Advertisement

CCC1 Suppresses Mitochondrial Damage in the Yeast Model of Friedreich's Ataxia by Limiting Mitochondrial Iron Accumulation*

  • Opal S. Chen
    Affiliations
    Division of Immunology and Cell Biology, Department of Pathology, School of Medicine, University of Utah, Salt Lake City, Utah 84132
    Search for articles by this author
  • Jerry Kaplan
    Correspondence
    To whom correspondence should be addressed. Tel.: 801-581-7427; Fax: 801-581-4517
    Affiliations
    Division of Immunology and Cell Biology, Department of Pathology, School of Medicine, University of Utah, Salt Lake City, Utah 84132
    Search for articles by this author
  • Author Footnotes
    * This work was supported by NIDDK, National Institutes of Health Grant 52380.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Open AccessPublished:March 17, 2000DOI:https://doi.org/10.1074/jbc.275.11.7626
      Deletion of YFH1 inSaccharomyces cerevisiae leads to a loss of respiratory competence due to excessive mitochondrial iron accumulation. A suppressor screen identified a gene, CCC1, that maintained respiratory function in a Δyfh1 yeast strain regardless of extracellular iron concentration. CCC1 expression prevented excessive mitochondrial iron accumulation by limiting mitochondrial iron uptake rather than by increasing mitochondrial iron egress. Expression of CCC1 did not result in sequestration of iron in membranous compartments or cellular iron export.CCC1 expression in wild type cells resulted in increased expression of the high affinity iron transport system composed ofFET3 and FTR1, suggesting that intracellular iron is not sensed by the iron-dependent transcription factor Aft1p. Introduction of AFT1 up, a constitutive allele of the iron transcription factor, AFT1, that also leads to increased high affinity iron transport did not prevent Δyfh1 cells from becoming respiratory-incompetent. Although the mechanism by whichCCC1 expression affects cytosolic iron is not known, the data suggest that excessive mitochondrial iron accumulation only occurs when cytosolic free iron levels are high.
      Friedreich's ataxia is a lethal disorder affecting the nervous system and heart. The disorder is due to a triplet expansion in the first intron of the Frataxin gene that results in decreased levels of Frataxin mRNA (
      • Campuzano V.
      • Montermini L.
      • Lutz Y.
      • Cova L.
      • Hindelang C.
      • Jiralerspong S.
      • Trottier Y.
      • Kish S.J.
      • Faucheux B.
      • Trouillas P.
      • Authier F.J.
      • Durr A.
      • Mandel J.L.
      • Vescovi A.
      • Pandolfo M.
      • Koenig M.
      ), which encodes a mitochondrial protein (
      • Koutnikova H.
      • Campuzano V.
      • Foury F.
      • Dolle P.
      • Cazzalini O.
      • Koenig M.
      ). Insight into the function of Frataxin resulted from studies on theSaccharomyces cerevisiae gene, yeastFrataxin homologue (YFH1), which is an orthologue of mammalian Frataxin. In the absence of Yfh1p, iron accumulates within mitochondria, leading to a loss of respiratory activity through the generation of mitochondrial DNA mutations. It is thought that decreased respiratory activity is a consequence of iron-induced oxygen radicals (
      • Babcock M.
      • de Silva D.
      • Oaks R.
      • Davis-Kaplan S.
      • Jiralerspong S.
      • Montermini L.
      • Pandolfo M.
      • Kaplan J.
      ,
      • Wilson R.B.
      • Roof D.M.
      ). The observation of increased iron deposits in Friedreich's ataxia heart biopsies (
      • Lamarche J.B.
      • Cote M.
      • Lemieux B.
      ) and the finding of increased iron in mitochondria of Friedreich's ataxia fibroblasts (
      • Delatycki M.B.
      • Camakaris J.
      • Brooks H.
      • Evans-Whipp T.
      • Thorburn D.R.
      • Williamson R.
      • Forrest S.M.
      ,
      • Wong A.
      • Yang J.
      • Cavadini P.
      • Gellera C.
      • Lonnerdal B.
      • Taroni F.
      • Cortopassi G.
      ) indicate that the yeast model may be an accurate reflection of the pathophysiology of Friedreich's ataxia.
      We initiated a search for genes that could suppress the respiratory deficit in yeast lacking Yfh1p. Here we report that CCC1,previously characterized as a putative Golgi Ca2+/Mn2+ transporter (
      • Fu D.
      • Beeler T.
      • Dunn T.
      ,
      • Lapinskas P.J.
      • Lin S.J.
      • Culotta V.C.
      ), allowedΔyfh1 cells to maintain respiratory activity by preventing the toxic accumulation of mitochondrial iron. The restriction on mitochondrial iron accumulation resulted from reduced mitochondrial iron uptake rather than increased mitochondrial iron efflux. The observation that respiratory activity as well as leucine biosynthesis was unaffected in Δyfh1 cells overexpressingCCC1 implies that Yfh1p is not significantly involved in the biosynthesis, assembly or export of iron-sulfur clusters. Overexpression of CCC1 in wild type cells activated the iron-dependent transcription factor Aft1p, resulting in an increase in iron uptake and cytosolic iron accumulation. Activation of Aft1p suggests that cells sense that cytosolic iron is low, indicating that much of the cytosolic iron was not bioavailable. This reduced concentration of “free” iron did not affect the growth ofΔyfh1 cells but did prevent excessive mitochondrial iron accumulation. These studies demonstrate that toxic mitochondrial iron levels resulting from the loss of Yfh1p may only occur when cytosolic free iron levels are high.

