport in the present study.

Mitoferrins are not the only mitochondrial metal transporters among the MCF members. To some extent, Rim2p (replication in mitochondria 2), the yeast mitochondrial pyrimidine nucleotide transporter, also transports divalent ions including iron [41,42]. More recently, Pic2p, which was initially described as a phosphate transporter, was shown to transport copper into the mitochondrion [34]. The biochemical approach used in the present study, based on recombinant carriers and proteoliposomes, could also be applied to these carriers to help elucidate their transport mechanism.

Two open questions remain from the present study. The first is what is the effective force driving iron transport by the mitoferrins? In vitro, the Fe2+ concentration gradient appears to be the main driving force, and transport is stimulated by a pH gradient and/or the resulting membrane potential [15] or electrostatic pressure (the present study). Since MCF members are classified as exchange carriers, this leads to the second open question, what is exchanged for the imported Fe2+?

Abbreviations

     
  • ANC2

    ADP/ATP carrier isoform 2

  •  
  • DEPC

    diethyl pyrocarbonate

  •  
  • DMT1

    divalent metal transporter 1

  •  
  • EDC

    1-ethyl-3-(3-dimethylaminopropyl)carbodi-imide

  •  
  • GdmCl

    guanidinium chloride

  •  
  • MCF

    mitochondrial carrier family

  •  
  • Mrs

    mitochondrial RNA splicing

  •  
  • Pic

    Pi carrier

  •  
  • SMP

    submitochondrial particle

AUTHOR CONTRIBUTION

Xavier Brazzolotto and Ludovic Pelosi designed the research. Xavier Brazzolotto, Fabien Pierrel and Ludovic Pelosi performed the experiments and analysed the data. Ludovic Pelosi wrote the paper.

We thank Dr Gérard Klein and Dr Gérard Brandolin for constructive discussions and technical assistance and Dr Françoise Foury for the gift of the W303 Δmrs3Δmrs4 strain. We thank Professor Marc Fontecave for critically reading the paper and Dr Maighread Gallagher-Gambarelli for editorial and language usage suggestions before submission.

FUNDING

This work was supported by the Université Joseph Fourier, the French Centre National de la Recherche Scientifique (CNRS) and the Commissariat à l’Energie Atomique et aux Energies Alternatives. Dr Xavier Brazzolotto was supported by a fellowship from the French Centre National de la Recherche Scientifique.

