- Email: [email protected]
The role of the mitochondrion in cellular iron homeostasis
Mitochondrion 1 (2001) 51±60
www.elsevier.com/locate/mito
The role of the mitochondrion in cellular iron homeostasis Nichole D. Schueck a,1, Michael Woontner a,1, David M. Koeller b,* a
Department of Pediatrics, University of Colorado Health Sciences Center, Denver, CO 80262, USA b Department of Pediatrics, Oregon Health Sciences University, Portland, OR 97201, USA
Received 12 December 2000; received in revised form 13 February 2001; accepted 16 February 2001
Abstract The yeast ATM1 protein is essential for normal mitochondrial iron homeostasis. Deletion of ATM1 results in mitochondrial iron accumulation and oxidative mitochondrial damage. Mutations in ABC7, the human homolog of ATM1, result in X-linked sideroblastic anemia and ataxia. Here we show that a deletion of ATM1 also has effects on extra-mitochondrial iron metabolism. ATM1-de®cient cells have an increased iron requirement for growth. When grown in iron-rich medium, mutant cells accumulate excess mitochondrial iron and have increased expression of the genes required for both high and low af®nity iron uptake. Thus, ATM1 mutant cells simultaneously demonstrate features of both iron overload and iron starvation. Yfh1p is the yeast homolog of the human frataxin protein, which is de®cient in Friedreich's ataxia. As in atm1 cells, a yfh1 deletion results in both mitochondrial iron accumulation and cytosolic iron starvation. In spite of their apparent roles in cellular iron homeostasis, we ®nd that the expression of neither ATM1 nor YFH1 is responsive to cellular iron status. Based on these observations, we propose a model in which cellular iron is prioritized for use by the mitochondrion, and available to the remainder of the cell only after mitochondrial needs have been ful®lled. q 2001 Elsevier Science B.V. and Mitochondria Research Society. All rights reserved. Keywords: Mitochondrion; Iron metabolism; Ataxia
1. Introduction Pathways for both high and low af®nity cellular iron uptake in yeast have been well characterized (reviewed in Eide, 1998). However, the mechanisms for the intracellular distribution of imported iron remain enigmatic. An essential requirement for effective intracellular iron traf®cking is the ability to prioritize iron delivery to the multiple sites within the cell where it is required, including the endoplasmic reticulum, mitochondria and peroxisomes. A means to minimize toxicity resulting from the iron-catalyzed * Corresponding author. Tel.: 11-503-494-2783; fax: 11-503494-2781. E-mail address: [email protected] (D.M. Koeller). 1 These authors contributed equally to this project.
formation of oxygen radical species is also imperative. A signi®cant fraction of the cell's iron requirement is for the assembly of iron sulfur clusters and heme biosynthesis in the mitochondrion. Because of its central role in cellular metabolism, the assurance of mitochondrial iron suf®ciency is predicted to have a high priority within any intracellular iron distribution network. However, proteins functioning in the import of iron to the mitochondrion have not been identi®ed. The ATM1 and YFH1 genes of Saccharomyces cerevisiae have roles in mitochondrial iron export (Allikmets et al., 1999; Radisky et al., 1999). Disruption of the ATM1 gene results in the accumulation of a 30fold excess of mitochondrial iron, loss of mitochondrial cytochromes, oxidative damage to the mitochon-
1567-7249/01/$20.00 q 2001 Elsevier Science B.V. and Mitochondria Research Society. All rights reserved. PII: S 1567-724 9(01)00004-6
52
N.D. Schueck et al. / Mitochondrion 1 (2001) 51±60
drial DNA, and decreased levels of cytosolic heme proteins (Leighton and Schatz, 1995; Kispal et al., 1997). Atm1p is a member of the ATP-binding cassette (ABC) superfamily of transporters. This family of transporters consists of a large number of evolutionarily related proteins involved in energydependent transport of substrates across cellular membranes (Dean and Allikmets, 1995). Atm1p is located in the mitochondrial inner membrane with its carboxyl terminus in the matrix (Leighton and Schatz, 1995). Based on this orientation it is predicted to export substrate from the matrix to the intermembranous space. The phenotype of cells with a deletion of the YFH1 gene is very similar to that resulting from disruption of ATM1. yfh1 cells accumulate a ten-fold excess of mitochondrial iron and become respiratory-de®cient secondary to oxidative damage to the mitochondrial DNA (Babcock et al., 1997; Foury and Cazzalini, 1997; Koutnikova et al., 1997; Wilson and Roof, 1997). Yfh1p has been localized to the mitochondrion (Babcock et al., 1997), where it is proposed to function as a regulator of mitochondrial iron export (Radisky et al., 1999). The similarity of the phenotypes resulting from ATM1 and YFH1 deletions has led to the hypothesis that Atm1p is a mitochondrial iron exporter, which is regulated by Yfh1p (Allikmets et al., 1999; Radisky et al., 1999). Homologs of the yeast ATM1 and YFH1 proteins are associated with inherited disease in man (Campuzano et al., 1996; Allikmets et al., 1999). A mutation in ABC7, the human homolog of ATM1, has been identi®ed as the cause of X-linked sideroblastic anemia and ataxia (Allikmets et al., 1999). Sideroblastic anemias are characterized by iron accumulation in the mitochondria of bone marrow erythroid precursors (Bottomley, 1993). The similarity between the biochemical phenotypes of ATM1 and ABC7 mutations (i.e. mitochondrial iron accumulation), and the ability of ABC7 to complement a yeast ATM1 mutation (Allikmets et al., 1999; Csere et al., 1998), indicate that these proteins perform a similar function in the mitochondrial inner membrane. Mutations in the gene for Frataxin, the human homolog of Yfh1p, result in Friedreich's ataxia, a progressive neurodegenerative disease characterized by gait ataxia, cardiomyopathy and diabetes (Campuzano et al., 1996). The observation that a deletion of the YFH1 gene
(YFH, Yeast Frataxin Homolog) is associated with mitochondrial iron accumulation and oxidative injury in yeast was the ®rst clue as to the possible function of frataxin (Babcock et al., 1997). Loss of the frataxin protein in humans is also associated with mitochondrial iron deposition and oxidative injury (Lamarche et al., 1980; Rotig et al., 1997). Prior studies have focused primarily on the effect of an ATM1 deletion on mitochondrial function. In this report we describe the effect of a deletion of ATM1 on extra-mitochondrial metal metabolism. We also present data on the in¯uence of cytosolic and mitochondrial iron status on ATM1 and YFH1 expression. Based on these experiments we propose a model in which iron entering the cell is prioritized ®rst for import into the mitochondrion, and only available to the remainder of the cell when mitochondrial needs have been ful®lled.