      EXPERIMENTAL PROCEDURES

      Strains and Growth Media

      DY150 (MATa ura3–52, leu2–3, 112, trp1–1, his3–11, 15, ade2–1, can1–100(oc)), 1457 (MATα ura3–52, leu2–3, 112, trp1–1, his3–11, ade6, can1–100(oc)), Δaft1 (MATα trp1-Δ63, leu2–3, 112 gcn4–101, his3–609, ura3–52 aft1::LEU2,) and Δfet3 (MATα ura3–52, leu2–3, 112, trp1–1, ade2–1, can1–100(oc), fet3::HIS3) were all isogenic to the W303 strain of S. cerevisiae. METYFH1 was generated by crossing Δyfh1 (MATa ura3–52, leu2–3, 112, trp1–1, 15, ade2–1, can1–100(oc),yfh1::HIS3) strain with 1457, as described previously (
      • Radisky D.C.
      • Babcock M.C.
      • Kaplan J.
      ). The resulting diploid was transformed with a plasmid containing a methionine-regulated YFH1 in pTF63, an episomal plasmid bearing the URA3 marker (pMetYFH1). Expression of Yfh1p is regulated by the MET3 promoter, aHindIII/EcoRV fragment of m2265 clone containing the MET3 promoter. In the presence of methionine, theYFH1 transcription is suppressed, whereas in the absence of methionine, YFH1 is transcribed. The transformed diploid was sporulated and further dissected on synthetic medium lacking methionine and uracil. A haploid Δyfh1 strain with respiratory activity, maintained by the pMetYFH1, was identified and is designated METYFH1 (Δyfh1, pMetYFH1). The respiratory competence of METYFH1 was maintained by growth in medium lacking methionine and uracil. All growth media used in this study were described previously (
      • Radisky D.C.
      • Babcock M.C.
      • Kaplan J.
      ). The Escherichia coli strain used for transformation was DH5α (F-, endA1, hsd-R17(rk-, mk+), supE44, thi-1, λ-, recA1, gyrA96, relA1, Δ(laczya-argF)U169, ϕ80dlacZΔM15deoR).

      Bacteria and Yeast Transformation

      The yeast genomic library used for the suppressor screen was obtained from American Type Culture Collection ATCC), ATCC 37323 (S. cerevisiae genomic library in YEp13, an episomal plasmid DNA bearing LEU2 marker). The genomic library was amplified in DH5α. DNA transformations ofE. coli and S. cerevisiae were performed by standard procedures (
      • Sambrook J.
      • Fritsch E.F.
      • Maniatis T.
      ,
      • Sherman F.
      • Fink G.R.
      • Lawrence C.W.
      ).

      Suppressor Screen

      A genomic library was transformed into METYFH1, and 30,000 LEU +, URA + transformants were selected on plates lacking uracil, leucine, and methionine. The transformants were replicated to uracil-supplemented plates, followed by replica-plating to 5′-fluoroorotic acid plates to select against URA3 cells. Suppressor candidates were identified by their growth on glycerol-ethanol plates. To exclude false positives, cells that still carried a functional YFH1 gene, polymerase chain reaction using primers specific to 5′ and 3′ end of YFH1 was performed. The two primers were 5REV, 5′-CGA GAT ATC TAG AGT GTA GCA ATG ATT AA-3′, and 3NTS1, 5′-CCC GAG CTC TTA AGC GGC CGC ACC TCC TTG GCT TTT AGA AAT GGC CT-3′. Cells showing a polymerase chain reaction product at 2.2 kilobases, but not 0.55 kilobases, the size of an intactYFH1 gene, were chosen for further study. Plasmids were rescued from candidate clones by electroporation into DH5α. DNA sequence analysis of the rescued plasmids utilized primers specific to the region upstream and downstream of BamHI site of YEp13. The sequencing primers were 026, 5′-CCA CTA TCG ACT ACG CGA-3′ and 027- 5′-ATG TCG GCG ATA TCA GGC G-3′.
      Sequence analysis revealed a single open reading frame,CCC1, shared by four library clones. One of the library clones, pS222, contained a genomic 4823-base pair segment consisting ofCCC1 and two other open reading frames: YLR219w and YLR218c. From this library clone, CCC1 was subcloned by removing a 1923-base pair EagI fragment that contained the two other open reading frames in part (YLR219w) and in whole (YLR218c) and part of the TET gene (tetracycline-resistant gene) in the vector. The remaining large EagI fragment (13562 base pair) that encompassed the sequence of 2243 base pair upstream and 252 base pair down stream of CCC1 was gel-purified and religated to construct pSCCC1.

