Deletion of the Mitochondrial Carrier Genes MRS3 andMRS4 Suppresses Mitochondrial Iron Accumulation in a Yeast Frataxin-deficient Strain
2002; Elsevier BV; Volume: 277; Issue: 27 Linguagem: Inglês
10.1074/jbc.m111789200
ISSN1083-351X
AutoresFrançoise Foury, Tiziana Roganti,
Tópico(s)RNA Research and Splicing
ResumoThe mitochondrial solute carriers Mrs3p and Mrs4p were originally isolated as multicopy suppressors of intron splicing defects. We show here that MRS4 is co-regulated with the iron regulon genes, and up-regulated in a strain deficient for Yfh1p, the yeast homologue of human frataxin. Using in vivo 55Fe cell radiolabeling we show that in glucose-grown cells mitochondrial iron accumulation is 5–15 times higher in ΔYFH1 than in wild-type strain. However, although in a ΔYFH1ΔMRS3ΔMRS4 strain, the intracellular 55Fe content is extremely high, the mitochondrial iron concentration is decreased to almost wild-type levels. Moreover, ΔYFH1ΔMRS3ΔMRS4 cells grown in high iron media do not lose their mitochondrial genome. Conversely, a ΔYFH1 strain overexpressing MRS4 has an increased mitochondrial iron content and no mitochondrial genome. Therefore, MRS4 is required for mitochondrial iron accumulation in ΔYFH1 cells. Expression of the iron regulon and intracellular 55Fe content are higher in a ΔMRS3ΔMRS4 strain than in the wild type. Nevertheless, the mitochondrial 55Fe content, a balance between iron uptake and exit, is decreased by a factor of two. Moreover, 55Fe incorporation into heme by ferrochelatase is increased in an MRS4-overexpressing strain. The function ofMRS4 in iron import into mitochondria is discussed. The mitochondrial solute carriers Mrs3p and Mrs4p were originally isolated as multicopy suppressors of intron splicing defects. We show here that MRS4 is co-regulated with the iron regulon genes, and up-regulated in a strain deficient for Yfh1p, the yeast homologue of human frataxin. Using in vivo 55Fe cell radiolabeling we show that in glucose-grown cells mitochondrial iron accumulation is 5–15 times higher in ΔYFH1 than in wild-type strain. However, although in a ΔYFH1ΔMRS3ΔMRS4 strain, the intracellular 55Fe content is extremely high, the mitochondrial iron concentration is decreased to almost wild-type levels. Moreover, ΔYFH1ΔMRS3ΔMRS4 cells grown in high iron media do not lose their mitochondrial genome. Conversely, a ΔYFH1 strain overexpressing MRS4 has an increased mitochondrial iron content and no mitochondrial genome. Therefore, MRS4 is required for mitochondrial iron accumulation in ΔYFH1 cells. Expression of the iron regulon and intracellular 55Fe content are higher in a ΔMRS3ΔMRS4 strain than in the wild type. Nevertheless, the mitochondrial 55Fe content, a balance between iron uptake and exit, is decreased by a factor of two. Moreover, 55Fe incorporation into heme by ferrochelatase is increased in an MRS4-overexpressing strain. The function ofMRS4 in iron import into mitochondria is discussed. 4′,6′-diamino-2-phenylindole Mitochondria utilize most of the cellular iron. First, the mitochondrial ferrochelatase catalyzes the insertion of ferrous iron into protoporphyrin IX, the heme precursor of cytochromes. Second, the mitochondrial matrix and respiratory chain contain several iron-sulfur proteins. Moreover, it has recently been discovered that iron-sulfur clusters are synthesized inside mitochondria by a specific machinery involving more than 10 proteins that have orthologues in bacteria (1Muhlenhoff U. Lill R. Biochim. Biophys. Acta. 2000; 1459: 370-382Crossref PubMed Scopus (181) Google Scholar,2Craig E.A. Voisine C. Schilke B. Biol. Chem. 1999; 380: 1167-1173Crossref PubMed Scopus (42) Google Scholar). These iron-sulfur clusters are used for both mitochondrial and cytosolic proteins, and their export into the cytosol is probably mediated by the ABC transporter Atm1p (3Kispal G. Csere P. Prohl C. Lill R. EMBO J. 1999; 18: 3981-3989Crossref PubMed Scopus (585) Google Scholar). Based on their capacity to restore the low iron growth defect of an erg25 mutant (4Li L. Kaplan J. J. Biol. Chem. 1997; 272: 28485-28493Abstract Full Text Full Text PDF PubMed Scopus (72) Google Scholar), two homologous transporters belonging to the cation efflux transporter family (5Paulsen I.T. Sliwinski M.K. Nelissen B. Goffeau A. Saier M.H., Jr. FEBS Lett. 1998; 430: 116-125Crossref PubMed Scopus (196) Google Scholar), Mmt1p and Mmt2p, have been reported to play a role in mitochondrial iron transport (4Li L. Kaplan J. J. Biol. Chem. 1997; 272: 28485-28493Abstract Full Text Full Text PDF PubMed Scopus (72) Google Scholar). However, the double deletion strain does not exhibit any respiratory defect, suggesting that the role played by MMT1 and MMT2 in mitochondrial iron metabolism is modest (4Li L. Kaplan J. J. Biol. Chem. 1997; 272: 28485-28493Abstract Full Text Full Text PDF PubMed Scopus (72) Google Scholar, 6Lange H. Kispal G. Lill R. J. Biol. Chem. 1999; 274: 18989-18996Abstract Full Text Full Text PDF PubMed Scopus (133) Google Scholar). Using DNA microarrays, we have shown that mitochondria play a key role in cellular iron homeostasis (7Foury F. Talibi D. J. Biol. Chem. 2001; 276: 7762-7768Abstract Full Text Full Text PDF PubMed Scopus (161) Google Scholar). This conclusion was reached by using a strain that has been deleted for YFH1, the yeast frataxin gene (8Babcock M. de Silva D. Oaks R. Davis-Kaplan S. Jiralerspong S. Montermini L. Pandolfo M. Kaplan J. Science. 