MCB Free Medline Searching
Home Help [Feedback] [For Subscribers] [Archive] [Search] [Contents]
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Courel, M.
Right arrow Articles by Blaiseau, P.-L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Courel, M.
Right arrow Articles by Blaiseau, P.-L.
Molecular and Cellular Biology, August 2005, p. 6760-6771, Vol. 25, No. 15
0270-7306/05/$08.00+0     doi:10.1128/MCB.25.15.6760-6771.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.

Direct Activation of Genes Involved in Intracellular Iron Use by the Yeast Iron-Responsive Transcription Factor Aft2 without Its Paralog Aft1

Maïté Courel, Sylvie Lallet, Jean-Michel Camadro, and Pierre-Louis Blaiseau*

Laboratoire d'Ingénierie des Protéines et Contrôle Métabolique, Département de Biologie des Génomes, Institut Jacques-Monod, UMR 7592 CNRS-Universités Paris 6 and 7, 2 Place Jussieu, F-75251 Paris cedex 05, France

Received 1 December 2004/ Returned for modification 18 January 2005/ Accepted 25 April 2005


   ABSTRACT
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
The yeast Saccharomyces cerevisiae contains a pair of paralogous iron-responsive transcription activators, Aft1 and Aft2. Aft1 activates the cell surface iron uptake systems in iron depletion, while the role of Aft2 remains poorly understood. This study compares the functions of Aft1 and Aft2 in regulating the transcription of genes involved in iron homeostasis, with reference to the presence/absence of the paralog. Cluster analysis of DNA microarray data identified the classes of genes regulated by Aft1 or Aft2, or both. Aft2 activates the transcription of genes involved in intracellular iron use in the absence of Aft1. Northern blot analyses, combined with chromatin immunoprecipitation experiments on selected genes from each class, demonstrated that Aft2 directly activates the genes SMF3 and MRS4 involved in mitochondrial and vacuolar iron homeostasis, while Aft1 does not. Computer analysis found different cis-regulatory elements for Aft1 and Aft2, and transcription analysis using variants of the FET3 promoter indicated that Aft1 is more specific for the canonical iron-responsive element TGCACCC than is Aft2. Finally, the absence of either Aft1 or Aft2 showed an iron-dependent increase in the amount of the remaining paralog. This may provide additional control of cellular iron homeostasis.


   INTRODUCTION
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Iron is an essential nutrient, but its accumulation can be highly cytotoxic owing to its chemical reactivity, which depends on its redox state (II or III). Prokaryotic and eukaryotic cells have therefore evolved various regulatory mechanisms to control iron homeostasis and to maintain a balance between nutritional deprivation and an excess of iron (12, 13). The yeast Saccharomyces cerevisiae has two paralogous iron-responsive transcription activators, Aft1 and Aft2, that play key roles in the response to a lack of iron in the environment by increasing the expression of genes involved in iron transport and its subcellular distribution and use (28). The N-terminal regions of Aft1 and Aft2, which contain the DNA binding domain (29, 38), are well conserved (3). These N-terminal regions interact with the same DNA promoter in vitro (29, 38). The replacement of a conserved cysteine residue in the N-terminal region by phenylalanine makes the gain of function mutant alleles AFT1-1up (37) and AFT2-1up (29) iron independent.

Aft1 is located in the cytosol of cells grown under iron-replete conditions, but in cells grown under iron-depleted conditions, it is in the nucleus, where it binds to DNA and activates transcription (39). Cells lacking AFT1 grow poorly under iron-depleted conditions (3, 29, 37). Consistent with this phenotype, Aft1 activates the transcription of genes involved in iron uptake at the plasma membrane. These include genes that encode the high-affinity ferrous transport complex composed of the multicopper oxidase (FET3) and iron permease (FTR1) (1, 34), the copper transporter responsible for delivering copper to Fet3 (CCC2) (40), plasma membrane metalloreductases (FRE1 to FRE4) (5, 10, 41), iron-siderophore transporters (ARN1 to ARN4) (17, 18, 42, 43), and cell wall mannoproteins, which facilitate the uptake of siderophore-bound iron (FIT1 to FIT3) (25). Aft1 is also involved in the activation of other genes, such as FTH1, which encodes a vacuolar iron transporter (35), and genes with functions not yet elucidated in iron metabolism, such as HMX1, the homolog of the gene encoding heme oxygenases (26, 33), two members of the FRE family (FRE5 and FRE6) (21), and CTH2, a gene recently shown to be involved in mRNA degradation under iron deficiency (27). Others genes were recently shown by DNA microarray analyses to be regulated by the constitutive AFT1-1up mutant allele, but the role of Aft1 in their regulation remains to be elucidated (30, 33).