      Construction of pMetCCC1 and METCCC1

      A plasmid containing a methionine-regulated CCC1-FLAG gene (pMetCCC1) was constructed using the MET3 promoter. A carboxyl-terminal FLAG epitope (underlined)-tagged CCC1 was generated by the polymerase chain reaction. The two primers were 035, 5′-TCC CCC GGG CAC AAA TAT TAT GTC CAT AGC AC-3′, and 037, 5′-TCC CCG CGG TTA CTT GTC ATC GTC ATC CTT GTA ATC GCC GCC ACC CAG TAA CTT AAC AAA GAA CCA-3′. The polymerase chain reaction product was digested withSmaI and SacII. The MET3 promoter and digested polymerase chain reaction product were cloned intoHindIII and SacII sites of pTF62 (YEp181 with replacement of Bluescript II polylinker), an episomal plasmid bearingLEU2 as a selectable marker. Methionine-dependent expression of Ccc1p-FLAG was confirmed by Western blot using a mouse monoclonal antibody against FLAG epitope. The FLAG epitope did not influence the ability of CCC1 to maintain respiratory activity in Δyfh1 cells. The METCCC1 strain was generated by transforming METYFH1 strain with pMetCCC1. The transformed cells were “cured” of the pMetYFH1 plasmid by growth on uracil-containing medium.

      Construction of YFH1- and AFT1up-containing Plasmids

      A plasmid containing YFH1 and a LEU2 gene as a selectable marker was constructed using polymerase chain reaction. Two primers were designed to amplify the up-stream 500 base pairs of YFH1. They are 028, 5′-CGC GGA TCC CTT ACC AGT CGC TGA TGC-3′, and 029, 5′-AA CTG CAG TCC CCC GGG TGC TAC ACT CTA TCT TCT CGC TTA G-3′. Another two primers were designed to amplify the open reading frame of YFH1 fused with FLAG sequence at 3′ end. They are 030, 5′-AA CTG CAG TCC CCC GGG ATG ATT AAG CGG TCT CTC GC-3′, and 031, 5′-CCC AAG CTT TTA CTT GTC ATC GTC ATC CTT GTA ATC GCC GCC TTG GCT TTT AGA AAT GGC CT-3′. The two amplified products were digested, purified, and ligated into pTF62 vector at BamHI and HindIII site. ASmaI site was inserted upstream of start codon ATG. Complementation analysis showed the FLAG epitope tag did not interfere with the protein function.
      An AFT1 up clone was generously provided by Dr. Dennis Winge. The original construct in a pRS313 vector, bearing aHIS3 gene, was digested with SpeI andSacI to release the AFT1 up fragment, which was cloned into pRS314, bearing the TRP1 marker.

      Iron-dependent Petite Formation

      METYFH1 transformed with vector only or pSCCC1 was selected on medium lacking uracil, leucine, and methionine. 2 × 104 cells grown in the selective medium were inoculated into 5.0 ml of liquid medium supplemented with methionine at 330 mg/liter and specified concentrations of iron (ferrous ammonium sulfate). After 18 h, cell number was determined, and about 100 colony-forming units were plated on medium lacking uracil, leucine, and methionine and incubated at 30 °C for 2 days. The incidence of petite colonies was determined by replica-plating to plates with 2% glycerol and 2% ethanol as a carbon source.

      Iron Uptake Assay

      Iron transport assays was performed as described (
      • Davis-Kaplan S.R.
      • Askwith C.C.
      • Bengtzen A.C.
      • Radisky D.
      • Kaplan J.
      ). In a 0.5-ml reaction, 5 × 106 cells were mixed with 59Fe (FeCl3; Dupont) at 0.5 μm in EDTA buffer with or without 1.0 mmascorbate. Cells were incubated at 30 °C for 10 min and then washed on filters with EDTA-containing buffer to remove unincorporated iron. The filter was air-dried, and radioactivity was determined. The uptake activity was expressed as fmol of 59Fe/min/106cells.