1997; 276: 1709-1712Crossref PubMed Scopus (817) Google Scholar, 9Foury F. Cazzalini O. FEBS Lett. 1997; 411: 373-377Crossref PubMed Scopus (342) Google Scholar). Frataxin is a small hydrophilic protein of unknown function conserved in the mitochondria of all eukaryotes. It has been reported that a multimeric form of frataxin is involved in iron sequestration (10Adamec J. Rusnak F. Owen W.G. Naylor S. Benson L.M. Gacy A.M. Isaya G. Am. J. Hum. Genet. 2000; 67: 549-562Abstract Full Text Full Text PDF PubMed Scopus (229) Google Scholar). In glucose-grown cells YFH1 deletion elicits mitochondrial iron overload (8Babcock M. de Silva D. Oaks R. Davis-Kaplan S. Jiralerspong S. Montermini L. Pandolfo M. Kaplan J. Science. 1997; 276: 1709-1712Crossref PubMed Scopus (817) Google Scholar, 9Foury F. Cazzalini O. FEBS Lett. 1997; 411: 373-377Crossref PubMed Scopus (342) Google Scholar) and prevents iron export from mitochondria (11Radisky D.C. Babcock M.C. Kaplan J. J. Biol. Chem. 1999; 274: 4497-4499Abstract Full Text Full Text PDF PubMed Scopus (248) Google Scholar). This phenotype is associated with increased expression of the genes involved in iron mobilization in anAFT1-dependent manner (7Foury F. Talibi D. J. Biol. Chem. 2001; 276: 7762-7768Abstract Full Text Full Text PDF PubMed Scopus (161) Google Scholar). Aft1p is an iron-sensing transcription factor that activates the transcription of a set of genes, the "iron regulon," under cellular iron starvation conditions (12Yamaguchi-Iwai Y. Dancis A. Klausner R.D. EMBO J. 1995; 14: 1231-1239Crossref PubMed Scopus (313) Google Scholar). By analyzing published data of gene expression profiles obtained under 300 different conditions (13Hughes T.R. Marton M.J. Jones A.R. Roberts C.J. Stoughton R. Armour C.D. Bennett HA Coffey E. Dai H., He, Y.D. Kidd M.J. King A.M. Meyer M.R. Slade D. Lum P.Y. Stepaniants S.B. Shoemaker D.D. Gachotte D. Chakraburtty K. Simon J. Bard M. Friend S.H. Cell. 2000; 102: 109-126Abstract Full Text Full Text PDF PubMed Scopus (2090) Google Scholar) we found that theMRS4 and MMT2 genes are co-regulated with severalAFT1-dependent genes. These data suggest thatMRS4 and MMT2 play a role in iron metabolism. The homologous Mrs3p and Mrs4p proteins are located in the mitochondrial inner membrane and belong to the mitochondrial solute carrier family (14Wiesenberger G. Link T.A. von Ahsen U. Waldherr M. Schweyen R.J. J. Mol. Biol. 1991; 217: 23-37Crossref PubMed Scopus (78) Google Scholar). Saccharomyces cerevisiae contains 35 of these carriers (15el Moualij B. Duyckaerts C. Lamotte-Brasseur J. Sluse F.E. Yeast. 1997; 13: 573-581Crossref PubMed Scopus (93) Google Scholar, 16Belenkiy R. Haefele A. Eisen M.B. Wohlrab H. Biochim. Biophys. Acta. 2000; 1467: 207-218Crossref PubMed Scopus (43) Google Scholar), the most studied of which is the mitochondrial ATP/ADP translocator. Mitochondrial carriers are characterized by three homologous domains, each containing two transmembrane spans.MRS3 and MRS4 were initially isolated as high copy number suppressors of mitochondrial mRNA splicing defects, particularly in mrs2 mutants (17Waldherr M. Ragnini A. Jank B. Teply R. Wiesenberger G. Schweyen R.J. Curr. Genet. 1993; 24: 301-306Crossref PubMed Scopus (60) Google Scholar). MRS2 encodes a mitochondrial inner membrane protein belonging to the bacterial CORA family of magnesium transporters and is involved in magnesium homeostasis (18Bui D.M. Gregan J. Jarosch E. Ragnini A. Schweyen R.J.J. Biol. Chem. 1999; 274: 20438-20443Abstract Full Text Full Text PDF Scopus (142) Google Scholar). Mrs3p and Mrs4p carriers have been conserved during evolution, and their human homologue can rescue the thermosensitive growth defect observed in a double ΔMRS3ΔMRS4 deletion yeast strain (19Li F.Y. Nikali K. Gregan J. Leibiger I. Leibiger B. Schweyen R. Larsson C. Suomalainen A. FEBS Lett. 2001; 494: 79-84Crossref PubMed Scopus (38) Google Scholar). It has been proposed that these carriers play a role in metal transport. We thus decided to investigate whether MRS3 andMRS4 genes play a role in mitochondrial iron import. S. cerevisiae strains used in this study were the wild-type strain W303–1B (MAT α ura3-52, leu2-3, 112, trp1-1, his3-11,15, ade2-1) and its isogenic derivative W303–1BΔYFH1 (MAT α ura3-52, leu2-3, 112, trp1-1, his3-11,15, ade2-1, yfh1Δ::KanR ) (9Foury F. Cazzalini O. FEBS Lett. 1997; 411: 373-377Crossref PubMed Scopus (342) Google Scholar). W303ΔYFH1–3C (MAT α ura3-52, leu2-3, 112, trp1-1, his3-11,15, ade2-1, yfh1Δ::KanR ) is a meiotic segregant of a W303 heterozygous diploid bearing a deleted copy ofYFH1. The ΔMRS3ΔMRS4 and ΔMMT1ΔMMT2 deletion strains were obtained by transformation of W303–1B with the loxP-kanR-loxP gene disruption cassette procedure (20Guldener U. Heck S. Fielder T. Beinhauer J. Hegemann J.H. Nucleic Acids Res. 1996; 24: 2519-2524Crossref PubMed Scopus (1346) Google Scholar). This cassette contains the Kanamycin resistance marker and can repeatedly be excised by homologous recombination between loxP sites upon yeast transformation with a plasmid containing the Cre recombinase under the control of the GAL1 promoter (20Guldener U. Heck S. Fielder T. Beinhauer J. Hegemann J.H. Nucleic Acids Res. 1996; 24: 2519-2524Crossref PubMed Scopus (1346) Google Scholar). First, yeast strains were transformed by the loxP-kanR-loxP cassette flanked on each side by 40–50 bp homologous to the 5′ and 3′ regions of the gene of interest (21Wach A. Brachat A. Pohlmann R. Philippsen P. Yeast. 