2. Materials and methods 2.1. Strains and culture methods The yeast strains were as follows: DY150 (MATa ura3-52 leu2-3,112 his3-11,15 trp1-1 ade2-1 can1100(oc)) and DY1457 (MATa ura3-52 leu2-3,112 his3-11,15 trp1-1 ade6 can1-100(oc)) (Askwith et al., 1994). AFT1 CM3260 (MATa trp1-63 leu23,112 gcn4-101 his3-609 ura3-52), and AFT11 up M2p (MATa trp1-63 leu2-3,112 gcn4-101 his3-609 ura3-52 AFT1-1 up) (Yamaguchi-Iwai et al., 1995). For growth in iron-rich medium, cells were cultured at 308C in 1% yeast extract, 2% peptone, and 2% glucose (YPD). For iron starvation, bathophenanthroline sulfonate (BPS) was added to 100 mM. Synthetic complete medium (SC) is 0.67% yeast nitrogen base with amino acids and 2% glucose. Low iron medium (LIM) is SC 1 1 mM EDTA and 20 mM citrate (Eide and Guarente, 1992). Media for assessment of transition metal sensitivity were made by addition of metal salts from stock solutions (ZnSO4, CuSO4, CoCl2, and MnSO4) to SC. Media were inoculated with a 1:100 dilution of cultures grown to saturation in YPD and analyzed for growth after 16±22 h. DNA transformations and manipulations of Escherichia coli and Saccharomyces cerevi-
N.D. Schueck et al. / Mitochondrion 1 (2001) 51±60
siae were done by standard techniques (Sambrook et al., 1989; Sherman, 1991). 2.2. Disruption of the ATM1 gene A disruption allele was made by inserting the LEU2 gene into a unique NheI site in the ATM1 gene and transformed into the diploid formed by mating DY150 £ DY1457. Transformed cells were selected for leucine prototrophy, and screened for disruption of the ATM1 gene by PCR and Southern blotting. Sporulation and dissection of tetrads from the heterozygous diploid yielded the haploid strain YNS1-5D (MATa ade2 ura3-52 his3-11,15 trp1-1 can1100(oc) leu2-3,112 atm1::LEU2). Sporulation of diploid strains and dissection of tetrads were as described (Sherman, 1991). 2.3. High af®nity Fe(II) uptake Whole cell iron uptake was measured by a modi®cation of the procedure of Eide and Guarente (1992). Cells were grown in YPD or YPD 1 100 mM BPS, harvested by centrifugation, washed twice with icecold 10 mM MES (pH 6.1), resuspended in 10 mM MES (pH 6.1)/2% glucose, and counted with a hemacytometer. All manipulations were done on ice or at 48C. Five million (DY150) or one million (YNS1-5D) cells were assayed in a total volume of 0.5 ml in 10 mM MES (pH 6.1), 2% glucose, 1 mM sodium ascorbate, and 1.0 mM 55FeCl3. Uptake was performed in duplicate at both 0 and 308C for 10 min. Cells were collected by ®ltration through Whatman GF/C ®lters, washed twice with ice-cold SSW (1 mM EDTA, 1 mM CaCl2, 20 mM sodium citrate (pH 4.2), 5 mM MgSO4, 1 mM KH2PO4, and 1 mM NaCl), air-dried and counted in a scintillation counter. The rate of uptake at 308C was calculated after subtraction of the non-speci®c cell associated radioactivity determined from the samples incubated at 08C. 2.4. Northern blot analysis RNA was isolated by glass bead lysis in the presence of phenol from log phase cultures grown in YPD. For each sample, 10 mg of total RNA was separated on a formaldehyde agarose gel and transferred to a nylon membrane by electroblotting. Probes for ATM1, YFH1, CTR1, FET3, FET4, and ZRT1 were
53 32
produced by random-primed a- P labeling using the RadPrime DNA Labeling System (Gibco BRL). For 18S rRNA a 32P end-labeled oligonucleotide probe was used. Hybridization signals were quantitated on a Molecular Dynamics Storm phosphorimager. mRNA signals were normalized to 18S rRNA signals. 2.5. Immunoblot analysis Mitochondria were isolated by differential centrifugation (Yaffe, 1991) and equal amounts of protein were separated by SDS-PAGE and transferred to a polyvinylidene ¯uoride (PVDF) membrane by electroblotting. Western blot analysis (Harlow and Lane, 1988) was performed with a rabbit polyclonal antibody generated against Yfh1p (Branda et al., 1999) and horseradish peroxidase-conjugated goat antirabbit IgG (Cappel). 3. Results A lack of the mitochondrial inner membrane ABC transporter Atm1p results in mitochondrial iron overload (Kispal et al., 1997; Csere et al., 1998). It is not known whether the excess iron in atm1 cells can be mobilized for utilization, or is metabolically inert. We hypothesized that atm1 cells would have increased tolerance to iron starvation if they are able to mobilize the accumulated mitochondrial iron. The ability to grow under iron-limited conditions was determined in LIM supplemented with varying amounts of iron. The growth of both wild-type (DY150) and atm1 cells is decreased in LIM with no added iron when compared to cultures grown with 500 mM FeCl3 (Fig. 1). Addition of as little as 5 mM FeCl3 restored the growth of wild-type cells to a level nearly equal to that in the high iron medium (500 mM FeCl3). In contrast, addition of as much as 20 mM FeCl3 failed to improve the growth of the atm1 mutant cells (Fig. 1), demonstrating an increased iron requirement for growth. These results suggest that the excess mitochondrial iron in atm1 cells is unavailable for utilization by the rest of the cell. Having determined that atm1 cells have an increased iron requirement for growth when measured under iron-limiting conditions, we evaluated the effect of an atm1 mutation on the iron status of cells grown
54
N.D. Schueck et al. / Mitochondrion 1 (2001) 51±60
Fig. 1. Effect of iron availability on the growth of wild-type and atm1 mutant cells. Wild-type (closed circles) and atm1 mutant (open circles) cells were pre-cultured in SC and diluted into low iron media containing varying amounts of FeCl3. Each point represents the mean of two experiments, each done in duplicate. Error bars, ^S.E.M.
Fig. 2. Comparison of high af®nity iron uptake by wild-type and atm1 mutant cells. Wild-type (DY150) and atm1 mutant (YNS15D) cells were grown to log phase in YPD, harvested, washed, and assayed for iron uptake in the presence of 1.0 mM 55FeCl3 and 1 mM ascorbate. To induce iron starvation DY150 cells were grown to log phase in YPD, diluted into fresh YPD containing BPS (100 mM), and grown for an additional 4 h. Data represent the mean of three experiments. Error bars, ^S.E.M.
in iron-rich medium. High af®nity cellular iron uptake is mediated by Ftr1p and Fet3p (Eide, 1998; Radisky and Kaplan, 1999). The expression of these genes is under the transcriptional control of Aft1p, a transcription factor whose activity is responsive to cellular iron levels (Yamaguchi-Iwai et al., 1995, 1996). Under conditions of iron starvation, Aft1p acts as a transcriptional activator to induce high levels of FET3 and FTR1 mRNA synthesis, leading to an increased rate of iron uptake. Measurement of high af®nity cellular iron uptake can therefore be used to assess cellular iron status. Wild-type cells grown in YPD made iron-de®cient by the addition of 100 mM BPS have a three-fold increase in high af®nity iron uptake (Fig. 2), and a corresponding increase in FET3 mRNA levels (Fig. 3), when compared to cells grown in YPD alone. The rate of iron uptake in atm1 cells grown in iron-rich medium (YPD) is similar to that of wild-type cells grown under conditions of iron starvation (Fig. 2). Measurement of FET3 mRNA levels in atm1 cells shows a 12-fold increase relative to wildtype cells grown in YPD (Fig. 3). The level of FET3 expression in atm1 cells is comparable to that seen in cells with a constitutively active allele of AFT1 (AFT1-1 up) (Fig. 3) (Yamaguchi-Iwai et al., 1995).
Fig. 3. Northern blot analysis of FET3 mRNA. Total RNA was isolated from log phase cultures of atm1 mutant (atm1), wild-type (DY150), AFT1-1 up and its wild-type parent (AFT1) grown in YPD, or YPD containing 100 mM BPS (DY150 1 BPS) and probed for FET3 and 18S rRNA. Quantitation of RNA was done with a phosphorimager and relative levels of FET3 mRNA calculated after normalization to 18S rRNA.
N.D. Schueck et al. / Mitochondrion 1 (2001) 51±60
Thus, when grown under conditions that result in mitochondrial iron accumulation (Csere et al., 1998; Kispal et al., 1997), atm1 mutant cells are sensing iron starvation. In addition to iron, normal mitochondrial function requires zinc for metalloproteases, manganese for Mn-superoxide dismutase, and copper for cytochrome oxidase. To determine the consequences of an atm1 mutation on the metabolism of other transition metals, we measured the growth of atm1 cells in media supplemented with varying levels of Cu, Zn, Co, and Mn. In comparison to wild-type, atm1 mutant cells are extremely sensitive to all four of these metals (Fig. 4). To determine whether the observed metal sensitivity was due to abnormal regulation of plasma membrane transporters, we measured the expression of genes involved in Cu and Zn uptake. Two plasma membrane transporters, Ctr1p and Ctr3p, mediate
55
high af®nity cellular copper uptake (Dancis et al., 1994; Knight et al., 1996). Both high and low af®nity transporters (Zrt1p and Zrt2p, respectively) function in zinc import (Zhao and Eide, 1996a,b). Northern blot analysis of CTR1 and ZRT1 mRNA levels showed less than a two-fold increase in atm1 mutant cells (Fig. 5A,B). This small increase in CTR1 and ZRT1 expression is comparable to that seen in wild-type cells grown in iron-limiting conditions, and seems unlikely to explain the extreme sensitivity of atm1 cells to these metals. An alternative hypothesis for the metal sensitivity of atm1 cells is an increased expression of the FET4 gene. Fet4p is a component of the low af®nity plasma membrane iron uptake mechanism, and is also capable of transporting several additional transition metals (Dix et al., 1994; Eide, 1998). The lack of iron speci®city of Fet4p results in generalized metal sensitivity
Fig. 4. Comparison of the effect of transition metals on the growth of wild-type and atm1 mutant cells. Wild-type (®lled diamonds) and atm1 mutant (open squares) cells were pre-cultured in YPD and then diluted into SC containing varying amounts of copper (A), zinc (B), cobalt (C), or manganese (D).
56
N.D. Schueck et al. / Mitochondrion 1 (2001) 51±60
Fig. 5. Northern blot analysis of copper, zinc, and iron transporter mRNA expression in wild-type and atm1 mutant cells. Total RNA was isolated from log phase cultures of atm1 mutant (atm1) and wild-type (DY150) cells grown in YPD, or YPD containing 100 mM BPS (DY150 1 BPS) and probed for CTR1 (A), ZRT1 (B), FET4 (C) and 18S rRNA. Quantitation of RNAs was done with a phosphorimager and relative levels of speci®c mRNAs calculated after normalization to 18S rRNA.
in cells lacking a high af®nity iron uptake pathway and therefore dependent on Fet4p for iron uptake (Li and Kaplan, 1998). Growth of wild-type cells in iron-limited medium has little effect on FET4 mRNA levels (Fig. 5C). In contrast, atm1 cells show a ®vefold increase in FET4 expression, supporting the hypothesis that elevated expression of FET4 is responsible for the metal sensitivity of atm1 mutant cells. These data also demonstrate that an atm1 mutation results in the induction of both the high and low af®nity iron transport pathways. In contrast, growth of wild-type cells in iron-de®cient medium only induces the genes required for high af®nity iron uptake (Figs. 3 and 5), indicating a unique character to the iron starvation resulting from an atm1 mutation. The dramatic effect of a mutation in the ATM1 gene on cellular iron homeostasis raises the question as to the effect of cellular iron status on ATM1 expression. Northern blot analysis shows that in wild-type cells, iron starvation induced by growth in YPD 1 BPS does not affect the level of ATM1 mRNA (Fig. 6). Similarly, there is only a slight change in the ATM1 mRNA level in an AFT1-1 up strain. Consistent with the lack of iron-dependent regulation there are no consensus Aft1p binding sites (Yamaguchi-Iwai et al., 1996) in the ATM1 promoter. Thus, the expression of ATM1 is not sensitive to changes in cellular iron status induced by conditions known to effect the expression of other iron-regulated genes. The apparent lack of iron-dependent regulation of ATM1 expression is surprising given the necessity of
Atm1p for normal cellular iron homeostasis. The localization of Atm1p to the mitochondrial inner membrane suggests that a signal derived from the mitochondrion may have a role in regulating its expression; however, no such signals are currently known. As an alternative, we chose to look at the affect of a yfh1 mutation on ATM1 expression. The YFH1 protein functions in the regulation of mitochondrial iron ef¯ux (Radisky et al., 1999); a lack of func-
Fig. 6. Northern blot analysis of ATM1 mRNA levels. Total RNA was isolated from log phase cultures of wild-type (DY150), AFT11 up and its wild-type parent (AFT1) grown in YPD, or YPD containing 100 mM BPS (DY150 1 BPS) and probed for ATM1 and 18S rRNA. Quantitation of RNA was done with a phosphorimager and relative levels of ATM1 mRNA calculated after normalization to 18S rRNA.