      59Fe Pulse-Chase and Subcellular Fractionation

      Exponentially growing cultures were harvested and washed three times with cold Lim-EDTA. 1 × 109 cells were incubated with 59Fe at 0.5 μm in the presence of ascorbate in 100 ml of Lim-EDTA at 30 °C for 10 min. Cells were harvested, washed three times with cold EDTA-containing buffer, and then incubated with 50 μm FeCl3and 1 mm ascorbate in the same culture medium for 2 h. Cells were harvested, spheroplasts were prepared, and subcellular organelles were recovered by centrifugation of a post-nuclear supernatant at 12,000 × g for 30 min as described by Radisky et al. (
      • Radisky D.C.
      • Babcock M.C.
      • Kaplan J.
      ). The membrane pellet was layered onto a 0–25% iodixanol gradient, which was centrifuged at 12,000 ×g for 2 h. The gradient was fractionated, and radioactivity in each fraction was determined. The location of organelles was determined by Western blot analysis using an antibody against mitochondrial porin, vacuolar alkaline phosphatase (Molecular Probes), and Pep12p (a generous gift from Dr. Scott Emr). To detect porin or Pep12p, 0.05-ml samples from each gradient fraction were applied to gel electrophoresis and then Western-blotted. To detect alkaline phosphatase, 0.75-ml samples from each fraction were precipitated with trchloroacetic acid before gel electrophoresis and Western blotting. The data is plotted as the radioactivity in each gradient fraction expressed as a fraction of the total radioactivity in the post-nuclear supernatant.

      RESULTS

      Identification of CCC1 as a High Copy Suppressor of Δyfh1

      Since Δyfh1 cells are respiratory-deficient, we used a plasmid shuttle approach to identify high copy suppressors that could rescue Δyfh1 cells from respiratory deficit (Fig. 1). A diploid heterozygous for chromosomal deletion of YFH1 was transformed with aURA3 plasmid in which YFH1 was placed under the control of the MET3 promoter (pMetYFH1). Diploids were sporulated in the absence of methionine, which permittedYFH1 gene expression and maintained respiratory competence as shown by growth on glycerol-ethanol. These haploids, designated METYFH1 (Δyfh1, pMetYFH1), were transformed with a high copy (YEp) genomic library that contained the LEU2 gene as a selectable marker. Transformants were selected on plates lacking leucine, uracil, and methionine. The cells were then plated under conditions that promoted the loss of the YFH1-containingURA3 plasmid: growth on medium lacking leucine but containing methionine and uracil followed by replica-plating to medium lacking leucine but containing 5-fluoroorotic acid. Cells capable of growth on glycerol-ethanol medium were examined for the presence or absence of an intact YFH1 gene by the polymerase chain reaction using YFH1-specific primers. Cells that did not contain an intact YFH1 gene and were capable of growth on glycerol-ethanol were selected for further study. Screening of 30,000 transformants resulted in five colonies that could grow on glycerol-ethanol plates but did not contain an intact YFH1. To demonstrate that growth on glycerol-ethanol required theLEU2 library plasmid, cells were grown in leucine-containing medium to permit plasmid loss. All cells that showed a leucine auxotroph were also found to be respiratory-incompetent. Plasmids were able to be rescued from transformed cells and, when re-transformed into the METYFH1 strain, permitted growth on respiratory substrates.
      Figure thumbnail gr1
      Figure 1Outline of suppressor screen. See text for details (m denotes mitochondria).
      Four of the five library clones contained overlapping regions of yeast chromosome XII. The only open reading frame in common was a previously identified gene CCC1 (
      • Fu D.
      • Beeler T.
      • Dunn T.
      ) (
      • Lapinskas P.J.
      • Lin S.J.
      • Culotta V.C.
      ). This gene was subcloned from one of the library clones and, when placed in a high copy vector, was able to support the growth of a Δyfh1 strain in glycerol-ethanol medium (Fig.2 A). Cells lackingYFH1 are viable but grow poorly in glucose medium, particularly in the presence of high levels of iron (
      • Radisky D.C.
      • Babcock M.C.
      • Kaplan J.
      ). Overexpression of CCC1 markedly enhanced growth in glucose-containing medium in the presence of high levels of iron (Fig.2 B).
      Figure thumbnail gr2
      Figure 2Analysis of CCC1 expression in Δyfh1 cells. METYFH1 (Δyfh1, pMetYFH1) transformed with vector alone or with pSCCC1 were allowed to lose the pMetYFH1 plasmid. A, the cells were then plated on glycerol-ethanol. METYFH1 transformed with YFH1, vector, pS222, or pSCCC1 were allowed to lose the pMetYFH1 plasmid. WT, wild type. B, approximately 1,000 cells of each strain were plated in glucose-containing synthetic medium supplemented with different concentrations of iron. C, cells were plated on yeast extract/peptone/dextrose medium containing different concentrations of H2O2.
      It is possible that overexpression of CCC1 is not sufficient for respiratory competence, as the selection system may have resulted in the generation of obligate endogenous suppressors. Additional mutations may have occurred during the process of selecting for cells that lost the pMetYFH1 plasmid. To test this possibility we took advantage of the ability to regulate YFH1 expression through the MET3 promoter. The addition of methionine repressedYFH1 expression, resulting in an iron concentration-dependent increase inrho cells. Cells overexpressing CCC1 maintained respiratory activity, even in the face of high iron levels (Fig. 3 A). If cells resistant to 500 μm iron were allowed to lose the CCC1plasmid through growth on leucine-containing medium, they also lost the ability to grow on glycerol-ethanol medium, becoming petite in an iron concentration-dependent manner (Fig. 3 B). These results suggest that growth of Δyfh1 cells on glycerol-ethanol is solely dependent on the CCC1 plasmid. The suppression of iron-dependent petite formation by overexpression of CCC1 demonstrates that CCC1 is both necessary and sufficient for maintaining respiratory activity ofΔyfh1 cells.
      Figure thumbnail gr3
      Figure 3Effect of CCC1 on petite formation. METYFH1 transformed with vector only or pSCCC1 were grown in glucose medium containing methionine and iron (ferrous ammonium sulfate) as indicated. A, iron-dependent petite formation was determined by replica-plating to glycerol-ethanol-containing media. METYFH1 transformed with pSCCC1, grown in 500 μm iron, was allowed to lose the pSCCC1 plasmid. B, iron-dependent petite formation was measured.