1994; 13: 1793-1808Crossref Scopus (2232) Google Scholar). The oligonucleotide sequences were as follows: MRS3, 5′-GTTTGCTGTAAATATATTTATGTTATAAAAAAAGAAGATTATG (up) and 5′-GATGCAAAATGAATGGAAAATAATATGACATGTAAGAACCAACT (down);MRS4, 5′-AGTGAAGAAGTAAAAAACTCAACCGAAATATCAGTTAATATTATG (up) and 5′-TAAGTTGTCAAAAAAAGTTGACCTCGAAAAGGGGAAAAAACATCA (down);MMT1, 5′-TTATATATTATTGCATCACACAAACATCGCTTTCTTTTTCGCATTTTTGACA (up); 5′-TATTTTCGACATTTCCCTCTCTTCTTCACTTGTTATTATTACTGTGGGAAG (down);MMT2, 5′-GATAAGTATTGACTCTATCAAGCAATTCGGTTCCTTTGTGAC (up); 5′-ACAAACTCGACGTCCACCTTCCCACGTTTGGCACCTTCATAC (down). Singly deleted strains were thus obtained. Then, the deletion cassette was excised by the Cre recombinase, and a second deletion cassette obtained by polymerase chain reaction (PCR) amplification of the genomic DNA of another deleted strain was used to transform the singly deleted strain and produce a double deletion strain. Strains W303-1BΔMMT1ΔMMT2, W303-1BΔMRS3ΔMRS4, and W303-1BΔMRS3ΔMRS4ΔMMT1ΔMMT2 were constructed. The mutants ΔYFH1ΔMMT1ΔMMT2 and ΔYFH1ΔMRS3ΔMRS4 are meiotic segregants resulting from crosses between W303-1BΔMMT1ΔMMT2 (or W303-1BΔMRS3ΔMRS4) and W303ΔYFH1-3C. In the ΔYFH1ΔMRS3ΔMRS4 strain, theMRS4 deletion was obtained by genomic integration of a linear deletion cassette containing the URA3 marker flanked by PCR-amplified DNA fragments corresponding to the 5′ and 3′ sides of the MRS4 gene. The strains were routinely grown in 2% glucose (or 3% glycerol), 2% yeast extract KAT, or in minimum glucose medium containing 2% glucose, 0.67% yeast nitrogen base Difco supplemented with required amino acids and bases. Synthetic raffinose medium contains 2% raffinose, 0.5% glucose, 0.67% yeast nitrogen base, 0.5% ammonium sulfate, 0.05% yeast extract and is supplemented with amino acids. After an overnight culture in the appropriate medium, cells were inoculated to a final concentration of 20 × 106 cells/ml in 100 ml of the same medium containing 2 μm55Fe (FeCl3) and incubated with shaking for 3 h at 30 °C. Cells were harvested and washed two times with a solution of 100 mm sodium citrate and 10 mm EDTA, pH 6.5. Preparation of spheroplasts and mitochondria was carried out as described previously (9Foury F. Cazzalini O. FEBS Lett. 1997; 411: 373-377Crossref PubMed Scopus (342) Google Scholar). The amount of 55Fe was estimated by scintillation counting of the radioactivity contained in the mitochondrial and post-mitochondrial fractions. The assay measuring55Fe import into isolated mitochondria by following the formation of 55Fe-deuteroheme was carried out as described (6Lange H. Kispal G. Lill R. J. Biol. Chem. 1999; 274: 18989-18996Abstract Full Text Full Text PDF PubMed Scopus (133) Google Scholar). At the time of the publication of their article (13Hughes T.R. Marton M.J. Jones A.R. Roberts C.J. Stoughton R. Armour C.D. Bennett HA Coffey E. Dai H., He, Y.D. Kidd M.J. King A.M. Meyer M.R. Slade D. Lum P.Y. Stepaniants S.B. Shoemaker D.D. Gachotte D. Chakraburtty K. Simon J. Bard M. Friend S.H. Cell. 2000; 102: 109-126Abstract Full Text Full Text PDF PubMed Scopus (2090) Google Scholar), Hughes et al. (13Hughes T.R. Marton M.J. Jones A.R. Roberts C.J. Stoughton R. Armour C.D. Bennett HA Coffey E. Dai H., He, Y.D. Kidd M.J. King A.M. Meyer M.R. Slade D. Lum P.Y. Stepaniants S.B. Shoemaker D.D. Gachotte D. Chakraburtty K. Simon J. Bard M. Friend S.H. Cell. 2000; 102: 109-126Abstract Full Text Full Text PDF PubMed Scopus (2090) Google Scholar) constructed a data base publicly available, in which one could search for the 30 genes most closely regulated with a query gene with a certain correlation coefficient (p < 0.01 top < 1). The number of data points used in the correlation and the number of times the subject gene was regulated at a certain p value were also given. A paste of cells grown in the exponential phase of growth was reduced with dithionite in 0.7-mm-thick cuvettes and frozen in liquid nitrogen. Low temperature cytochrome absorption spectra were carried out, and cytochrome concentration was estimated as reported (22Claisse M.L. Pere-Aubert G.A. Clavilier L.P. Slonimski P.P. Eur. J. Biochem. 1970; 16: 430-438Crossref PubMed Scopus (95) Google Scholar). RNA peparation, Northern blotting analysis, and enzyme activities were carried out as described previously (7Foury F. Talibi D. J. Biol. Chem. 2001; 276: 7762-7768Abstract Full Text Full Text PDF PubMed Scopus (161) Google Scholar). Magnesium and cobalt concentrations were determined by atomic absorption spectrometry on samples treated with 5:2 nitric acid/perchloric acid at 80 °C for 2 h. To discover new genes co-regulated with theAFT1-dependent iron regulon, we performed a search in the data base generated by Hughes et al. (13Hughes T.R. Marton M.J. Jones A.R. Roberts C.J. Stoughton R. Armour C.D. Bennett HA Coffey E. Dai H., He, Y.D. Kidd M.J. King A.M. Meyer M.R. Slade D. Lum P.Y. Stepaniants S.B. Shoemaker D.D. Gachotte D. Chakraburtty K. Simon J. Bard M. Friend S.H. Cell. 2000; 102: 109-126Abstract Full Text Full Text PDF PubMed Scopus (2090) Google Scholar). Using DNA