N.D. Schueck et al. / Mitochondrion 1 (2001) 51±60
tional Yfh1p results in mitochondrial iron accumulation (Babcock et al., 1997; Foury and Cazzalini, 1997). Northern blot analysis revealed no difference in ATM1 mRNA levels in yfh1 cells as compared to wild-type (data not shown). In contrast, the level of YFH1 mRNA and protein is decreased four-fold in atm1 mutant cells (Fig. 7). This result suggested the possibility that the iron accumulation seen in atm1 cells was secondary to a de®ciency of Yfh1p, as seen in cells with an ssq1 mutation (Knight et al., 1996). Expression of YFH1 on a high copy plasmid using either its own promoter or a heterologous promoter had no affect on the phenotype of atm1 cells (data not shown). However, overexpression of yfh1p in these experiments was not con®rmed by Western blotting. We did not see any changes in YFH1 expression in cells grown under iron-limiting conditions, or in the presence of the AFT1-1 up allele (Fig. 7), indicating that a signal other than cytosolic iron starvation is responsible for the observed decrease in YFH1 expression.
57
4. Discussion We have shown that in addition to abnormal mitochondrial iron homeostasis, atm1 cells have anomalous cytosolic iron metabolism. A deletion of ATM1 results in an increased sensitivity to iron starvation. In iron-rich medium atm1 cells accumulate toxic levels of mitochondrial iron and simultaneously show an iron-starved pattern of nuclear gene expression. Thus, deletion of ATM1 is affecting the cell's ability to ensure that suf®cient iron is available for both mitochondrial and extra-mitochondrial needs, suggesting that Atm1p plays a role in the intracellular distribution of iron. The orientation of Atm1p in the mitochondrial inner membrane predicts that it functions as an exporter (Leighton and Schatz, 1995). The accretion of iron by mitochondria in atm1 cells suggests that the substrate for Atm1p is iron, for which an export pathway has been demonstrated (Radisky et al., 1999). Because of the affect of an atm1 mutation on the cytosolic Fe-S enzyme isopropyl malate isomerase
Fig. 7. Expression of YFH1 in atm1 mutant cells. (A) Total RNA was isolated from log phase cultures of atm1 mutant (atm1), wild-type (DY150), AFT1-1 up and its wild-type parent (AFT1) grown in YPD, or YPD containing 100 mM BPS (DY150 1 BPS) and probed for YFH1 and 18S rRNA. Quantitation of RNA was done with a phosphorimager and relative levels of YFH1 mRNA calculated after normalization to 18S rRNA. (B) Equal amounts of mitochondrial protein from wild-type (DY150) and atm1 mutant cells (atm1) were separated by SDS-PAGE, transferred to a PVDF membrane, and analyzed by Western blotting with a polyclonal rabbit anti-Yfh1p antibody. Proteins were visualized by ECL using a peroxidase-conjugated goat anti-rabbit secondary antibody.
58
N.D. Schueck et al. / Mitochondrion 1 (2001) 51±60
(Leu1p), it has been proposed that the function of Atm1p is to export Fe-S clusters assembled in the mitochondrion (Kispal et al., 1999). However, our data demonstrate that in addition to an inability to form cytosolic iron sulfur clusters, atm1 cells have a generalized cytosolic de®ciency of iron, as demonstrated by an increased expression of the genes required for high af®nity iron uptake. Simultaneous mitochondrial iron accumulation and cytosolic iron starvation has previously been demonstrated in cells with a deletion of NFS1, a cysteine desulfurase that provides sul®de for mitochondrial iron sulfur cluster formation (Li et al., 1999), or a deletion of ISU1, which is believed to function in the mobilization of iron for mitochondrial Fe-S cluster formation (unpublished data). Deletion of the YFH1 gene also results in mitochondrial iron overload and cytosolic iron starvation, as demonstrated by increased FET3 expression (Babcock et al., 1997). Using an allele of YFH1 controlled by an inducible promoter, Radisky et al. (1999) showed that in cells grown under conditions that repress YFH1 expression, mitochondria accumulate iron and FET3 expression is high. When switched to permissive growth conditions, the appearance of Yfh1p is rapidly followed by a decrease in both mitochondrial iron content and FET3 mRNA levels. The decrease in the expression of FET3 occurs without any change in the iron content of the medium, indicating that the iron released from the mitochondrion is being made available to the rest of the cell. Based on the similar effects of ATM1, NFS1, ISU1 and YFH1 deletions on mitochondrial and cellular iron metabolism, we hypothesize that after entry into the cell, iron is prioritized for use by the mitochondrion, and is only available to the rest of the cell after mitochondrial needs have been ful®lled. A model in which imported iron is speci®cally targeted to the mitochondrion has previously been proposed for mammalian erythroid cells (Ponka, 1997). The nature of the mechanism for sensing mitochondrial iron status and regulation of iron export is unknown. Deletion of ATM1 also results in increased expression of the low af®nity iron importer FET4. Wild-type cells grown in iron-limited medium or expressing the AFT1-1 up allele have elevated expression of the genes for high af®nity iron uptake, but do not show signi®cant induction of FET4 (Fig. 5) (Casas et al., 1997). In contrast, in cells lacking high af®nity iron uptake due
to a FET3 deletion, FET4 expression is induced by iron starvation (Dix et al., 1997). The increased sensitivity to transition metal toxicity of atm1 cells is also seen in cells de®cient in high af®nity iron import, and results from the ability of other metals to compete with iron for import via Fet4p (Dix et al., 1994; Li and Kaplan, 1998). The requirement of atm1 cells for greater than 20 mM iron for growth, which is greater than 100 times the KM of the high af®nity import pathway, further supports their dependence on iron imported via Fet4p (K M 30 mM). Thus, ATM1 mutant cells have phenotypic features similar to cells lacking high af®nity iron. Interestingly, deletion of FET3 prevents the accumulation of excess mitochondrial iron in yfh1 cells (Radisky et al., 1999), raising the possibility that the route of entry (i.e. the high vs. low af®nity pathway) may in part determine where imported iron is delivered. Manipulations known to affect the expression of iron-regulated genes, including iron starvation and the AFT1-1 up allele, have only a small effect on the level of ATM1 mRNA. Elevation of mitochondrial iron secondary to a YFH1 deletion was also without effect on ATM1 expression. Alternatively, it may be the transport activity of Atm1p that is subject to regulation. The ATM1 protein is an ABC transporter. A general mechanism for the function of ABC transporters is that upon binding of substrate to the membrane spanning domain there is activation of the ATPase and subsequent transfer of the bound substrate across the membrane (Holland and Blight, 1999). Transport activity can also be regulated in response to the metabolic state of the cell and modi®cations such as phosphorylation. It is highly unlikely that there would be free iron in the mitochondrion that could bind to Atm1p. Studies with copper have shown that all cellular copper is protein-bound (Pufahl et al., 1997), and that its intracellular distribution is mediated by metal binding chaperone proteins (Glerum et al., 1996; Culotta et al., 1997; Lin et al., 1997). We would predict that all intracellular iron is also proteinbound, and that some type of iron chaperone would mediate delivery of iron to Atm1p. No iron binding chaperones have been identi®ed. However, recently Yfh1p was demonstrated to bind iron, and proposed to serve a physiological role in mitochondrial iron sequestration and bioavailability (Adamec et al., 2000). The ability of Yfh1p to bind iron, and to facil-
N.D. Schueck et al. / Mitochondrion 1 (2001) 51±60