      CCC1 Prevents Mitochondrial Iron Accumulation

      Our previous studies indicated that mitochondrial iron accumulation was the proximal cause of respiratory incompetence in Δyfh1 cells (
      • Radisky D.C.
      • Babcock M.C.
      • Kaplan J.
      ). IfCCC1 maintains respiratory competence in Δyfh1cells, it could do so by either increasing antioxidant defenses or by limiting mitochondrial iron accumulation. Growth of cells in medium containing H2O2 showed that CCC1overexpression had at best a marginal protective effect. This effect was seen with the plasmid that contained just CCC1 (pSCCC1) but not with the original library clone (pS222). Yet both plasmids could maintain the respiratory activity of Δyfh1 cells. These results suggest that it is unlikely that the respiratory sparing ability of CCC1 is due to increased antioxidant defenses (Fig. 2 C). We then examined the effect of CCC1overexpression on mitochondrial iron accumulation. When METYFH1 cells were grown in the presence of methionine (Yfh1p “off”), there was a 10-fold increase in the total amount of cell-associated59Fe, and more than 50% of that was localized in mitochondria (Fig. 4). The accumulation of mitochondrial iron was not seen if cells were grown in the absence of methionine (Yfh1p “on”). In CCC1-overexpressing METYFH1 cells, the amount of cell-associated 59Fe was similar in the presence or absence of methionine (see below). Yet regardless of whether Yfh1p was expressed or not,CCC1-overexpressing cells did not accumulate mitochondrial iron.
      Figure thumbnail gr4
      Figure 4The effect of CCC1expression on mitochondrial iron accumulation. METYFH1 transformed with vector only or pSCCC1 were grown in the presence or absence of methionine for 8 h. Cells were washed, incubated with59Fe for 10 min, and then resuspended in growth medium containing nonradioactive iron for 2 h. Cells were harvested, spheroplasted, and homogenized, and the membrane fraction was applied to an iodixanol gradient. A, the amount of radioactivity in each fraction of the gradient was determined. B, distribution of organelle markers on iodixanol gradients was determined by Western blot analysis using antibodies directed against mitochondrial porin, prevacuole Pep12p, and vacuolar alkaline phosphatase (ALP).
      The inability of CCC1-overexpressing cells to accumulate mitochondrial iron could reflect a restriction on mitochondrial iron import or an acceleration of mitochondrial iron export. We took advantage of the ability to regulate CCC1 expression to distinguish between these possibilities. A strain, METCCC1 (Δyfh1, pMetCCC1), was constructed that had a chromosomal deletion of YFH1 and contained a high copy plasmid ofCCC1 under the control of the MET3 promoter. Cells grown in the presence of methionine overnight did not show evidence of CCC1 mRNA by Northern blot analysis. Maximal levels of CCC1 mRNA, however, were reached within 2 h of methionine removal (data not shown). To examine the effect ofCCC1 on mitochondrial iron efflux, METCCC1 cells were grown in the presence of methionine overnight to turn off expression of Ccc1p. The cells were washed with assay buffer, incubated for 10 min with 59Fe2+, washed, and then incubated for 2 h in the same medium containing methionine but without radioactive iron. At the end of this incubation, the cells were divided into two aliquots, one aliquot was again incubated in the same methionine-containing medium (Ccc1p off), and the other was incubated in the medium lacking methionine (Ccc1p on). Cells were harvested at specific times, spheroplasted, and homogenized, and organelles were applied to an iodixanol gradient. The amount of cell-associated iron was similar in Δyfh1 cells regardless of whether they synthesized Ccc1p. Yet cells grown in the absence of methionine (Ccc1p on) had less than 20% cell associated iron within mitochondria compared with greater than 60% cell-associated iron in mitochondria of cells grown in the presence of methionine (Ccc1p off). Once iron had accumulated within mitochondria, re-expression of Ccc1p did not promote the loss of that accumulated iron (Fig.5, A and B). In contrast, re-expression of Yfh1p in Δyfh1 cells resulted in an acceleration of mitochondrial iron egress (Fig. 5, Cand D). These results indicate that, unlike YFH1,CCC1 is not involved in mitochondrial iron efflux, suggesting that Ccc1p limits iron entry into mitochondria.
      Figure thumbnail gr5
      Figure 5CCC1 expression does not promote mitochondrial iron efflux. METCCC1 (A and B) and METYFH1 (C and D) were grown in growth media with methionine (Ccc1p or Yfh1p, off) or without methionine (Ccc1p or Yfh1p, on) for 12 h. At the end of that incubation, cells were washed, incubated with 59Fe for 10 min, and then incubated in the same medium in which they were grown. After 2 h, cells previously grown in methionine containing media were either maintained in that medium (off to off) or transferred to methionine-free medium (off to on). At the designated times (0 and 6 h), samples were harvested, and the intracellular distribution of iron was determined.
      CCC1 was originally identified as a Mn2+/Ca2+Golgi transporter, and it is possible that it may transport iron as well. One explanation for the decrease in mitochondrial iron may be that Ccc1p transports iron out of the cytosol. Decreased cytosolic iron may limit iron entry into mitochondria. Analysis of the distribution of59Fe within cells, however, did not provide evidence thatCCC1 overexpression led to accumulation of iron in any membranous compartment (Figs. 4 and 5). A representative gradient of subcellular organelles is shown in Fig. 4 B. We did not detect any accumulation of iron within vacuoles, prevacuoles, or Golgi vesicles. Our fractionation procedure resulted in less than 5.0% release of vacuolar carboxypeptidase Y to the cytosolic fraction (data not shown). Greater than 75% of accumulated iron was found to be cytosolic in both wild type and Δyfh1 cells overexpressingCCC1 (data not shown). We also considered the possibility that CCC1 overexpression resulted in the appearance of Ccc1p on the cell surface, where it might act as an efflux channel. Pulse-chase experiments, however, provided no evidence that Ccc1p overexpression resulted in efflux of cellular iron (data not shown).