microarrays, the authors have obtained expression profiles for the complete genome of S. cerevisiae under 300 different conditions. A search for the genes most closely regulated with typical iron regulon genes such as FTR1, FET3,FRE2, or YOR383c led to detection ofMRS4 and MMT2 genes. Reciprocally, a search for genes closely regulated with MRS4 identified iron regulon genes. We have previously shown that the iron regulon is up-regulated in a ΔYFH1 strain (7Foury F. Talibi D. J. Biol. Chem. 2001; 276: 7762-7768Abstract Full Text Full Text PDF PubMed Scopus (161) Google Scholar). Therefore, we analyzed the expression ofMRS4 and MMT2 in wild-type and ΔYFH1 strains grown in glucose medium. By Northern blotting experiments we found that the mRNA levels of MRS4 and MMT2 were ∼2.5 times increased in a ΔYFH1 strain (Fig.1 A). It must be noted thatMMT2 expression was low compared with MRS4. In contrast with known iron regulon genes such as YOR382w (23Protchenko O. Ferea T. Rashford J. Tiedeman J. Brown P.O. Botstein D. Philpott C. J. Biol. Chem. 2001; 276: 49244-49250Abstract Full Text Full Text PDF PubMed Scopus (133) Google Scholar),MRS4 was expressed in ΔAFT1 strains (Fig. 1 B), and moreover, at a higher level than in Aft1+ strains. Recently, DNA microarray experiments have shown that MRS4expression is controlled by another iron regulatory pathway involvingAFT2, a paralogue of AFT1 (24Blaiseau P.L. Lesuisse E. Camadro J.M. J. Biol. Chem. 2001; 276: 34221-34226Abstract Full Text Full Text PDF PubMed Scopus (161) Google Scholar, 25Rutherford J.C. Jaron S. Ray E. Brown P.O. Winge D.R. Proc. Natl. Sci. U. S. A. 2001; 98: 14322-14327Crossref PubMed Scopus (127) Google Scholar). Double and quadruple deletion strains were constructed for MRS3, MRS4,MMT1, and MMT2 genes in wild-type and ΔYFH1 nuclear backgrounds. All strains were viable. However, cellular growth was decreased both in glucose- and glycerol-rich media (Fig.2 A). The slower growth on glucose medium did not result from formation of petite mutants (rho−). This indicates thatMRS3/MRS4 deletions impair cell metabolism not only inside mitochondria but also outside. ΔMRS3ΔMRS4 cells grew poorly in low iron media containing bathophenantroline sulfonate, an iron chelator (Fig. 2 A). Moreover, optimal growth of ΔMRS3ΔMRS4 and ΔYFH1ΔMRS3ΔMRS4 cells on minimum medium was only obtained upon addition of iron in the culture medium (Fig. 2, A and B). The high iron concentration (1 mm) used here inhibits the growth of ΔYFH1 cells, and therefore, the ΔMRS3ΔMRS4 deletion suppresses the iron sensitivity trait characterizing the ΔYFH1 strain. In contrast, after transformation of ΔYFH1 cells with anMRS4-bearing multicopy plasmid these cells were extremely sick, and moreover, had lost their mitochondrial genome. Loss of the mitochondrial DNA was shown by DAPI1 staining experiments (Fig. 3) and in crosses with a rhoo tester strain of opposite mating type, which gave no restoration of growth on glycerol medium (data not shown). These data suggest a link between mitochondrial iron metabolism and Mrs3p/Mrs4p transporters. No growth defect was observed in the ΔMMT1ΔMMT2 strain. Surprisingly, overexpression ofMMT2 suppressed iron sensitivity in ΔYFH1 cells (Fig. 2 C). The quadruple ΔMRS3ΔMRS4ΔMMT1ΔMMT2 deletion strain was not lethal, and its phenotype was identical to that of ΔMRS3ΔMRS4 strains (data not shown).Figure 3DAPI staining of mitochondrial DNA.Wild-type, ΔYFH1, and ΔYFH1[YEplac195/MRS4] strains grown in glucose-rich medium were suspended in 100 mm Tris-HCl, pH 8.0, and 1 μg/ml DAPI and incubated on ice for 10 min. Under these conditions, mitochondrial DNA is preferentially stained. The mitochondrial DNA spots are visible in wild-type and ΔYFH1 cells; nuclei are barely visible. Only the nuclei can be detected inMRS4-overexpressing ΔYFH1 cells. A Leica (DMR) microscope was used with a 340–380-nm band-pass excitation filter, a 400-nm dichroic reflector, and an emission filter at 425 nm.View Large Image Figure ViewerDownload Hi-res image Download (PPT) If MRS4 and MMT2 genes play a role in the transport of iron into mitochondria, their deletion should alter mitochondrial iron fluxes. Wild-type and mutant cells were radiolabeled with 55Fe for 3 h, and the radioactivity associated with mitochondria and post-mitochondrial supernatant was measured (Fig.4). The iron that accumulates in the mitochondria of ΔYFH1 cells grown in glucose-rich medium cannot be extruded (11Radisky D.C. Babcock M.C. Kaplan J. J. Biol. Chem. 1999; 274: 4497-4499Abstract Full Text Full Text PDF PubMed Scopus (248) Google Scholar). Therefore, defects in iron import into mitochondria should more easily be detected in a ΔYFH1 strain than in a frataxin-containing strain in which the regulation of the balance between uptake and efflux of mitochondrial iron is complex. Mitochondria of glucose-grown ΔYFH1 cells contained 40–45% of the cellular 55Fe compared with 5–10% for wild-type mitochondria (Fig. 4 A). In contrast, the fraction of55Fe present in ΔYFH1ΔMRS3ΔMRS4 mitochondria was low, close to the level found in wild-type mitochondria (Fig.4 A). These differences were still amplified when cells were cultivated in high iron media (50 μm) (Fig.4 B). Under these conditions, the intracellular55Fe content was similar in wild-type and ΔYFH1 strains. However, in the ΔYFH1 strain