References

References
1
Lill
R.
Function and biogenesis of iron–sulphur proteins
Nature
2009
, vol. 
460
 (pg. 
831
-
838
)
[PubMed]
2
Ajioka
R. S.
Phillips
J. D.
Kushner
J. P.
Biosynthesis of heme in mammals
Biochim. Biophys. Acta
2006
, vol. 
1763
 (pg. 
723
-
736
)
[PubMed]
3
Lill
R.
Hoffmann
B.
Molik
S.
Pierik
A. J.
Rietzschel
N.
Stehling
O.
Uzarska
M. A.
Webert
H.
Wilbrecht
C.
Muhlenhoff
U.
The role of mitochondria in cellular iron–sulfur protein biogenesis and iron metabolism
Biochim. Biophys. Acta
2012
, vol. 
1823
 (pg. 
1491
-
1508
)
[PubMed]
4
Hamza
I.
Dailey
H. A.
One ring to rule them all: trafficking of heme and heme synthesis intermediates in the metazoans
Biochim. Biophys. Acta
2012
, vol. 
1823
 (pg. 
1617
-
1632
)
[PubMed]
5
Richardson
D. R.
Lane
D. J.
Becker
E. M.
Huang
M. L.
Whitnall
M.
Suryo Rahmanto
Y.
Sheftel
A. D.
Ponka
P.
Mitochondrial iron trafficking and the integration of iron metabolism between the mitochondrion and cytosol
Proc. Natl. Acad. Sci. U.S.A.
2010
, vol. 
107
 (pg. 
10775
-
10782
)
[PubMed]
6
Palmieri
F.
Pierri
C. L.
De Grassi
A.
Nunes-Nesi
A.
Fernie
A. R.
Evolution, structure and function of mitochondrial carriers: a review with new insights
Plant J.
2013
, vol. 
66
 (pg. 
161
-
181
)
7
Walker
J. E.
Runswick
M. J.
The mitochondrial transport protein superfamily
J. Bioenerg. Biomembr.
1993
, vol. 
25
 (pg. 
435
-
446
)
[PubMed]
8
Pebay-Peyroula
E.
Dahout-Gonzalez
C.
Kahn
R.
Trézéguet
V.
Lauquin
G. J.
Brandolin
G.
Structure of mitochondrial ADP/ATP carrier in complex with carboxyatractyloside
Nature
2003
, vol. 
426
 (pg. 
39
-
44
)
[PubMed]
9
Palmieri
F.
Agrimi
G.
Blanco
E.
Castegna
A.
Di Noia
M. A.
Iacobazzi
V.
Lasorsa
F. M.
Marobbio
C. M.
Palmieri
L.
Scarcia
P.
, et al. 
Identification of mitochondrial carriers in Saccharomyces cerevisiae by transport assay of reconstituted recombinant proteins
Biochim. Biophys. Acta
2006
, vol. 
1757
 (pg. 
1249
-
1262
)
[PubMed]
10
Foury
F.
Cazzalini
O.
Deletion of the yeast homologue of the human gene associated with Friedreich's ataxia elicits iron accumulation in mitochondria
FEBS Lett.
1997
, vol. 
411
 (pg. 
373
-
377
)
[PubMed]
11
Foury
F.
Roganti
T.
Deletion of the mitochondrial carrier genes MRS3 and MRS4 suppresses mitochondrial iron accumulation in a yeast frataxin-deficient strain
J. Biol. Chem.
2002
, vol. 
277
 (pg. 
24475
-
24483
)
[PubMed]
12
Muhlenhoff
U.
Stadler
J. A.
Richhardt
N.
Seubert
A.
Eickhorst
T.
Schweyen
R. J.
Lill
R.
Wiesenberger
G.
A specific role of the yeast mitochondrial carriers MRS3/4p in mitochondrial iron acquisition under iron-limiting conditions
J. Biol. Chem.
2003
, vol. 
278
 (pg. 
40612
-
40620
)
[PubMed]
13
Zhang
Y.
Lyver
E. R.
Knight
S. A.
Lesuisse
E.
Dancis
A.
Frataxin and mitochondrial carrier proteins, Mrs3p and Mrs4p, cooperate in providing iron for heme synthesis
J. Biol. Chem.
2005
, vol. 
280
 (pg. 
19794
-
19807
)
[PubMed]
14
Zhang
Y.
Lyver
E. R.
Knight
S. A.
Pain
D.
Lesuisse
E.
Dancis
A.
Mrs3p, Mrs4p, and frataxin provide iron for Fe–S cluster synthesis in mitochondria
J. Biol. Chem.
2006
, vol. 
281
 (pg. 
22493
-
22502
)
[PubMed]
15
Froschauer
E. M.
Schweyen
R. J.
Wiesenberger
G.
The yeast mitochondrial carrier proteins Mrs3p/Mrs4p mediate iron transport across the inner mitochondrial membrane
Biochim. Biophys. Acta
2009
, vol. 
1788
 (pg. 
1044
-
1050
)
[PubMed]
16
Shaw
G. C.
Cope
J. J.
Li
L.
Corson
K.
Hersey
C.
Ackermann
G. E.
Gwynn
B.
Lambert
A. J.
Wingert
R. A.
Traver
D.
, et al. 
Mitoferrin is essential for erythroid iron assimilation
Nature
2006
, vol. 
440
 (pg. 
96
-
100
)
[PubMed]
17
Metzendorf
C.
Wu
W.
Lind
M. I.
Overexpression of Drosophila mitoferrin in l(2)mbn cells results in dysregulation of Fer1HCH expression
Biochem. J.
2009
, vol. 
421
 (pg. 
463
-
471
)
[PubMed]
18
Metzendorf
C.
Lind
M. I.
Drosophila mitoferrin is essential for male fertility: evidence for a role of mitochondrial iron metabolism during spermatogenesis
BMC Dev. Biol.
2010
, vol. 
10
 pg. 
68
 
[PubMed]
19
Ren
Y.
Yang
S.
Tan
G.
Ye
W.
Liu
D.
Qian
X.
Ding
Z.
Zhong
Y.
Zhang
J.
Jiang
D.
, et al. 
Reduction of mitoferrin results in abnormal development and extended lifespan in Caenorhabditis elegans
PLoS ONE
2012
, vol. 
7
 pg. 
e29666
 