      Overexpression of CCC1 Affects the Iron Regulon

      Cells with aYFH1 deletion show increased expression of genes regulated by the iron-dependent transcription factor Aft1p (
      • Babcock M.
      • de Silva D.
      • Oaks R.
      • Davis-Kaplan S.
      • Jiralerspong S.
      • Montermini L.
      • Pandolfo M.
      • Kaplan J.
      ). This increased expression results from sequestration of iron within mitochondria, resulting in a decreased cytosolic iron concentration. Overexpression of CCC1 did not prevent the increased rate of iron uptake in Δyfh1 cells and resulted in increased iron transport in wild type cells (Fig.6 A). Increased iron uptake was mediated by the high affinity iron transport system, as deletion ofFET3 resulted in no iron transport inCCC1-overexpressing cells. No increased uptake activity was seen in a CCC1-overexpressing Δaft1 strain, indicating that the effect of CCC1 required the iron transcription factor Aft1p. CCC1-overexpressing cells had a higher iron content than wild type cells, as measured by atomic absorption spectroscopy (control cells, 60.8 fmol of iron/106 cells; pSCCC1-transformed cells, 99.0 fmol iron/106 cells). The increase in iron content was similar to that seen in Δyfh1 cells grown in the same medium but was less than expected, based on the measurement of iron transporter activity. The high affinity iron transport system is regulated primarily by iron through Aft1p. Ferrireductase activity, which is required to convert ferric iron to ferrous iron (the substrate for the high affinity transport system) is regulated by several factors including iron (Aft1p), copper (Mac1p), and cyclic AMP (
      • Lesuisse E.
      • Horion B.
      • Labbe P.
      • Hilger F.
      ). That ferrireductase activity was rate-limiting was shown by measuring iron transport in the presence and absence of ascorbate (Fig.6 B). In the absence of ascorbate, which bypasses the need for a reductase, CCC1-overexpressing cells took up more iron than control cells (vector only) but much less iron than in the presence of ascorbate. A similar pattern of iron transport was observed for cells that had increased expression of the high affinity iron transport system through the action of a constitutive allele ofAFT1 (AFT1 up ).
      Figure thumbnail gr6
      Figure 6Induction of high affinity iron uptake system by CCC1 overexpression. A, wild type (DY150), Δfet3, and Δaft1 were transformed with vector or CCC1 subcloned into the vector pTF63, and high affinity iron uptake activity was determined. B, iron uptake activity was determined in the presence and absence of ascorbate in wild type cells transformed with vector only,AFT1 up, or pSCCC1.
      The finding that overexpression of CCC1 results in induction of at least some iron-regulated genes leads to the question of whether activation of the iron regulon will suppress the respiratory phenotype of Δyfh1. Introduction of theAFT1 up allele into wild type cells resulted in an increase in high affinity iron transport to the same extent asCCC1 overexpression (Fig. 6 B), yetAFT1 up did not prevent iron-dependent petite formation in Δyfh1 cells (Fig. 7).
      Figure thumbnail gr7
      Figure 7Effect of AFT1 upon iron-dependent petite formation.Iron-dependent petite formation was determined in METYFH1 strain transformed with vector only or with plasmids containing eitherAFT1 up or CCC1.