iron accumulated in mitochondria to the detriment of the non-mitochondrial fraction. In contrast, in the ΔYFH1ΔMRS3ΔMRS4 strain iron preferentially accumulated in a non-mitochondrial fraction (Fig. 4 B). Conversely, in a rhoo ΔYFH1 strain overproducing Mrs4p, the mitochondrial55Fe content was ∼2.5 times higher than in a rhoo ΔYFH1 strain (Fig. 4 F). The mitochondrial 55Fe content of ΔMRS3ΔMRS4 cells grown either in glucose or glycerol medium was approximately half of that present in the wild-type (Fig. 4, C and E). It must be noted that although intracellular iron concentration was increased in ΔMRS3ΔMRS4 cells compared with wild-type, extra iron was not sent to mitochondria (Fig. 4, C and E). As a consequence, the ratio of the mitochondrial to intracellular iron content decreased from 0.04 in the wild-type to 0.015 in the ΔMRS3ΔMRS4 strain grown in glucose-rich medium. We also used isolated energized mitochondria to follow the incorporation of 55Fe into protoporphyrin by mitochondrial ferrochelatase in wild-type and MRS4-overexpressing strains. This assay reflects the fraction of iron imported into mitochondria that can be used by the ferrochelatase. A greater fraction of55Fe was recovered in the heme ofMRS4-overexpressing cells (Fig.5). Altogether, these experiments show that loss of the Mrs3p and Mrs4p carriers is a limiting factor for iron import into mitochondria, specifically in ΔYFH1 cells grown in high iron media (Fig.4 B). However, a substantial amount of iron is still imported into mitochondria in the absence of Mrs3p/Mrs4p transporters. Therefore, if Mrs3p/Mrs4p transport iron efficient alternative iron uptake pathways that have yet to be identified are also present.MMT1 and MMT2 deletions had no major effect on mitochondrial iron incorporation and accumulation (Fig. 4,A, B, and E). However, MMT2overexpression in ΔYFH1 cells dramatically decreased the recovery of55Fe in mitochondria (Fig. 4 D), suggesting that Mmt2p does not import iron into mitochondria. Under standard conditions, mitochondrial DNA is stably maintained in our ΔYFH1 strain (9Foury F. Cazzalini O. FEBS Lett. 1997; 411: 373-377Crossref PubMed Scopus (342) Google Scholar). However, non-physiological high iron concentrations induce accumulation of cells devoid of mitochondrial DNA (9Foury F. Cazzalini O. FEBS Lett. 1997; 411: 373-377Crossref PubMed Scopus (342) Google Scholar). No loss of mitochondrial DNA was observed in ΔYFH1ΔMRS3ΔMRS4 cells even in the presence of very high iron concentrations (Fig.6 A), in agreement with the observation that iron does not accumulate in ΔYFH1ΔMRS3ΔMRS4 mitochondria. In contrast, the introduction of ΔMMT1ΔMMT2 deletions in a ΔYFH1 strain elicited significant increase in mitochondrial DNA loss when cells were cultivated in the presence of high iron concentrations (Fig. 6 A). Using Northern blot analysis we have found that mitochondrial DNA loss in ΔYFH1ΔMMT1ΔMMT2 mitochondria did not result from increased expression of the MRS4 gene (data not shown). Similarly,MMT2 was not overexpressed in ΔMRS3ΔMRS4ΔYFH1 cells. We found that the expression of typicalAFT1-dependent genes such as YOR382w,YLR136c, YHL040c, or COT1 was significantly increased in a ΔMRS3ΔMRS4 strain, although to a lower extent than in a ΔYFH1 strain (Fig. 7). The expression of the iron regulon genes was also higher in a ΔYFH1ΔMRS3ΔMRS4 strain than in a ΔYFH1 strain (data not shown). These data show that the loss of Mrs3p/Mrs4p is associated with activation of the AFT1-dependent regulon, implying that Aft1p encounters low iron concentrations. However, the fraction of 55Fe recovered in the post-mitochondrial supernatant of ΔMRS3ΔMRS4 (ΔYFH1) strains is high (Fig.4 C), indicating that this iron species is not perceived by Aft1p. Iron might be sequestered in a specific compartment such as the vacuole (26Raguzzi F. Lesuisse E. Crighton R.R. FEBS Lett. 1988; 231: 253-258Crossref PubMed Scopus (127) Google Scholar, 27Li L. Chen O.S. mcvey Wardzz D. Kaplan J. J. Biol. Chem. 2001; 276: 29515-29519Abstract Full Text Full Text PDF PubMed Scopus (268) Google Scholar). The expression ofAFT1-dependent genes was not increased in ΔMMT1ΔMMT2 cells. Wild-type, ΔYFH1, ΔMRS3ΔMRS4, and ΔYFH1ΔMRS3 ΔMRS4 strains were grown in glycerol-containing medium, and their cytochrome content as well as their aconitase, cytochrome coxidase, and oligomycin-sensitive ATPase activities were determined. The cytochrome content was not significantly modified in ΔMRS3ΔMRS4 cells (Fig. 8). Cytochrome coxidase and aconitase activities measured in isolated mitochondria were slightly decreased (Fig. 9, Aand D). Therefore, the amount of mitochondrial iron provided by the alternative import pathway was sufficient to ensure a large part of heme and iron-sulfur cluster biosynthesis in these cells growing at a slower rate. In ΔYFH1 cells, the cytochrome content was also normal, and cytochrome c oxidase was slightly affected, yet aconitase activity remained substantially lower than in wild-type mitochondria, although the apoprotein concentration estimated by Western blotting analysis was not modified (Fig.9 C). The decline in cytochrome content and in cytochrome oxidase and aconitase activities was more pronounced whenMRS3 and MRS4 deletions were introduced in a ΔYFH1 strain (Figs. 8 and 9). Therefore, in ΔYFH1ΔMRS3ΔMRS4 cells the decrease in