[PubMed]
20
Paradkar
P. N.
Zumbrennen
K. B.
Paw
B. H.
Ward
D. M.
Kaplan
J.
Regulation of mitochondrial iron import through differential turnover of mitoferrin 1 and mitoferrin 2
Mol. Cell. Biol.
2009
, vol. 
29
 (pg. 
1007
-
1016
)
[PubMed]
21
Huang
M. L.
Becker
E. M.
Whitnall
M.
Suryo Rahmanto
Y.
Ponka
P.
Richardson
D. R.
Elucidation of the mechanism of mitochondrial iron loading in Friedreich's ataxia by analysis of a mouse mutant
Proc. Natl. Acad. Sci. U.S.A.
2009
, vol. 
106
 (pg. 
16381
-
16386
)
[PubMed]
22
Richardson
D. R.
Huang
M. L.
Whitnall
M.
Becker
E. M.
Ponka
P.
Suryo Rahmanto
Y.
The ins and outs of mitochondrial iron-loading: the metabolic defect in Friedreich's ataxia
J. Mol. Med.
2010
, vol. 
88
 (pg. 
323
-
329
)
[PubMed]
23
Carroll
C. B.
Zeissler
M. L.
Chadborn
N.
Gibson
K.
Williams
G.
Zajicek
J. P.
Morrison
K. E.
Hanemann
C. O.
Changes in iron-regulatory gene expression occur in human cell culture models of Parkinson's disease
Neurochem. Int.
2011
, vol. 
59
 (pg. 
73
-
80
)
[PubMed]
24
Miroux
B.
Walker
J. E.
Over-production of proteins in Escherichia coli: mutant hosts that allow synthesis of some membrane proteins and globular proteins at high levels
J. Mol. Biol.
1996
, vol. 
260
 (pg. 
289
-
298
)
[PubMed]
25
Clemencon
B.
Rey
M.
Dianoux
A. C.
Trezeguet
V.
Lauquin
G. J.
Brandolin
G.
Pelosi
L.
Structure–function relationships of the C-terminal end of the Saccharomyces cerevisiae ADP/ATP carrier isoform 2
J. Biol. Chem.
2008
, vol. 
283
 (pg. 
11218
-
11225
)
[PubMed]
26
Palmieri
F.
Indiveri
C.
Bisaccia
F.
Iacobazzi
V.
Mitochondrial metabolite carrier proteins: purification, reconstitution, and transport studies
Methods Enzymol.
1995
, vol. 
260
 (pg. 
349
-
369
)
[PubMed]
27
Dahout-Gonzalez
C.
Ramus
C.
Dassa
E. P.
Dianoux
A. C.
Brandolin
G.
Conformation-dependent swinging of the matrix loop m2 of the mitochondrial Saccharomyces cerevisiae ADP/ATP carrier
Biochemistry
2005
, vol. 
44
 (pg. 
16310
-
16320
)
[PubMed]
28
Riemer
J.
Hoepken
H. H.
Czerwinska
H.
Robinson
S. R.
Dringen
R.
Colorimetric ferrozine-based assay for the quantitation of iron in cultured cells
Anal. Biochem.
2004
, vol. 
331
 (pg. 
370
-
375
)
[PubMed]
29
Pena-Diaz
P.
Pelosi
L.
Ebikeme
C.
Colasante
C.
Gao
F.
Bringaud
F.
Voncken
F.
Functional characterization of TbMCP5, a conserved and essential ADP/ATP carrier present in the mitochondrion of the human pathogen Trypanosoma brucei
J. Biol. Chem.
2012
, vol. 
287
 (pg. 
41861
-
41874
)
[PubMed]
30
Connor
J.
Yatvin
M. B.
Huang
L.
pH-sensitive liposomes: acid-induced liposome fusion
Proc. Natl. Acad. Sci. U.S.A.
1984
, vol. 
81
 (pg. 
1715
-
1718
)
[PubMed]
31
Majima
E.
Koike
H.
Hong
Y. M.
Shinohara
Y.
Terada
H.
Characterization of cysteine residues of mitochondrial ADP/ATP carrier with the SH-reagents eosin 5-maleimide and N-ethylmaleimide
J. Biol. Chem.
1993
, vol. 
268
 (pg. 
22181
-
22187
)
[PubMed]
32
Schroers