The role of Aft2 is still unclear, unlike that of Aft1. No phenotype is associated with the lack of AFT2 alone. Consistent with this lack of phenotype, the genes involved in the iron uptake systems are expressed similarly in the wild type and in the aft2{Delta} mutant (3; unpublished results). However, plasmids expressing AFT2 in the aft1 aft2 mutant activate the transcription of Aft1 target genes in an iron-regulated manner (3, 29). The deletion of AFT2 exacerbates the phenotype of the aft1 mutant, rendering the aft1 aft2 double mutant unable to grow under iron-depleted conditions, and it abolishes the residual transcription of genes such as FET3 and CTH2 that still occurs in the single aft1 mutant (3, 27). The aft1 aft2 mutant also has many oxidative stress-related phenotypes that are not present in the aft1 mutant (3). These results suggested that the roles of Aft2 and Aft1 overlap to some extent.

DNA microarray data have defined a set of genes that is activated by the constitutive AFT2-1up (29, 30). A few of these genes encode proteins that are involved in iron homeostasis, such as the vacuolar iron transporter SMF3 (23, 24), the mitochondrial iron transporter MRS4 (7), and a protein involved in the mitochondrial iron-sulfur cluster assembly, ISU1 (9, 32). This work was designed to define the involvement of Aft1 and Aft2 in the transcriptional regulation of iron homeostasis in regard to the presence/absence of the paralog. DNA microarray clustering allowed us to identify several classes of genes that are regulated by Aft1 and/or Aft2, and computer analyses highlighted different consensus sequences for each class. A combination of Northern blotting and chromatin immunoprecipitation experiments with the iron-regulated genes FET3, FTR1, SMF3, and MRS4 demonstrated that the direct transcription activation mediated by either Aft1 or Aft2 is gene specific and iron dependent. Aft2 directly activates the transcription of the iron intracellular use genes SMF3 and MRS4, while Aft1 does not. We show that Aft2 functions in the absence of Aft1. We have also obtained evidence that the amounts of Aft1 and Aft2 are increased in the absence of the paralog and that iron represses the amounts of Aft1 and Aft2 in these genetic contexts.