mitochondrial iron concentration became a severely limiting factor. This indicates that a strain that has no frataxin needs more mitochondrial iron than a wild-type strain for heme and iron-sulfur cluster biosynthesis. The mitochondrial oligomycin-sensitive ATPase activity was similar in all strains (Fig.9 B), suggesting that iron-containing proteins were specifically affected. The deletion of MMT1 andMMT2 genes had no great effect on aconitase activity (Fig.9 A) and heme content (data not shown).Figure 9Mitochondrial enzyme activities in ΔMRS3ΔMRS4 and ΔMMT1ΔMMT2 cells. Activities were measured in mitochondria of glycerol-grown cells. Aconitase, cytochrome oxidase, and oligomycin-sensitive ATPase activities have been normalized to wild-type activities (100%). Aconitase activity is given as the ratio of aconitase to isocitrate dehydrogenase activity. Standard deviations were estimated using independent cell cultures and mitochondrial preparations. In the Western blot duplicates of wild-type and ΔYFH1 samples are shown.View Large Image Figure ViewerDownload Hi-res image Download (PPT) MRS3 and MRS4 are multicopy suppressors of mrs2 mutants (17Waldherr M. Ragnini A. Jank B. Teply R. Wiesenberger G. Schweyen R.J. Curr. Genet. 1993; 24: 301-306Crossref PubMed Scopus (60) Google Scholar). The latter have defects in magnesium transport into mitochondria (18Bui D.M. Gregan J. Jarosch E. Ragnini A. Schweyen R.J.J. Biol. Chem. 1999; 274: 20438-20443Abstract Full Text Full Text PDF Scopus (142) Google Scholar), and it has recently been shown that mitochondrial magnesium concentrations are restored to wild-type levels in an mrs2 mutant overexpressingMRS3 or MRS4 (28Gregan J. Kolisek M. Schweyen R.J. Genes Dev. 2001; 15: 2229-2237Crossref PubMed Scopus (61) Google Scholar). We did not find significant modifications in the mitochondrial magnesium concentration of strains deleted for MRS3/MRS4 or overexpressingMRS4 (Table I). We also tested the effect of heavy metals on cellular growth. The most striking effect was obtained with cobalt. Growth of wild-type cells was reduced in the presence of 2.5 mm cobalt (Fig.10, lane 1). Inhibition was stronger for ΔYFH1 cells and total for ΔMMT1ΔMMT2 cells (Fig. 10,lanes 2 and 4). In all strains inhibition of cellular growth by cobalt was associated with induction of rho− mutants (Fig. 6). In contrast, ΔMRS3ΔMRS4 cells were extremely resistant to cobalt (Fig. 10, lane 3), and no loss of mitochondrial DNA was observed in ΔYFH1ΔMRS3ΔMRS4 cells. Cells expressing MRS4 on a multicopy plasmid had increased sensitivity to cobalt (Fig. 10, lanes 5 and 6). We were not able to detect cobalt inside mitochondria of wild-type or mutant strains grown in the presence of 25 μm cobalt (Table I). However, cobalt was detected in mitochondria overexpressing Mrs4p (Table I), indicating cobalt accumulation in these mitochondria. Curiously, the cobalt resistant ΔMRS3ΔMRS4 strain exhibits a higher intracellular cobalt concentration than the wild-type strain (Table I). Since this cobalt is not toxic, it might be sequestered in the vacuole.MRS3/MRS4 deletions were associated with increased cellular sensitivity to hydrogen peroxide and heavy metals such as copper, cadmium, and to a lesser extent, zinc (Fig. 10,lanes 3).Table ICobalt and magnesium concentrations in mitochondrial and non-mitochondrial fractions of wild-type and mutant strainsStrainCobaltMagnesiumMitochondriaSupernatantMitochondriaSupernatant μg Co/mg protein μg Mg/mg proteinWild-type<0.20.72.134ΔMRS3ΔMRS4<0.23.02.234ΔMMT1ΔMMT2<0.20.82.030Wild-type*<0.2NDNDNDWt [MRS4]*0.7NDNDNDCells were cultivated in glucose-rich medium or raffinose minimum medium (*) in the presence of 25 μm cobalt chloride. Mitochondrial pellets and post-mitochondrial supernatants of the cobalt-treated cells were kept for cobalt and magnesium determination. Metal concentrations below 1 μg/ml were under detection limits. Wt [MRS4] is the wild-type strain carrying the multicopy Yeplac195/MRS4 plasmid. ND, not determined. Open table in a new tab Cells were cultivated in glucose-rich medium or raffinose minimum medium (*) in the presence of 25 μm cobalt chloride. Mitochondrial pellets and post-mitochondrial supernatants of the cobalt-treated cells were kept for cobalt and magnesium determination. Metal concentrations below 1 μg/ml were under detection limits. Wt [MRS4] is the wild-type strain carrying the multicopy Yeplac195/MRS4 plasmid. ND, not determined. On the basis of the capacity ofMRS3/MRS4-overexpressing strains to suppress the mitochondrial magnesium transport defect of mrs2 mutants, it has been proposed that these mitochondrial carriers transport metals. The discovery that MRS4 expression was specifically co-regulated with genes involved in iron mobilization from the external medium and intracellular stores led us to investigate whether Mrs4p mediates iron import into mitochondria. The expression of the AFT1-dependent iron regulon and intracellular iron content are increased in a ΔMRS3ΔMRS4 strain. Nevertheless, the fraction of iron recovered in mitochondria is ∼2-fold lower in a ΔMRS3ΔMRS4 mutant. Therefore, the extra iron pumped from the external medium by ΔMRS3ΔMRS4 cells is not perceived by Aft1p and does not enter mitochondria. It might be sequestered in the vacuole, which is an important iron store in yeast (25Rutherford J.C. Jaron