itate mitochondrial iron export (Radisky et al., 1999), is consistent with a role for Yfh1p in the delivery of iron to Atm1p; however, a speci®c interaction between Atm1p and Yfh1p has not been demonstrated. The level of Yfh1p is decreased four-fold in atm1 cells. The lack of effect of either iron starvation or the AFT1-1 up allele on YFH1 expression suggests that an alternative regulatory signal must be involved in its control. The ability of Yfh1p to bind and sequester iron predicts that decreasing its level of expression would result in more iron being available for utilization within the mitochondrion. In this regard, the decreased expression of YFH1 in atm1-de®cient cells is reminiscent of the decreased production of ferritin seen in mammalian cells when iron-depleted. The mechanism for the decreased expression of YFH1 is unknown. Human homologs of both Atm1p (ABC7) and Yfh1p (Frataxin) have been associated with inherited disease (Babcock et al., 1997; Allikmets et al., 1999). A conservation of function between the yeast and human proteins has also been demonstrated (Wilson and Roof, 1997; Csere et al., 1998; Allikmets et al., 1999), including the accumulation of mitochondrial iron in affected patients. The observation that mutations in a growing number of yeast genes result in simultaneous mitochondrial iron accumulation and cytosolic iron starvation suggests that abnormal cytosolic iron metabolism may also have a role in the pathophysiology of Friedreich's ataxia and X-linked sideroblastic anemia and ataxia. Whether the mitochondrial iron accumulation resulting from mutations in ABC7 and Frataxin is associated with disturbances of cytosolic iron metabolism has not yet been determined. Acknowledgements We thank D. Eide, A. Dancis and G. Isaya for sharing reagents, and J. Kaplan for reagents and advice on measurement of iron uptake. This work was supported by National Institutes of Health Grant HD08315. References Adamec, J., Rusnak, F., Owen, W.G., et al., 2000. Iron-dependent
59
self-assembly of recombinant yeast frataxin: implications for Friedreich ataxia. Am. J. Hum. Genet. 67, 549±562. Allikmets, R., Raskind, W.H., Hutchinson, A., Schueck, N.D., Dean, M., Koeller, D.M., 1999. Mutation of a putative mitochondrial iron transporter gene (ABC7) in X- linked sideroblastic anemia and ataxia (XLSA/A). Hum. Mol. Genet. 8, 743±749. Askwith, C., Eide, D., Van Ho, A., et al., 1994. The FET3 gene of S. cerevisiae encodes a multicopper oxidase required for ferrous iron uptake. Cell 76, 403±410. Babcock, M., de Silva, D., Oaks, R., et al., 1997. Regulation of mitochondrial iron accumulation by Yfh1p, a putative homolog of frataxin. Science 276, 1709±1712. Bottomley, S.S., 1993. Wintrobes clinical hematology, Vol. 1. In: Lee, G.R., Bithell, T.C., Foerster, J., Athens, J.W., Lukens, J.N. (Eds.), Wintrobes Clinical Hematology, Vol. 1. Lea & Febiger, Philadelphia, PA, pp. 852±871. Branda, S.S., Cavadini, P., Adamec, J., Kalousek, F., Taroni, F., Isaya, G., 1999. Yeast and human frataxin are processed to mature form in two sequential steps by the mitochondrial processing peptidase. J. Biol. Chem. 274, 22763±22769. Campuzano, V., Montermini, L., Molto, M.D., et al., 1996. Friedreich's ataxia: autosomal recessive disease caused by an intronic GAA triplet repeat expansion. Science 271, 1423±1427. Casas, C., Aldea, M., Espinet, C., Gallego, C., Gil, R., Herrero, E., 1997. The AFT1 transcriptional factor is differentially required for expression of high-af®nity iron uptake genes in Saccharomyces cerevisiae. Yeast 13, 621±637. Csere, P., Lill, R., Kispal, G., 1998. Identi®cation of a human mitochondrial ABC transporter, the functional orthologue of yeast Atm1p. FEBS Lett. 441, 266±270. Culotta, V.C., Klomp, L.W., Strain, J., Casareno, R.L., Krems, B., Gitlin, J.D., 1997. The copper chaperone for superoxide dismutase. J. Biol. Chem. 272, 23469±23472. Dancis, A., Yuan, D.S., Haile, D., et al., 1994. Molecular characterization of a copper transport protein in S. cerevisiae: an unexpected role for copper in iron transport. Cell 76, 393±402. Dean, M., Allikmets, R., 1995. Evolution of ATP-binding cassette transporter genes. Curr. Opin. Genet. Dev. 5, 779±785. Dix, D.R., Bridgham, J.T., Broderius, M.A., Byersdorfer, C.A., Eide, D.J., 1994. The FET4 gene encodes the low af®nity Fe(II) transport protein of Saccharomyces cerevisiae. J. Biol. Chem. 269, 26092±26099. Dix, D., Bridgham, J., Broderius, M., Eide, D., 1997. Characterization of the FET4 protein of yeast. Evidence for a direct role in the transport of iron. J. Biol. Chem. 272, 11770±11777. Eide, D.J., 1998. The molecular biology of metal ion transport in Saccharomyces cerevisiae. Annu. Rev. Nutr. 18, 441±469. Eide, D., Guarente, L., 1992. Increased dosage of a transcriptional activator gene enhances iron-limited growth of Saccharomyces cerevisiae. J. Gen. Microbiol. 138, 347±354. Foury, F., Cazzalini, O., 1997. Deletion of the yeast homologue of the human gene associated with Friedreich's ataxia elicits iron accumulation in mitochondria. FEBS Lett. 411, 373±377. Glerum, D.M., Shtanko, A., Tzagoloff, A., 1996. Characterization of COX17, a yeast gene involved in copper metabolism and assembly of cytochrome oxidase. J. Biol. Chem. 271, 14504± 14509.
60
N.D. Schueck et al. / Mitochondrion 1 (2001) 51±60
Harlow, E., Lane, D., 1988. Antibodies: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Holland, I.B., Blight, M.A., 1999. ABC-ATPases, adaptable energy generators fuelling transmembrane movement of a variety of molecules in organisms from bacteria to humans. J. Mol. Biol. 293, 381±399. Kispal, G., Csere, P., Guiard, B., Lill, R., 1997. The ABC transporter Atm1p is required for mitochondrial iron homeostasis. FEBS Lett. 418, 346±350. Kispal, G., Csere, P., Prohl, C., Lill, R., 1999. The mitochondrial proteins Atm1p and Nfs1p are essential for biogenesis of cytosolic Fe/S proteins. EMBO J. 18, 3981±3989. Knight, S.A., Labbe, S., Kwon, L.F., Kosman, D.J., Thiele, D.J., 1996. A widespread transposable element masks expression of a yeast copper transport gene. Genes Dev. 10, 1917±1929. Koutnikova, H., Campuzano, V., Foury, F., Dolle, P., Cazzalini, O., Koenig, M., 1997. Studies of human, mouse and yeast homologues indicate a mitochondrial function for frataxin. Nat. Genet. 16, 345±351. Lamarche, J.B., Cote, M., Lemieux, B., 1980. The cardiomyopathy of Friedreich's ataxia morphological observations in 3 cases. Can. J. Neurol. Sci. 7, 389±396. Leighton, J., Schatz, G., 1995. An ABC transporter in the mitochondrial inner membrane is required for normal growth of yeast. EMBO J. 14, 188±195. Li, L., Kaplan, J., 1998. Defects in the yeast high af®nity iron transport system result in increased metal sensitivity because of the increased expression of transporters with a broad transition metal speci®city. J. Biol. Chem. 273, 22181±22187. Li, J., Kogan, M., Knight, S.A., Pain, D., Dancis, A., 1999. Yeast mitochondrial protein, Nfs1p, coordinately regulates iron-sulfur cluster proteins, cellular iron uptake, and iron distribution. J. Biol. Chem. 274, 33025±33034. Lin, S.J., Pufahl, R.A., Dancis, A., O'Halloran, T.V., Culotta, V.C., 1997. A role for the Saccharomyces cerevisiae ATX1 gene in copper traf®cking and iron transport. J. Biol. Chem. 272, 9215± 9220.
Ponka, P., 1997. Tissue-speci®c regulation of iron metabolism and heme synthesis: distinct control mechanisms in erythroid cells [see comments]. Blood 89, 1±25. Pufahl, R.A., Singer, C.P., Peariso, S.-J.L., et al., 1997. Metal ion chaperone function of the soluble Cu(I) receptor Atx1. Science 278, 853±856. Radisky, D., Kaplan, J., 1999. Regulation of Transition Metal Transport across the Yeast Plasma Membrane. J. Biol. Chem. 274, 4481±4484. Radisky, D.C., Babcock, M.C., Kaplan, J., 1999. The yeast frataxin homologue mediates mitochondrial iron ef¯ux. Evidence for a mitochondrial iron cycle. J. Biol. Chem. 274, 4497±4499. Rotig, A., de Lonlay, P., Chretien, D., et al., 1997. Aconitase and mitochondrial iron-sulphur protein de®ciency in Friedreich ataxia. Nat. Genet. 17, 215±217. Sambrook, J., Fritsch, E.F., Maniatis, T., 1989. Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Sherman, F., 1991. Getting started with yeast. Methods Enzymol. 194, 3±21. Wilson, R.B., Roof, D.M., 1997. Respiratory de®ciency due to loss of mitochondrial DNA in yeast lacking the frataxin homologue. Nat. Genet. 16, 352±357. Yaffe, M.P., 1991. Analysis of mitochondrial function and assembly. Methods Enzymol. 194, 627±643. Yamaguchi-Iwai, Y., Dancis, A., Klausner, R.D., 1995. AFT1: a mediator of iron regulated transcriptional control in Saccharomyces cerevisiae. EMBO J. 14, 1231±1239. Yamaguchi-Iwai, Y., Stearman, R., Dancis, A., Klausner, R.D., 1996. Iron-regulated DNA binding by the Aft1 protein controls the iron regulon in yeast. EMBO J. 15, 3377±3384. Zhao, H., Eide, D., 1996a. The yeast ZRT1 gene encodes the zinc transporter protein of a high- af®nity uptake system induced by zinc limitation. Proc. Natl. Acad. Sci. USA 93, 2454±2458. Zhao, H., Eide, D., 1996b. The ZRT2 gene encodes the low af®nity zinc transporter in Saccharomyces cerevisiae. J. Biol. Chem. 271, 23203±23210.