   MATERIALS AND METHODS
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Yeast strains, plasmids, and growth conditions. All strains used are listed in Table 1. Transcriptome analysis experiments were performed with strains CM3260, Y18, and Y18aft2{Delta}. The strains used for other experiments were derivatives from strains obtained from Research Genetics (Huntsville, AL). The haploid strain SCMC01 (aft1{Delta} aft2{Delta}) was constructed as follows. Y01090 and Y14438 were mated, and the diploid strain was selected on medium lacking lysine and methionine and was made to sporulate. Tetrads were dissected, and spores showing resistance to Geneticin and hypersensitivity to copper were characterized. AFT1 and AFT2 deletions were verified by PCR, and the known phenotypes of the Y18aft2{Delta} double mutant strain (3) were confirmed. Strains SCMC05 (AFT2, AFT1-HA), SCMC11, SCMC18 (AFT1, AFT2-HA), SCMC10 (aft2{Delta}, AFT1-HA), and SCMC13 (aft1{Delta}, AFT2-HA) carry three tandem copies of the influenza virus hemagglutinin (HA) epitope at the very carboxy terminus of AFT1 or AFT2. The HA epitope tags for AFT1 and AFT2 were amplified from the template pFA6a-3HA-HIS3MX6 as described previously (20), using the following primer sets: AFT1-3HA, 5'-AATGGTGAACGGCGAGTTGAAGTATGTGAAGCCAGAAGATCGGATCCCCGGGTTAATTAA-3' and 5'-ATGGACGAGAGATACGTCTAAGTTTGATTTCATCTATATGGAATTCGAGCTCGTTTAAAC-3'; AFT2-3HA, 5'-TGAATTAAATTCTATTGACCCAGCCTTAATATCAAAATATCGGATCCCCGGGTTAATTAA-3' and 5'-TTAAACGTGATACCGTTTTAATGAGTTGAAAACTAAATAAGAATTCGAGCTCGTTTAAAC-3'. The AFT1-3HA PCR products were transformed into BY4742 and those of AFT2-3HA were transformed into BY4741 to generate strains SCMC05 and SCMC11. Strains SCMC18, SCMC10, and SCMC13 were isolated after mating strains SCMC11 and BY4742, SCMC05 and Y01090, and SCMC11 and Y14438, respectively. Epitope-tagged strains were verified by PCR, sequencing, and protein synthesis. The plasmids pEG202-AFT1 and pEG202-AFT2 have been described previously (3). Plasmid pFC-W was kindly provided by Y. Yamaguchi-Iwai; it contains the –263/–234 upstream activation sequence of the FET3 gene that has been inserted into the CYC1 promoter and fused to the lacZ gene (38). Plasmids pFC-M1, pFC-M2, and pFC-M3, containing site-directed nucleotide substitutions introduced into the FET3 core sequence (–252/–243) to resemble to the SMF3 sequence (–362/–353), were constructed as follows. The entire SalI-BamHI fragment from the pFC-W was first subcloned into the pUC-18 vector (Stratagene), and the resulting plasmid was used as a PCR template for the QuikChange mutagenesis kit (Stratagene) according to the manufacturer's instructions. The primers used were 5'-GGCTCGACCTTCAAAACCGCACCCATTTGCAGGTGC-3' and its complement for M1 substitutions, 5'-CCTTCAAAAGTGCACCCTGTTGCAGGTGCTCGTCG-3' and its complement for M2 substitutions, and 5'-GGCTCGACCTTCAAAACCGCACCCTGTTGC-AGGTGCTCGTCG-3' and its complement for M3 substitutions. Then the entire SalI-BamHI fragment from the pUC-18 vector was reinserted into the SalI-BamHI sites of the pFC-W vector to obtain the pFC-M1, pFC-M2, and pFC-M3 plasmids. All PCR-generated sequences were confirmed by DNA sequencing.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Saccharomyces cerevisiae strains used

 
All yeast transformations were performed by the lithium acetate method. The iron-depleted or iron-replete conditions were created by first growing cells at 30°C in a defined medium consisting of an iron-limiting and copper-limiting yeast nitrogen base (catalog no. 4027-122; Bio101) plus 1 µM ferric ammonium sulfate. These cells (optical density at 600 nm [OD600] = 0.3) were placed in the defined medium with or without 100 µM ferric ammonium sulfate, and the cultures were grown for 5 h to an OD600 of 1.0. They were then used for DNA microarray assays, Northern blotting, chromatin immunoprecipitation (ChIP) experiments, or ß-galactosidase assays.

RNA isolation, Northern analysis, and ß-galactosidase assay. Total RNA extraction, Northern blotting (15 µg total RNA), and hybridization were performed in duplicate, as described previously (14, 31). The 32P-labeled DNA fragments used as probes corresponded to the open reading frame of each gene. A 1.2-kb BamHI-HindIII fragment was used for the ACT1 gene. The 623-bp fragment of FET3, the 759-bp fragment of FTR1, the 744-bp fragment of MRS4, and the 924-bp fragment of SMF3 were obtained by PCR using the primer sets listed in Table 2. The membranes were exposed for 2 days. Data were quantified using ImageQuant software and normalized using the ACT1 mRNA signal. ß-Galactosidase was assayed using o-nitrophenyl-D-galactopyranoside (11).