S. Ray E. Brown P.O. Winge D.R. Proc. Natl. Sci. U. S. A. 2001; 98: 14322-14327Crossref PubMed Scopus (127) Google Scholar, 26Raguzzi F. Lesuisse E. Crighton R.R. FEBS Lett. 1988; 231: 253-258Crossref PubMed Scopus (127) Google Scholar). Mitochondrial iron content is a balance between uptake and exit, and therefore, decreased mitochondrial iron uptake in the ΔMRS3ΔMRS4 strain may be compensated by lower exit so that the final mitochondrial iron content is only modestly decreased. Moreover, the slower growth of ΔMRS3ΔMRS4 cells contributes to maintain mitochondrial iron concentration as constant as possible. Our data indicate that Mrs3p/Mrs4p contribute for approximately half of the mitochondrial iron content. The increase in the incorporation of iron into protoporphyrin IX by ferrochelatase in isolated mitochondria overexpressing Mrs4p suggests in a more direct way that Mrs4p plays a role in mitochondrial iron transport in the wild-type strain. The situation is quite different for ΔYFH1 cells cultivated in glucose-rich medium. The iron regulon is up-regulated even when cells are grown in high iron media, and a large part of the iron imported into the cell is directed to mitochondria. Moreover, no iron is exported from mitochondria (11Radisky D.C. Babcock M.C. Kaplan J. J. Biol. Chem. 1999; 274: 4497-4499Abstract Full Text Full Text PDF PubMed Scopus (248) Google Scholar). Therefore, the rate of mitochondrial iron uptake is more faithfully reflected in a ΔYFH1 strain. Although the intracellular iron is substantially increased in a ΔYFH1ΔMRS3ΔMRS4 strain grown in the presence of 50 μm iron, the mitochondrial iron content is five times lower than in a ΔYFH1 strain. Moreover, although frataxin-deficient cells cultivated in high iron media lose mitochondrial DNA, this effect is completely relieved by MRS3/MRS4 deletions. These data show that iron import into mitochondria is severely limited in ΔYFH1ΔMRS3ΔMRS4 cells. Conversely, MRS4 overexpression in ΔYFH1 cells elicits a 2.5-fold additional increase in the mitochondrial iron content. These cells have no mitochondrial DNA even under physiological conditions and grow very poorly, probably as a result of toxic mitochondrial iron overload. Therefore, in ΔYFH1 cells Mrs3p/Mrs4p carriers significantly contribute to mitochondrial iron accumulation. Up-regulation of the MRS4 transcript in the ΔYFH1 strain probably contributes to the process. Altogether, our experiments show that Mrs4p plays a role in mitochondrial iron uptake. This function is shared with at least another transporter that has not yet been identified. This unknown transporter might have high affinity for iron, since mitochondrial iron concentrations are only slightly increased in ΔYFH1ΔMRS3ΔMRS4 cells grown in high iron media. Reciprocally, Mrs3p and Mrs4p might be low affinity carriers. Increased expression of the iron regulon in ΔMRS3ΔMRS4 cells is associated with increased sensitivity to heavy metals, in particular cadmium and copper. It has been shown that iron starvation is associated with increased expression of broad specificity metal transporters in the plasma membrane (29Li L. Kaplan J. J. Biol. Chem. 1998; 273: 22181-22187Abstract Full Text Full Text PDF PubMed Scopus (135) Google Scholar). Increased hydrogen peroxide sensitivity of ΔMRS3ΔMRS4 cells could be explained by the increased oxidative damage produced by heavy metal intracellular overload. However, ΔMRS3ΔMRS4 cells exhibit an extreme resistance to cobalt. The loss of mitochondrial DNA, which is observed in wild-type and ΔYFH1 strains treated with high cobalt concentrations, is completely prevented by MRS3/MRS4 deletions. Curiously, these cobalt-resistant cells accumulate more cobalt than wild type. This suggests that cobalt is sequestered, perhaps in the vacuole, since the expression of COT1, which encodes a vacuolar cobalt transporter is increased in ΔMRS3ΔMRS4 cells (29Li L. Kaplan J. J. Biol. Chem. 1998; 273: 22181-22187Abstract Full Text Full Text PDF PubMed Scopus (135) Google Scholar). Mitochondrial cobalt concentration was under detection threshold in wild-type, ΔMRS3ΔMRS4, or ΔYFH1 cells. However, cobalt was detectable in mitochondria of MRS4-overexpressing cells. The latter exhibit increased sensitivity to cobalt. Whether cobalt plays a role in yeast cells, and more specifically in mitochondria, remains obscure. Inhibitory cobalt concentrations are extremely high, and toxicity might result from competition between iron and cobalt in prosthetic groups of iron-containing enzymes, such as ferrochelatase (30Shibayama N. Morimoto H. Kitagawa T. J. Mol. Biol. 1986; 192: 331-336Crossref PubMed Scopus (63) Google Scholar, 31Kwast K.E. Burke P.V. Staahl B.T. Poyton R.O. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 5446-5451Crossref PubMed Scopus (146) Google Scholar). It has previously been found that overexpressed Mrs4p can substitute for Mrs2p, a mitochondrial inner membrane protein involved in magnesium transport (18Bui D.M. Gregan J. Jarosch E. Ragnini A. Schweyen R.J.J. Biol. Chem. 1999; 274: 20438-20443Abstract Full Text Full Text PDF Scopus (142) Google Scholar), suggesting that Mrs4p can also be used for import of magnesium into mitochondria (28Gregan J. Kolisek M. Schweyen R.J. Genes Dev. 2001; 15: 2229-2237Crossref PubMed Scopus (61) Google Scholar). Two other transporters in the inner mitochondrial membrane that belong to the family of cation efflux facilitators, Mmt1p and Mmt2p, have also been reported to play a role in mitochondrial iron import (4Li L. Kaplan J. J. Biol. Chem. 1997; 272: 28485-28493Abstract Full Text Full Text PDF PubMed Scopus (72) Google Scholar). Moreover, like MRS4, MMT2 is co-regulated with several iron regulon genes. MMT1 and MMT2 were isolated as multicopy suppressors of the growth defect of anerg25 mutant in low iron media (4Li L. Kaplan J. J. Biol. Chem. 1997; 272: 28485-28493Abstract Full Text Full Text PDF PubMed Scopus (72) Google Scholar). The authors found that the double deletion mutant grown in low iron media reaches a stationary phase for a lower cell density. Moreover, cells overexpressingMMT1 or MMT2 in high iron (250 μm) media have increased levels of iron both in cytosol and mitochondria. We also found that MMT2 overexpression increases the intracellular iron content, but we did not observe significant change in the mitochondrial iron level. It must be noted, however, that our culture medium contained lower iron concentrations (5–50 μm) to avoid aspecific binding of iron on yeast membranes (6Lange H. Kispal G. Lill R. J. Biol. Chem. 1999; 274: 18989-18996Abstract Full Text Full Text PDF PubMed Scopus (133) Google Scholar). Moreover, as previously reported by others (6Lange H. Kispal G. Lill R. J. Biol. Chem. 1999; 274: 18989-18996Abstract Full Text Full Text PDF PubMed Scopus (133) Google Scholar), we did not observe a clear phenotype linked to iron in ΔMMT1ΔMMT2 or ΔYFH1ΔMMT1ΔMMT2 deletion strains. This may be explained by the low expression of these transporters. In contrast to a previous report (4Li L. Kaplan J. J. Biol. Chem. 1997; 272: 28485-28493Abstract Full Text Full Text PDF PubMed Scopus (72) Google Scholar), our deletion strains were sensitive to cobalt. Finally, the dramatic decrease in the mitochondrial 55Fe level observed in ΔYFH1 cells overexpressing MMT2 does not support the hypothesis that Mmt2p imports iron into mitochondria but agrees with our observation that MMT2 behaves as a multicopy suppressor of the iron sensitivity trait characterizing ΔYFH1 cells. These data suggest that Mmt1p and Mmt2p play a role in the efflux of iron from mitochondria. It has often been admitted that an increase in toxic free radicals generated by mitochondrial iron overload is a cause of the mitochondrial dysfunction in frataxin-deficient cells (11Radisky D.C. Babcock M.C. Kaplan J. J. Biol. Chem. 1999; 274: 4497-4499Abstract Full Text Full Text PDF PubMed Scopus (248) Google Scholar). This is probably true in ΔYFH1 cells cultivated in the presence of high non-physiological iron concentrations or when Mrs4p is overexpressed. We also found that ΔYFH1 cells that do not eliminate superoxide anions because they lack Sod2p, the mitochondrial superoxide dismutase, are extremely sick. 2F. Foury, unpublished data. Under these conditions, the mitochondrial genome is destroyed. However, in glycerol-grown ΔYFH1 cells, there is no mitochondrial iron overload, and a specific defect in the activity of iron-sulfur proteins such as aconitase is still observed. Moreover, a decrease in mitochondrial iron concentration is associated with a further decrease in aconitase activity. This means that in a frataxin-deficient strain more iron is required to synthesize iron-sulfur clusters, implying that mitochondrial iron is not efficiently used. This conclusion is in agreement with the report that multimers of yeast frataxin sequester large amounts of iron in a bioavailable form (10Adamec J. Rusnak F. Owen W.G. Naylor S. Benson L.M. Gacy A.M. Isaya G. Am. J. Hum. Genet. 2000; 67: 549-562Abstract Full Text Full Text PDF PubMed Scopus (229) Google Scholar), suggesting that frataxin could regulate the delivery of iron for the biosynthesis of iron-sulfur clusters. A role for frataxin in iron-sulfur cluster metabolism is also supported by data obtained with conditional frataxin-deficient mice showing that a decline in the activity of iron-sulfur cluster proteins occurs prior to mitochondrial iron deposits (32Puccio H. Simon D. Cossee M. Criqui-Filipe P. Tiziano F. Melki J. Hindelang C. Matyas R. Rustin P. Koenig M. Nat. Genet. 2001; 27: 181-186Crossref PubMed Scopus (590) Google Scholar). Moreover, on the basis of phylogenetic studies run on 56 completely sequenced genomes, it has recently been proposed that frataxin plays a role in iron-sulfur cluster biosynthesis (33Huynen M.A. Snel B. Bork P. Gibson T.J. Hum. Mol. Genet. 2001; 10: 2463-2468Crossref PubMed Scopus (136) Google Scholar). We thus propose that two defective mechanisms co-exist in frataxin deficient cells. First, biosynthesis of iron-sulfur clusters is impaired (34Foury F. FEBS Lett. 1999; 456: 281-284Crossref PubMed Scopus (143) Google Scholar), and second, oxidative damage occurs under conditions of excess mitochondrial iron. Anne-Marie Faber performed the DAPI staining experiments. We thank Dr. Roland Lill (Marburg, Germany) for the gift of the aconitase antibody. F. Foury thanks the University Pierre and Marie Curie for a 1-month stay as Invited Professor in the laboratory of Maurice Claisse in the CNRS Center of Molecular Genetics (Gif-sur-Yvette, France). Dr. M. Claisse is greatfuly acknowledged for cytochrome spectra facilities and discussions.
Referência(s)