www.elsevier.com/locate/mito
The role of the mitochondrion in cellular iron homeostasis Nichole D. Schueck a,1, Michael Woontner a,1, David M. Koeller b,* a
Department of Pediatrics, University of Colorado Health Sciences Center, Denver, CO 80262, USA b Department of Pediatrics, Oregon Health Sciences University, Portland, OR 97201, USA
Received 12 December 2000; received in revised form 13 February 2001; accepted 16 February 2001
Abstract The yeast ATM1 protein is essential for normal mitochondrial iron homeostasis. Deletion of ATM1 results in mitochondrial iron accumulation and oxidative mitochondrial damage. Mutations in ABC7, the human homolog of ATM1, result in X-linked sideroblastic anemia and ataxia. Here we show that a deletion of ATM1 also has effects on extra-mitochondrial iron metabolism. ATM1-de®cient cells have an increased iron requirement for growth. When grown in iron-rich medium, mutant cells accumulate excess mitochondrial iron and have increased expression of the genes required for both high and low af®nity iron uptake. Thus, ATM1 mutant cells simultaneously demonstrate features of both iron overload and iron starvation. Yfh1p is the yeast homolog of the human frataxin protein, which is de®cient in Friedreich's ataxia. As in atm1 cells, a yfh1 deletion results in both mitochondrial iron accumulation and cytosolic iron starvation. In spite of their apparent roles in cellular iron homeostasis, we ®nd that the expression of neither ATM1 nor YFH1 is responsive to cellular iron status. Based on these observations, we propose a model in which cellular iron is prioritized for use by the mitochondrion, and available to the remainder of the cell only after mitochondrial needs have been ful®lled. q 2001 Elsevier Science B.V. and Mitochondria Research Society. All rights reserved. Keywords: Mitochondrion; Iron metabolism; Ataxia
1. Introduction Pathways for both high and low af®nity cellular iron uptake in yeast have been well characterized (reviewed in Eide, 1998). However, the mechanisms for the intracellular distribution of imported iron remain enigmatic. An essential requirement for effective intracellular iron traf®cking is the ability to prioritize iron delivery to the multiple sites within the cell where it is required, including the endoplasmic reticulum, mitochondria and peroxisomes. A means to minimize toxicity resulting from the iron-catalyzed * Corresponding author. Tel.: 11-503-494-2783; fax: 11-503494-2781. E-mail address: [email protected] (D.M. Koeller). 1 These authors contributed equally to this project.
formation of oxygen radical species is also imperative. A signi®cant fraction of the cell's iron requirement is for the assembly of iron sulfur clusters and heme biosynthesis in the mitochondrion. Because of its central role in cellular metabolism, the assurance of mitochondrial iron suf®ciency is predicted to have a high priority within any intracellular iron distribution network. However, proteins functioning in the import of iron to the mitochondrion have not been identi®ed. The ATM1 and YFH1 genes of Saccharomyces cerevisiae have roles in mitochondrial iron export (Allikmets et al., 1999; Radisky et al., 1999). Disruption of the ATM1 gene results in the accumulation of a 30fold excess of mitochondrial iron, loss of mitochondrial cytochromes, oxidative damage to the mitochon-
1567-7249/01/$20.00 q 2001 Elsevier Science B.V. and Mitochondria Research Society. All rights reserved. PII: S 1567-724 9(01)00004-6
52
N.D. Schueck et al. / Mitochondrion 1 (2001) 51±60
drial DNA, and decreased levels of cytosolic heme proteins (Leighton and Schatz, 1995; Kispal et al., 1997). Atm1p is a member of the ATP-binding cassette (ABC) superfamily of transporters. This family of transporters consists of a large number of evolutionarily related proteins involved in energydependent transport of substrates across cellular membranes (Dean and Allikmets, 1995). Atm1p is located in the mitochondrial inner membrane with its carboxyl terminus in the matrix (Leighton and Schatz, 1995). Based on this orientation it is predicted to export substrate from the matrix to the intermembranous space. The phenotype of cells with a deletion of the YFH1 gene is very similar to that resulting from disruption of ATM1. yfh1 cells accumulate a ten-fold excess of mitochondrial iron and become respiratory-de®cient secondary to oxidative damage to the mitochondrial DNA (Babcock et al., 1997; Foury and Cazzalini, 1997; Koutnikova et al., 1997; Wilson and Roof, 1997). Yfh1p has been localized to the mitochondrion (Babcock et al., 1997), where it is proposed to function as a regulator of mitochondrial iron export (Radisky et al., 1999). The similarity of the phenotypes resulting from ATM1 and YFH1 deletions has led to the hypothesis that Atm1p is a mitochondrial iron exporter, which is regulated by Yfh1p (Allikmets et al., 1999; Radisky et al., 1999). Homologs of the yeast ATM1 and YFH1 proteins are associated with inherited disease in man (Campuzano et al., 1996; Allikmets et al., 1999). A mutation in ABC7, the human homolog of ATM1, has been identi®ed as the cause of X-linked sideroblastic anemia and ataxia (Allikmets et al., 1999). Sideroblastic anemias are characterized by iron accumulation in the mitochondria of bone marrow erythroid precursors (Bottomley, 1993). The similarity between the biochemical phenotypes of ATM1 and ABC7 mutations (i.e. mitochondrial iron accumulation), and the ability of ABC7 to complement a yeast ATM1 mutation (Allikmets et al., 1999; Csere et al., 1998), indicate that these proteins perform a similar function in the mitochondrial inner membrane. Mutations in the gene for Frataxin, the human homolog of Yfh1p, result in Friedreich's ataxia, a progressive neurodegenerative disease characterized by gait ataxia, cardiomyopathy and diabetes (Campuzano et al., 1996). The observation that a deletion of the YFH1 gene
(YFH, Yeast Frataxin Homolog) is associated with mitochondrial iron accumulation and oxidative injury in yeast was the ®rst clue as to the possible function of frataxin (Babcock et al., 1997). Loss of the frataxin protein in humans is also associated with mitochondrial iron deposition and oxidative injury (Lamarche et al., 1980; Rotig et al., 1997). Prior studies have focused primarily on the effect of an ATM1 deletion on mitochondrial function. In this report we describe the effect of a deletion of ATM1 on extra-mitochondrial metal metabolism. We also present data on the in¯uence of cytosolic and mitochondrial iron status on ATM1 and YFH1 expression. Based on these experiments we propose a model in which iron entering the cell is prioritized ®rst for import into the mitochondrion, and only available to the remainder of the cell when mitochondrial needs have been ful®lled.
2. Materials and methods 2.1. Strains and culture methods The yeast strains were as follows: DY150 (MATa ura3-52 leu2-3,112 his3-11,15 trp1-1 ade2-1 can1100(oc)) and DY1457 (MATa ura3-52 leu2-3,112 his3-11,15 trp1-1 ade6 can1-100(oc)) (Askwith et al., 1994). AFT1 CM3260 (MATa trp1-63 leu23,112 gcn4-101 his3-609 ura3-52), and AFT11 up M2p (MATa trp1-63 leu2-3,112 gcn4-101 his3-609 ura3-52 AFT1-1 up) (Yamaguchi-Iwai et al., 1995). For growth in iron-rich medium, cells were cultured at 308C in 1% yeast extract, 2% peptone, and 2% glucose (YPD). For iron starvation, bathophenanthroline sulfonate (BPS) was added to 100 mM. Synthetic complete medium (SC) is 0.67% yeast nitrogen base with amino acids and 2% glucose. Low iron medium (LIM) is SC 1 1 mM EDTA and 20 mM citrate (Eide and Guarente, 1992). Media for assessment of transition metal sensitivity were made by addition of metal salts from stock solutions (ZnSO4, CuSO4, CoCl2, and MnSO4) to SC. Media were inoculated with a 1:100 dilution of cultures grown to saturation in YPD and analyzed for growth after 16±22 h. DNA transformations and manipulations of Escherichia coli and Saccharomyces cerevi-
N.D. Schueck et al. / Mitochondrion 1 (2001) 51±60
siae were done by standard techniques (Sambrook et al., 1989; Sherman, 1991). 2.2. Disruption of the ATM1 gene A disruption allele was made by inserting the LEU2 gene into a unique NheI site in the ATM1 gene and transformed into the diploid formed by mating DY150 £ DY1457. Transformed cells were selected for leucine prototrophy, and screened for disruption of the ATM1 gene by PCR and Southern blotting. Sporulation and dissection of tetrads from the heterozygous diploid yielded the haploid strain YNS1-5D (MATa ade2 ura3-52 his3-11,15 trp1-1 can1100(oc) leu2-3,112 atm1::LEU2). Sporulation of diploid strains and dissection of tetrads were as described (Sherman, 1991). 2.3. High af®nity Fe(II) uptake Whole cell iron uptake was measured by a modi®cation of the procedure of Eide and Guarente (1992). Cells were grown in YPD or YPD 1 100 mM BPS, harvested by centrifugation, washed twice with icecold 10 mM MES (pH 6.1), resuspended in 10 mM MES (pH 6.1)/2% glucose, and counted with a hemacytometer. All manipulations were done on ice or at 48C. Five million (DY150) or one million (YNS1-5D) cells were assayed in a total volume of 0.5 ml in 10 mM MES (pH 6.1), 2% glucose, 1 mM sodium ascorbate, and 1.0 mM 55FeCl3. Uptake was performed in duplicate at both 0 and 308C for 10 min. Cells were collected by ®ltration through Whatman GF/C ®lters, washed twice with ice-cold SSW (1 mM EDTA, 1 mM CaCl2, 20 mM sodium citrate (pH 4.2), 5 mM MgSO4, 1 mM KH2PO4, and 1 mM NaCl), air-dried and counted in a scintillation counter. The rate of uptake at 308C was calculated after subtraction of the non-speci®c cell associated radioactivity determined from the samples incubated at 08C. 2.4. Northern blot analysis RNA was isolated by glass bead lysis in the presence of phenol from log phase cultures grown in YPD. For each sample, 10 mg of total RNA was separated on a formaldehyde agarose gel and transferred to a nylon membrane by electroblotting. Probes for ATM1, YFH1, CTR1, FET3, FET4, and ZRT1 were