      DISCUSSION

      A genetic screen revealed that CCC1 is a high copy suppressor of the respiratory deficit of Δyfh1 cells. High copy plasmid expression of CCC1 prevented Δyfh1from becoming petite, even in the presence of high iron concentrations.CCC1 overexpression does not prevent cellular iron uptake but does prevent the increased mitochondrial iron accumulation expected of a Δyfh1 strain. This result confirms previous studies demonstrating that the respiratory defect in Δyfh1 cells is due to excessive accumulation of iron within mitochondria (
      • Radisky D.C.
      • Babcock M.C.
      • Kaplan J.
      ). Heart biopsies from patients with Friedreich's ataxia have shown a selective defect in the activity of mitochondrial iron-sulfur proteins (
      • Rotig A.
      • de Lonlay P.
      • Chretien D.
      • Foury F.
      • Koenig M.
      • Sidi D.
      • Munnich A.
      • Rustin P.
      ). A recent report indicated that in the absence of Yfh1p, even in low iron conditions, there was a decrease in aconitase activity, suggesting that Yfh1p was involved in regulating formation of iron-sulfur clusters (
      • Foury F.
      ). A caveat to those experiments is that aconitase protein levels were not measured. In our strain of yeast, however, deletion of YFH1 does not result in significant loss of iron-sulfur-containing enzyme activities, as determined by biologic function. For the studies reported here, we employedΔyfh1 cells that had a deletion in the LEU2gene. These cells could be converted to leucine prototroph by aLEU2-containing plasmid. Leucine biosynthesis requires the activity of Leu1p, a cytosolic enzyme that relies on iron-sulfur clusters for catalytic activity (
      • Kispal G.
      • Csere P.
      • Prohl C.
      • Lill R.
      ). Deletion of YFH1 does not confer leucine auxotroph, implying that Yfh1p is not required for the formation of cytosolic iron-sulfur clusters. Respiration also requires the activity of iron-sulfur cluster proteins, andCCC1 can preserve respiratory activity in aΔyfh1 strain. These observations suggest thatYFH1 may not be significantly involved in iron-sulfur cluster formation.
      Overexpression of CCC1 preserves respiratory activity by preventing mitochondrial iron accumulation. We determined that Ccc1p does not accelerate the loss of iron from mitochondria. Thus, Ccc1p prevents or limits iron transported into mitochondria. The limitation on mitochondrial iron accumulation is clearly not absolute, as enough iron enters the mitochondria for the synthesis of heme and iron-sulfur clusters. The mechanism by which Ccc1p limits mitochondrial iron uptake is not clear. Ccc1p was characterized as a Mn2+/Ca2+ transporter, based on genetic studies (
      • Fu D.
      • Beeler T.
      • Dunn T.
      ,
      • Lapinskas P.J.
      • Lin S.J.
      • Culotta V.C.
      ). Examination of the data base reveals homologous genes in plants, bacteria, and the archea. No homologues have been identified in either invertebrates (Caenorhabditis elegans) or vertebrates. It is possible that Ccc1p may transport iron, as most Fe2+ transporters are capable of transporting other divalent transition metals, particularly Mn2+ (
      • Gunshin H.
      • Mackenzie B.
      • Berger U.V.
      • Gunshin Y.
      • Romero M.F.
      • Boron W.F.
      • Nussberger S.
      • Gollan J.L.
      • Hediger M.A.
      ,
      • Korshunova Y.O.
      • Eide D.
      • Clark W.G.
      • Guerinot M.L.
      • Pakrasi H.B.
      ,
      • Supek F.
      • Supekova L.
      • Nelson H.
      • Nelson N.
      ). We have shown that Ccc1p does not lead to iron export from cells or sequestration of iron in membranous compartments. Our results indicate that iron accumulates in cytosol and is not recognized by the iron-sensing transcription factor Aft1p. We do not know the form or species (Fe2+ or Fe3+) of this stored iron. Cells overexpressing Ccc1p perceive an apparent “low” iron content and increase the transcription of the iron regulon, resulting in increased activity of the high affinity iron transport system.
      Increased activity of the high affinity iron transport inCCC1-overexpressing cells is an indication that cells “sense” low iron levels. Expression of the high affinity iron transport system, however, does not necessarily result in iron-limited growth; the activities of a number of essential iron-containing proteins, such as methyl sterol oxidase, Δ9-fatty acid desaturase, are unaffected as shown by robust rates of cell growth. The mechanism by which mitochondria can adjust iron accumulation with respect to cytosolic iron levels is unknown. Restriction of cytosolic “free iron” either by environmental iron restriction (
      • Radisky D.C.
      • Babcock M.C.
      • Kaplan J.
      ,
      • Foury F.
      ), genetic ablation of high affinity iron transport systems (
      • Radisky D.C.
      • Babcock M.C.
      • Kaplan J.
      ), or genetic manipulation of intracellular iron content (CCC1overexpression) prevents excessive mitochondrial iron accumulation in cells lacking YFH1.