A.
Kramer
R.
Wohlrab
H.
The reversible antiport-uniport conversion of the phosphate carrier from yeast mitochondria depends on the presence of a single cysteine
J. Biol. Chem.
1997
, vol. 
272
 (pg. 
10558
-
10564
)
[PubMed]
33
Tonazzi
A.
Giangregorio
N.
Palmieri
F.
Indiveri
C.
Relationships of cysteine and lysine residues with the substrate binding site of the mitochondrial ornithine/citrulline carrier: an inhibition kinetic approach combined with the analysis of the homology structural model
Biochim. Biophys. Acta
2005
, vol. 
1718
 (pg. 
53
-
60
)
[PubMed]
34
Vest
K. E.
Leary
S. C.
Winge
D. R.
Cobine
P. A.
Copper import into the mitochondrial matrix in Saccharomyces cerevisiae is mediated by Pic2, a mitochondrial carrier family protein
J. Biol. Chem.
2013
, vol. 
288
 (pg. 
23884
-
23892
)
[PubMed]
35
Niviere
V.
Fontecave
M.
Discovery of superoxide reductase: an historical perspective
J. Biol. Inorg. Chem.
2004
, vol. 
9
 (pg. 
119
-
123
)
[PubMed]
36
Hakemian
A. S.
Rosenzweig
A. C.
The biochemistry of methane oxidation
Annu. Rev. Biochem.
2007
, vol. 
76
 (pg. 
223
-
241
)
[PubMed]
37
Zuris
J. A.
Harir
Y.
Conlan
A. R.
Shvartsman
M.
Michaeli
D.
Tamir
S.
Paddock
M. L.
Onuchic
J. N.
Mittler
R.
Cabantchik
Z. I.
, et al. 
Facile transfer of [2Fe–2S] clusters from the diabetes drug target mitoNEET to an apo-acceptor protein
Proc. Natl. Acad. Sci. U.S.A.
2011
, vol. 
108
 (pg. 
13047
-
13052
)
[PubMed]
38
Rey
M.
Man
P.
Clémencon
B.
Trézéguet
V.
Brandolin
G.
Forest
E.
Pelosi
L.
Conformational dynamics of the bovine mitochondrial ADP/ATP carrier isoform 1 revealed by hydrogen/deuterium exchange coupled to mass spectrometry
J. Biol. Chem.
2010
, vol. 
285
 (pg. 
34981
-
34990
)
[PubMed]
39
David
C.
Arnou
B.
Sanchez
J. F.
Pelosi
L.
Brandolin
G.
Lauquin
G. J.
Trézéguet
V.
Two residues of a conserved aromatic ladder of the mitochondrial ADP/ATP carrier are crucial to nucleotide transport
Biochemistry
2008
, vol. 
47
 (pg. 
13223
-
13231
)
[PubMed]
40
Shawki
A.
Knight
P. B.
Maliken
B. D.
Niespodzany
E. J.
Mackenzie
B.
H+-coupled divalent metal-ion transporter-1: functional properties, physiological roles and therapeutics
Curr. Top. Membr.
2012
, vol. 
70
 (pg. 
169
-
214
)
[PubMed]
41
Yoon
H.
Zhang
Y.
Pain
J.
Lyver
E. R.
Lesuisse
E.
Pain
D.
Dancis
A.
Rim2, a pyrimidine nucleotide exchanger, is needed for iron utilization in mitochondria
Biochem. J.
2011
, vol. 
440
 (pg. 
137
-
146
)
[PubMed]
42
Froschauer
E. M.
Rietzschel
N.
Hassler
M. R.
Binder
M.
Schweyen
R. J.
Lill
R.
Muhlenhoff
U.
Wiesenberger
G.
The mitochondrial carrier Rim2 co-imports pyrimidine nucleotides and iron
Biochem. J.
2013
, vol. 
455
 (pg. 
57
-
65
)
[PubMed]
43
Waterhouse
A. M.
Procter
J. B.
Martin
D. M.
Clamp
M.
Barton
G. J.
Jalview Version 2: a multiple sequence alignment editor and analysis workbench
Bioinformatics
2009
, vol. 
25
 (pg. 
1189
-
1191
)
[PubMed]

Supplementary data