53 32
produced by random-primed a- P labeling using the RadPrime DNA Labeling System (Gibco BRL). For 18S rRNA a 32P end-labeled oligonucleotide probe was used. Hybridization signals were quantitated on a Molecular Dynamics Storm phosphorimager. mRNA signals were normalized to 18S rRNA signals. 2.5. Immunoblot analysis Mitochondria were isolated by differential centrifugation (Yaffe, 1991) and equal amounts of protein were separated by SDS-PAGE and transferred to a polyvinylidene ¯uoride (PVDF) membrane by electroblotting. Western blot analysis (Harlow and Lane, 1988) was performed with a rabbit polyclonal antibody generated against Yfh1p (Branda et al., 1999) and horseradish peroxidase-conjugated goat antirabbit IgG (Cappel). 3. Results A lack of the mitochondrial inner membrane ABC transporter Atm1p results in mitochondrial iron overload (Kispal et al., 1997; Csere et al., 1998). It is not known whether the excess iron in atm1 cells can be mobilized for utilization, or is metabolically inert. We hypothesized that atm1 cells would have increased tolerance to iron starvation if they are able to mobilize the accumulated mitochondrial iron. The ability to grow under iron-limited conditions was determined in LIM supplemented with varying amounts of iron. The growth of both wild-type (DY150) and atm1 cells is decreased in LIM with no added iron when compared to cultures grown with 500 mM FeCl3 (Fig. 1). Addition of as little as 5 mM FeCl3 restored the growth of wild-type cells to a level nearly equal to that in the high iron medium (500 mM FeCl3). In contrast, addition of as much as 20 mM FeCl3 failed to improve the growth of the atm1 mutant cells (Fig. 1), demonstrating an increased iron requirement for growth. These results suggest that the excess mitochondrial iron in atm1 cells is unavailable for utilization by the rest of the cell. Having determined that atm1 cells have an increased iron requirement for growth when measured under iron-limiting conditions, we evaluated the effect of an atm1 mutation on the iron status of cells grown
54
N.D. Schueck et al. / Mitochondrion 1 (2001) 51±60
Fig. 1. Effect of iron availability on the growth of wild-type and atm1 mutant cells. Wild-type (closed circles) and atm1 mutant (open circles) cells were pre-cultured in SC and diluted into low iron media containing varying amounts of FeCl3. Each point represents the mean of two experiments, each done in duplicate. Error bars, ^S.E.M.
Fig. 2. Comparison of high af®nity iron uptake by wild-type and atm1 mutant cells. Wild-type (DY150) and atm1 mutant (YNS15D) cells were grown to log phase in YPD, harvested, washed, and assayed for iron uptake in the presence of 1.0 mM 55FeCl3 and 1 mM ascorbate. To induce iron starvation DY150 cells were grown to log phase in YPD, diluted into fresh YPD containing BPS (100 mM), and grown for an additional 4 h. Data represent the mean of three experiments. Error bars, ^S.E.M.
in iron-rich medium. High af®nity cellular iron uptake is mediated by Ftr1p and Fet3p (Eide, 1998; Radisky and Kaplan, 1999). The expression of these genes is under the transcriptional control of Aft1p, a transcription factor whose activity is responsive to cellular iron levels (Yamaguchi-Iwai et al., 1995, 1996). Under conditions of iron starvation, Aft1p acts as a transcriptional activator to induce high levels of FET3 and FTR1 mRNA synthesis, leading to an increased rate of iron uptake. Measurement of high af®nity cellular iron uptake can therefore be used to assess cellular iron status. Wild-type cells grown in YPD made iron-de®cient by the addition of 100 mM BPS have a three-fold increase in high af®nity iron uptake (Fig. 2), and a corresponding increase in FET3 mRNA levels (Fig. 3), when compared to cells grown in YPD alone. The rate of iron uptake in atm1 cells grown in iron-rich medium (YPD) is similar to that of wild-type cells grown under conditions of iron starvation (Fig. 2). Measurement of FET3 mRNA levels in atm1 cells shows a 12-fold increase relative to wildtype cells grown in YPD (Fig. 3). The level of FET3 expression in atm1 cells is comparable to that seen in cells with a constitutively active allele of AFT1 (AFT1-1 up) (Fig. 3) (Yamaguchi-Iwai et al., 1995).
Fig. 3. Northern blot analysis of FET3 mRNA. Total RNA was isolated from log phase cultures of atm1 mutant (atm1), wild-type (DY150), AFT1-1 up and its wild-type parent (AFT1) grown in YPD, or YPD containing 100 mM BPS (DY150 1 BPS) and probed for FET3 and 18S rRNA. Quantitation of RNA was done with a phosphorimager and relative levels of FET3 mRNA calculated after normalization to 18S rRNA.
N.D. Schueck et al. / Mitochondrion 1 (2001) 51±60
Thus, when grown under conditions that result in mitochondrial iron accumulation (Csere et al., 1998; Kispal et al., 1997), atm1 mutant cells are sensing iron starvation. In addition to iron, normal mitochondrial function requires zinc for metalloproteases, manganese for Mn-superoxide dismutase, and copper for cytochrome oxidase. To determine the consequences of an atm1 mutation on the metabolism of other transition metals, we measured the growth of atm1 cells in media supplemented with varying levels of Cu, Zn, Co, and Mn. In comparison to wild-type, atm1 mutant cells are extremely sensitive to all four of these metals (Fig. 4). To determine whether the observed metal sensitivity was due to abnormal regulation of plasma membrane transporters, we measured the expression of genes involved in Cu and Zn uptake. Two plasma membrane transporters, Ctr1p and Ctr3p, mediate
55
high af®nity cellular copper uptake (Dancis et al., 1994; Knight et al., 1996). Both high and low af®nity transporters (Zrt1p and Zrt2p, respectively) function in zinc import (Zhao and Eide, 1996a,b). Northern blot analysis of CTR1 and ZRT1 mRNA levels showed less than a two-fold increase in atm1 mutant cells (Fig. 5A,B). This small increase in CTR1 and ZRT1 expression is comparable to that seen in wild-type cells grown in iron-limiting conditions, and seems unlikely to explain the extreme sensitivity of atm1 cells to these metals. An alternative hypothesis for the metal sensitivity of atm1 cells is an increased expression of the FET4 gene. Fet4p is a component of the low af®nity plasma membrane iron uptake mechanism, and is also capable of transporting several additional transition metals (Dix et al., 1994; Eide, 1998). The lack of iron speci®city of Fet4p results in generalized metal sensitivity
Fig. 4. Comparison of the effect of transition metals on the growth of wild-type and atm1 mutant cells. Wild-type (®lled diamonds) and atm1 mutant (open squares) cells were pre-cultured in YPD and then diluted into SC containing varying amounts of copper (A), zinc (B), cobalt (C), or manganese (D).
56
N.D. Schueck et al. / Mitochondrion 1 (2001) 51±60
Fig. 5. Northern blot analysis of copper, zinc, and iron transporter mRNA expression in wild-type and atm1 mutant cells. Total RNA was isolated from log phase cultures of atm1 mutant (atm1) and wild-type (DY150) cells grown in YPD, or YPD containing 100 mM BPS (DY150 1 BPS) and probed for CTR1 (A), ZRT1 (B), FET4 (C) and 18S rRNA. Quantitation of RNAs was done with a phosphorimager and relative levels of speci®c mRNAs calculated after normalization to 18S rRNA.
in cells lacking a high af®nity iron uptake pathway and therefore dependent on Fet4p for iron uptake (Li and Kaplan, 1998). Growth of wild-type cells in iron-limited medium has little effect on FET4 mRNA levels (Fig. 5C). In contrast, atm1 cells show a ®vefold increase in FET4 expression, supporting the hypothesis that elevated expression of FET4 is responsible for the metal sensitivity of atm1 mutant cells. These data also demonstrate that an atm1 mutation results in the induction of both the high and low af®nity iron transport pathways. In contrast, growth of wild-type cells in iron-de®cient medium only induces the genes required for high af®nity iron uptake (Figs. 3 and 5), indicating a unique character to the iron starvation resulting from an atm1 mutation. The dramatic effect of a mutation in the ATM1 gene on cellular iron homeostasis raises the question as to the effect of cellular iron status on ATM1 expression. Northern blot analysis shows that in wild-type cells, iron starvation induced by growth in YPD 1 BPS does not affect the level of ATM1 mRNA (Fig. 6). Similarly, there is only a slight change in the ATM1 mRNA level in an AFT1-1 up strain. Consistent with the lack of iron-dependent regulation there are no consensus Aft1p binding sites (Yamaguchi-Iwai et al., 1996) in the ATM1 promoter. Thus, the expression of ATM1 is not sensitive to changes in cellular iron status induced by conditions known to effect the expression of other iron-regulated genes. The apparent lack of iron-dependent regulation of ATM1 expression is surprising given the necessity of
Atm1p for normal cellular iron homeostasis. The localization of Atm1p to the mitochondrial inner membrane suggests that a signal derived from the mitochondrion may have a role in regulating its expression; however, no such signals are currently known. As an alternative, we chose to look at the affect of a yfh1 mutation on ATM1 expression. The YFH1 protein functions in the regulation of mitochondrial iron ef¯ux (Radisky et al., 1999); a lack of func-
Fig. 6. Northern blot analysis of ATM1 mRNA levels. Total RNA was isolated from log phase cultures of wild-type (DY150), AFT11 up and its wild-type parent (AFT1) grown in YPD, or YPD containing 100 mM BPS (DY150 1 BPS) and probed for ATM1 and 18S rRNA. Quantitation of RNA was done with a phosphorimager and relative levels of ATM1 mRNA calculated after normalization to 18S rRNA.