      Acknowledgments

      We express our appreciation to our colleagues for help in preparing this manuscript.

      REFERENCES

        • Campuzano V.
        • Montermini L.
        • Lutz Y.
        • Cova L.
        • Hindelang C.
        • Jiralerspong S.
        • Trottier Y.
        • Kish S.J.
        • Faucheux B.
        • Trouillas P.
        • Authier F.J.
        • Durr A.
        • Mandel J.L.
        • Vescovi A.
        • Pandolfo M.
        • Koenig M.
        Hum. Mol. Genet. 1997; 6: 1771-1780
        • Koutnikova H.
        • Campuzano V.
        • Foury F.
        • Dolle P.
        • Cazzalini O.
        • Koenig M.
        Nat. Genet. 1997; 16: 345-351
        • Babcock M.
        • de Silva D.
        • Oaks R.
        • Davis-Kaplan S.
        • Jiralerspong S.
        • Montermini L.
        • Pandolfo M.
        • Kaplan J.
        Science. 1997; 276: 1709-1712
        • Wilson R.B.
        • Roof D.M.
        Nat. Genet. 1997; 16: 352-357
        • Lamarche J.B.
        • Cote M.
        • Lemieux B.
        Can. J. Neurol. Sci. 1980; 7: 389-396
        • Delatycki M.B.
        • Camakaris J.
        • Brooks H.
        • Evans-Whipp T.
        • Thorburn D.R.
        • Williamson R.
        • Forrest S.M.
        Ann. Neurol. 1999; 45: 673-675
        • Wong A.
        • Yang J.
        • Cavadini P.
        • Gellera C.
        • Lonnerdal B.
        • Taroni F.
        • Cortopassi G.
        Hum. Mol. Genet. 1999; 8: 425-430
        • Fu D.
        • Beeler T.
        • Dunn T.
        Yeast. 1994; 10: 515-521
        • Lapinskas P.J.
        • Lin S.J.
        • Culotta V.C.
        Mol. Microbiol. 1996; 21: 519-528
        • Radisky D.C.
        • Babcock M.C.
        • Kaplan J.
        J. Biol. Chem. 1999; 274: 4497-4499
        • Sambrook J.
        • Fritsch E.F.
        • Maniatis T.
        Molecular Cloning: A Laboratory Manual. 2nd Ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1989
        • Sherman F.
        • Fink G.R.
        • Lawrence C.W.
        Methods in Yeast Genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1989
        • Davis-Kaplan S.R.
        • Askwith C.C.
        • Bengtzen A.C.
        • Radisky D.
        • Kaplan J.
        Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 13641-13645
        • Lesuisse E.
        • Horion B.
        • Labbe P.
        • Hilger F.
        Biochem. J. 1991; 280: 545-548
        • Rotig A.
        • de Lonlay P.
        • Chretien D.
        • Foury F.
        • Koenig M.
        • Sidi D.
        • Munnich A.
        • Rustin P.
        Nat. Genet. 1997; 17: 215-217
        • Foury F.
        FEBS Lett. 1999; 456: 281-284
        • Kispal G.
        • Csere P.
        • Prohl C.
        • Lill R.
        EMBO J. 1999; 18: 3981-3989
        • Gunshin H.
        • Mackenzie B.
        • Berger U.V.
        • Gunshin Y.
        • Romero M.F.
        • Boron W.F.
        • Nussberger S.
        • Gollan J.L.
        • Hediger M.A.
        Nature. 1997; 388: 482-488
        • Korshunova Y.O.
        • Eide D.
        • Clark W.G.
        • Guerinot M.L.
        • Pakrasi H.B.
        Plant. Mol. Biol. 1999; 40: 37-44
        • Supek F.
        • Supekova L.
        • Nelson H.
        • Nelson N.
        Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 5105-5110