N.D. Schueck et al. / Mitochondrion 1 (2001) 51±60
tional Yfh1p results in mitochondrial iron accumulation (Babcock et al., 1997; Foury and Cazzalini, 1997). Northern blot analysis revealed no difference in ATM1 mRNA levels in yfh1 cells as compared to wild-type (data not shown). In contrast, the level of YFH1 mRNA and protein is decreased four-fold in atm1 mutant cells (Fig. 7). This result suggested the possibility that the iron accumulation seen in atm1 cells was secondary to a de®ciency of Yfh1p, as seen in cells with an ssq1 mutation (Knight et al., 1996). Expression of YFH1 on a high copy plasmid using either its own promoter or a heterologous promoter had no affect on the phenotype of atm1 cells (data not shown). However, overexpression of yfh1p in these experiments was not con®rmed by Western blotting. We did not see any changes in YFH1 expression in cells grown under iron-limiting conditions, or in the presence of the AFT1-1 up allele (Fig. 7), indicating that a signal other than cytosolic iron starvation is responsible for the observed decrease in YFH1 expression.
57
4. Discussion We have shown that in addition to abnormal mitochondrial iron homeostasis, atm1 cells have anomalous cytosolic iron metabolism. A deletion of ATM1 results in an increased sensitivity to iron starvation. In iron-rich medium atm1 cells accumulate toxic levels of mitochondrial iron and simultaneously show an iron-starved pattern of nuclear gene expression. Thus, deletion of ATM1 is affecting the cell's ability to ensure that suf®cient iron is available for both mitochondrial and extra-mitochondrial needs, suggesting that Atm1p plays a role in the intracellular distribution of iron. The orientation of Atm1p in the mitochondrial inner membrane predicts that it functions as an exporter (Leighton and Schatz, 1995). The accretion of iron by mitochondria in atm1 cells suggests that the substrate for Atm1p is iron, for which an export pathway has been demonstrated (Radisky et al., 1999). Because of the affect of an atm1 mutation on the cytosolic Fe-S enzyme isopropyl malate isomerase
Fig. 7. Expression of YFH1 in atm1 mutant cells. (A) Total RNA was isolated from log phase cultures of atm1 mutant (atm1), wild-type (DY150), AFT1-1 up and its wild-type parent (AFT1) grown in YPD, or YPD containing 100 mM BPS (DY150 1 BPS) and probed for YFH1 and 18S rRNA. Quantitation of RNA was done with a phosphorimager and relative levels of YFH1 mRNA calculated after normalization to 18S rRNA. (B) Equal amounts of mitochondrial protein from wild-type (DY150) and atm1 mutant cells (atm1) were separated by SDS-PAGE, transferred to a PVDF membrane, and analyzed by Western blotting with a polyclonal rabbit anti-Yfh1p antibody. Proteins were visualized by ECL using a peroxidase-conjugated goat anti-rabbit secondary antibody.
58
N.D. Schueck et al. / Mitochondrion 1 (2001) 51±60
(Leu1p), it has been proposed that the function of Atm1p is to export Fe-S clusters assembled in the mitochondrion (Kispal et al., 1999). However, our data demonstrate that in addition to an inability to form cytosolic iron sulfur clusters, atm1 cells have a generalized cytosolic de®ciency of iron, as demonstrated by an increased expression of the genes required for high af®nity iron uptake. Simultaneous mitochondrial iron accumulation and cytosolic iron starvation has previously been demonstrated in cells with a deletion of NFS1, a cysteine desulfurase that provides sul®de for mitochondrial iron sulfur cluster formation (Li et al., 1999), or a deletion of ISU1, which is believed to function in the mobilization of iron for mitochondrial Fe-S cluster formation (unpublished data). Deletion of the YFH1 gene also results in mitochondrial iron overload and cytosolic iron starvation, as demonstrated by increased FET3 expression (Babcock et al., 1997). Using an allele of YFH1 controlled by an inducible promoter, Radisky et al. (1999) showed that in cells grown under conditions that repress YFH1 expression, mitochondria accumulate iron and FET3 expression is high. When switched to permissive growth conditions, the appearance of Yfh1p is rapidly followed by a decrease in both mitochondrial iron content and FET3 mRNA levels. The decrease in the expression of FET3 occurs without any change in the iron content of the medium, indicating that the iron released from the mitochondrion is being made available to the rest of the cell. Based on the similar effects of ATM1, NFS1, ISU1 and YFH1 deletions on mitochondrial and cellular iron metabolism, we hypothesize that after entry into the cell, iron is prioritized for use by the mitochondrion, and is only available to the rest of the cell after mitochondrial needs have been ful®lled. A model in which imported iron is speci®cally targeted to the mitochondrion has previously been proposed for mammalian erythroid cells (Ponka, 1997). The nature of the mechanism for sensing mitochondrial iron status and regulation of iron export is unknown. Deletion of ATM1 also results in increased expression of the low af®nity iron importer FET4. Wild-type cells grown in iron-limited medium or expressing the AFT1-1 up allele have elevated expression of the genes for high af®nity iron uptake, but do not show signi®cant induction of FET4 (Fig. 5) (Casas et al., 1997). In contrast, in cells lacking high af®nity iron uptake due
to a FET3 deletion, FET4 expression is induced by iron starvation (Dix et al., 1997). The increased sensitivity to transition metal toxicity of atm1 cells is also seen in cells de®cient in high af®nity iron import, and results from the ability of other metals to compete with iron for import via Fet4p (Dix et al., 1994; Li and Kaplan, 1998). The requirement of atm1 cells for greater than 20 mM iron for growth, which is greater than 100 times the KM of the high af®nity import pathway, further supports their dependence on iron imported via Fet4p (K M 30 mM). Thus, ATM1 mutant cells have phenotypic features similar to cells lacking high af®nity iron. Interestingly, deletion of FET3 prevents the accumulation of excess mitochondrial iron in yfh1 cells (Radisky et al., 1999), raising the possibility that the route of entry (i.e. the high vs. low af®nity pathway) may in part determine where imported iron is delivered. Manipulations known to affect the expression of iron-regulated genes, including iron starvation and the AFT1-1 up allele, have only a small effect on the level of ATM1 mRNA. Elevation of mitochondrial iron secondary to a YFH1 deletion was also without effect on ATM1 expression. Alternatively, it may be the transport activity of Atm1p that is subject to regulation. The ATM1 protein is an ABC transporter. A general mechanism for the function of ABC transporters is that upon binding of substrate to the membrane spanning domain there is activation of the ATPase and subsequent transfer of the bound substrate across the membrane (Holland and Blight, 1999). Transport activity can also be regulated in response to the metabolic state of the cell and modi®cations such as phosphorylation. It is highly unlikely that there would be free iron in the mitochondrion that could bind to Atm1p. Studies with copper have shown that all cellular copper is protein-bound (Pufahl et al., 1997), and that its intracellular distribution is mediated by metal binding chaperone proteins (Glerum et al., 1996; Culotta et al., 1997; Lin et al., 1997). We would predict that all intracellular iron is also proteinbound, and that some type of iron chaperone would mediate delivery of iron to Atm1p. No iron binding chaperones have been identi®ed. However, recently Yfh1p was demonstrated to bind iron, and proposed to serve a physiological role in mitochondrial iron sequestration and bioavailability (Adamec et al., 2000). The ability of Yfh1p to bind iron, and to facil-
N.D. Schueck et al. / Mitochondrion 1 (2001) 51±60
itate mitochondrial iron export (Radisky et al., 1999), is consistent with a role for Yfh1p in the delivery of iron to Atm1p; however, a speci®c interaction between Atm1p and Yfh1p has not been demonstrated. The level of Yfh1p is decreased four-fold in atm1 cells. The lack of effect of either iron starvation or the AFT1-1 up allele on YFH1 expression suggests that an alternative regulatory signal must be involved in its control. The ability of Yfh1p to bind and sequester iron predicts that decreasing its level of expression would result in more iron being available for utilization within the mitochondrion. In this regard, the decreased expression of YFH1 in atm1-de®cient cells is reminiscent of the decreased production of ferritin seen in mammalian cells when iron-depleted. The mechanism for the decreased expression of YFH1 is unknown. Human homologs of both Atm1p (ABC7) and Yfh1p (Frataxin) have been associated with inherited disease (Babcock et al., 1997; Allikmets et al., 1999). A conservation of function between the yeast and human proteins has also been demonstrated (Wilson and Roof, 1997; Csere et al., 1998; Allikmets et al., 1999), including the accumulation of mitochondrial iron in affected patients. The observation that mutations in a growing number of yeast genes result in simultaneous mitochondrial iron accumulation and cytosolic iron starvation suggests that abnormal cytosolic iron metabolism may also have a role in the pathophysiology of Friedreich's ataxia and X-linked sideroblastic anemia and ataxia. Whether the mitochondrial iron accumulation resulting from mutations in ABC7 and Frataxin is associated with disturbances of cytosolic iron metabolism has not yet been determined. Acknowledgements We thank D. Eide, A. Dancis and G. Isaya for sharing reagents, and J. Kaplan for reagents and advice on measurement of iron uptake. This work was supported by National Institutes of Health Grant HD08315. References Adamec, J., Rusnak, F., Owen, W.G., et al., 2000. Iron-dependent
59
self-assembly of recombinant yeast frataxin: implications for Friedreich ataxia. Am. J. Hum. Genet. 67, 549±562. Allikmets, R., Raskind, W.H., Hutchinson, A., Schueck, N.D., Dean, M., Koeller, D.M., 1999. Mutation of a putative mitochondrial iron transporter gene (ABC7) in X- linked sideroblastic anemia and ataxia (XLSA/A). Hum. Mol. Genet. 8, 743±749. Askwith, C., Eide, D., Van Ho, A., et al., 1994. The FET3 gene of S. cerevisiae encodes a multicopper oxidase required for ferrous iron uptake. Cell 76, 403±410. Babcock, M., de Silva, D., Oaks, R., et al., 1997. Regulation of mitochondrial iron accumulation by Yfh1p, a putative homolog of frataxin. Science 276, 1709±1712. Bottomley, S.S., 1993. Wintrobes clinical hematology, Vol. 1. In: Lee, G.R., Bithell, T.C., Foerster, J., Athens, J.W., Lukens, J.N. (Eds.), Wintrobes Clinical Hematology, Vol. 1. Lea & Febiger, Philadelphia, PA, pp. 852±871. Branda, S.S., Cavadini, P., Adamec, J., Kalousek, F., Taroni, F., Isaya, G., 1999. Yeast and human frataxin are processed to mature form in two sequential steps by the mitochondrial processing peptidase. J. Biol. Chem. 274, 22763±22769. Campuzano, V., Montermini, L., Molto, M.D., et al., 1996. Friedreich's ataxia: autosomal recessive disease caused by an intronic GAA triplet repeat expansion. Science 271, 1423±1427. Casas, C., Aldea, M., Espinet, C., Gallego, C., Gil, R., Herrero, E., 1997. The AFT1 transcriptional factor is differentially required for expression of high-af®nity iron uptake genes in Saccharomyces cerevisiae. Yeast 13, 621±637. Csere, P., Lill, R., Kispal, G., 1998. Identi®cation of a human mitochondrial ABC transporter, the functional orthologue of yeast Atm1p. FEBS Lett. 441, 266±270. Culotta, V.C., Klomp, L.W., Strain, J., Casareno, R.L., Krems, B., Gitlin, J.D., 1997. The copper chaperone for superoxide dismutase. J. Biol. Chem. 272, 23469±23472. Dancis, A., Yuan, D.S., Haile, D., et al., 1994. Molecular characterization of a copper transport protein in S. cerevisiae: an unexpected role for copper in iron transport. Cell 76, 393±402. Dean, M., Allikmets, R., 1995. Evolution of ATP-binding cassette transporter genes. Curr. Opin. Genet. Dev. 5, 779±785. Dix, D.R., Bridgham, J.T., Broderius, M.A., Byersdorfer, C.A., Eide, D.J., 1994. The FET4 gene encodes the low af®nity Fe(II) transport protein of Saccharomyces cerevisiae. J. Biol. Chem. 269, 26092±26099. Dix, D., Bridgham, J., Broderius, M., Eide, D., 1997. Characterization of the FET4 protein of yeast. Evidence for a direct role in the transport of iron. J. Biol. Chem. 272, 11770±11777. Eide, D.J., 1998. The molecular biology of metal ion transport in Saccharomyces cerevisiae. Annu. Rev. Nutr. 18, 441±469. Eide, D., Guarente, L., 1992. Increased dosage of a transcriptional activator gene enhances iron-limited growth of Saccharomyces cerevisiae. J. Gen. Microbiol. 138, 347±354. Foury, F., Cazzalini, O., 1997. Deletion of the yeast homologue of the human gene associated with Friedreich's ataxia elicits iron accumulation in mitochondria. FEBS Lett. 411, 373±377. Glerum, D.M., Shtanko, A., Tzagoloff, A., 1996. Characterization of COX17, a yeast gene involved in copper metabolism and assembly of cytochrome oxidase. J. Biol. Chem. 271, 14504± 14509.
60
N.D. Schueck et al. / Mitochondrion 1 (2001) 51±60
Harlow, E., Lane, D., 1988. Antibodies: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Holland, I.B., Blight, M.A., 1999. ABC-ATPases, adaptable energy generators fuelling transmembrane movement of a variety of molecules in organisms from bacteria to humans. J. Mol. Biol. 293, 381±399. Kispal, G., Csere, P., Guiard, B., Lill, R., 1997. The ABC transporter Atm1p is required for mitochondrial iron homeostasis. FEBS Lett. 418, 346±350. Kispal, G., Csere, P., Prohl, C., Lill, R., 1999. The mitochondrial proteins Atm1p and Nfs1p are essential for biogenesis of cytosolic Fe/S proteins. EMBO J. 18, 3981±3989. Knight, S.A., Labbe, S., Kwon, L.F., Kosman, D.J., Thiele, D.J., 1996. A widespread transposable element masks expression of a yeast copper transport gene. Genes Dev. 10, 1917±1929. Koutnikova, H., Campuzano, V., Foury, F., Dolle, P., Cazzalini, O., Koenig, M., 1997. Studies of human, mouse and yeast homologues indicate a mitochondrial function for frataxin. Nat. Genet. 16, 345±351. Lamarche, J.B., Cote, M., Lemieux, B., 1980. The cardiomyopathy of Friedreich's ataxia morphological observations in 3 cases. Can. J. Neurol. Sci. 7, 389±396. Leighton, J., Schatz, G., 1995. An ABC transporter in the mitochondrial inner membrane is required for normal growth of yeast. EMBO J. 14, 188±195. Li, L., Kaplan, J., 1998. Defects in the yeast high af®nity iron transport system result in increased metal sensitivity because of the increased expression of transporters with a broad transition metal speci®city. J. Biol. Chem. 273, 22181±22187. Li, J., Kogan, M., Knight, S.A., Pain, D., Dancis, A., 1999. Yeast mitochondrial protein, Nfs1p, coordinately regulates iron-sulfur cluster proteins, cellular iron uptake, and iron distribution. J. Biol. Chem. 274, 33025±33034. Lin, S.J., Pufahl, R.A., Dancis, A., O'Halloran, T.V., Culotta, V.C., 1997. A role for the Saccharomyces cerevisiae ATX1 gene in copper traf®cking and iron transport. J. Biol. Chem. 272, 9215± 9220.
Ponka, P., 1997. Tissue-speci®c regulation of iron metabolism and heme synthesis: distinct control mechanisms in erythroid cells [see comments]. Blood 89, 1±25. Pufahl, R.A., Singer, C.P., Peariso, S.-J.L., et al., 1997. Metal ion chaperone function of the soluble Cu(I) receptor Atx1. Science 278, 853±856. Radisky, D., Kaplan, J., 1999. Regulation of Transition Metal Transport across the Yeast Plasma Membrane. J. Biol. Chem. 274, 4481±4484. Radisky, D.C., Babcock, M.C., Kaplan, J., 1999. The yeast frataxin homologue mediates mitochondrial iron ef¯ux. Evidence for a mitochondrial iron cycle. J. Biol. Chem. 274, 4497±4499. Rotig, A., de Lonlay, P., Chretien, D., et al., 1997. Aconitase and mitochondrial iron-sulphur protein de®ciency in Friedreich ataxia. Nat. Genet. 17, 215±217. Sambrook, J., Fritsch, E.F., Maniatis, T., 1989. Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Sherman, F., 1991. Getting started with yeast. Methods Enzymol. 194, 3±21. Wilson, R.B., Roof, D.M., 1997. Respiratory de®ciency due to loss of mitochondrial DNA in yeast lacking the frataxin homologue. Nat. Genet. 16, 352±357. Yaffe, M.P., 1991. Analysis of mitochondrial function and assembly. Methods Enzymol. 194, 627±643. Yamaguchi-Iwai, Y., Dancis, A., Klausner, R.D., 1995. AFT1: a mediator of iron regulated transcriptional control in Saccharomyces cerevisiae. EMBO J. 14, 1231±1239. Yamaguchi-Iwai, Y., Stearman, R., Dancis, A., Klausner, R.D., 1996. Iron-regulated DNA binding by the Aft1 protein controls the iron regulon in yeast. EMBO J. 15, 3377±3384. Zhao, H., Eide, D., 1996a. The yeast ZRT1 gene encodes the zinc transporter protein of a high- af®nity uptake system induced by zinc limitation. Proc. Natl. Acad. Sci. USA 93, 2454±2458. Zhao, H., Eide, D., 1996b. The ZRT2 gene encodes the low af®nity zinc transporter in Saccharomyces cerevisiae. J. Biol. Chem. 271, 23203±23210.