View this table:
[in this window]
[in a new window]
  		TABLE 2. Primer sets for PCR amplifications used in ChIP assays and Northern blots

 
Microarray hybridization and analysis. RNA extracted from strains grown exponentially in iron-depleted medium was used to hybridize a yeast GeneFilter microarray (Research Genetics, Invitrogen Corporation) according to the manufacturer's instructions, except that 5 µg of total RNA was reverse transcribed. Hybridizations were done in duplicate using two separate sets of filters. Images were acquired after 3 days of exposure using a Storm 860 phosphorimager (Molecular Dynamics) and analyzed with PATHWAYS 3 software (Research Genetics). Those genes previously found to be activated by AFT1-1up and/or AFT2-1up were selected and subjected to a k mean clustering (k = 5) (6) (http://rana.stanford.edu/clustering). The JavaTreeView program (http://genome-www.stanford.edu/~alok/TreeView) was used to visualize the clustered data. Multiple expectation maximization for motif elicitation (MEME) (2) was used to identify shared motifs in the 700 bp of the promoters of similarly regulated genes. Additional information and a version of MEME running on a parallel supercomputer are available at http://meme.sdsc.edu/meme/website/intro.html.

Chromatin immunoprecipitation. Cells were grown exponentially in 100 ml iron-depleted or iron-replete medium to an OD600 of 1. The chromatin was then prepared (15), and the resulting supernatant volume was adjusted to 4 ml before storage at –80°C. Immunoprecipitations were performed in duplicate. Five hundred microliters of the cross-linked chromatin solution was added to 8 µg of anti-HA monoclonal antibodies (Santa Cruz Biotechnology) prebound to 10 mg protein A-Sepharose CL-4B (Sigma) and incubated for 1.5 h at room temperature. Protein A-Sepharose CL-4B without antibody was used for background control. Beads were washed twice with 1.6 ml FA lysis buffer (15) with 500 mM NaCl; once with 1.6 ml 10 mM Tris-HCl, pH 8, 0.25 mM LiCl, 1 mM EDTA, 0.5% NP-40, and 0.5% sodium deoxycholate; and once with 1.6 ml Tris-EDTA, for 15 min each. Chromatin complexes were released from the beads by incubation in 500 µl of 25 mM Tris-HCl, pH 7.5, 10 mM EDTA, and 0.5% sodium dodecyl sulfate for 15 min at 65°C. Cross-links from eluates and crude chromatin solution (50 µl) were reversed by incubation with 600 µg proteinase K (Sigma) for 1 h at 37°C and overnight at 65°C. The resulting DNA was purified on PCR purification kit columns (QIAGEN).

Real-time quantitative PCR analysis. The QIAGEN Quantitect SYBR Green PCR kit was used for quantitative real-time PCR in a LightCycler (Roche Diagnostics). Primer pairs (Table 2) were designed with Oligo 4.0-s software to generate products of 90 to 130 bp. For the FET3, SMF3, and MRS4 promoters, PCR fragments were amplified with primers flanking the iron regulatory sites defined by promoter deletion analyses (23, 30, 38). For the FTR1 promoter, we designed primers to amplify the region (190 bp) between the two TGCACCC sequences. For POL1 and RPO21, used as negatives controls, we designed primers within their coding sequences.

PCRs were carried out in 15-µl reaction mixtures with 2 µM concentrations of each primer and 1x Quantitect SYBR Green PCR kit. The DNA templates added to the reaction mixture were 1/150 of the immunoprecipitated or background DNA and 1/50,000 of the input DNA. The LightCycler protocol was denaturation at 95°C for 15 min, 45 cycles of amplification and quantification (95°C for 20 s, 55°C for 20 s, 72°C for 25 s, with a single measurement), and a melting curve of 60 to 95°C, with a heating rate of 0.1°C per second and continuous fluorescence measurement.

Data were analyzed using the Roche LightCycler 3.5 software and the fit point method. The crossing point (CP) was defined as the point at which the fluorescence was 10 times the background fluorescence. The efficiency (E) of each primer pair was calculated from the slope of the linear standard curve (E = 10–1/slope) generated with a fivefold dilution of a DNA input mix. The protein occupancy of each DNA fragment was then calculated as previously described (4): protein occupancy = E(CP input – CP immunoprecipitation)/E(CP input – CP background). The data were averaged over two independent experiments, with real-time PCR performed at least in duplicate. The relative enrichment of a selected DNA fragment was obtained by dividing the protein occupancy at this DNA fragment by the average protein occupancy at the negative controls (coding sequences of POL1 and RPO21).

Protein extraction and Western blotting. Total protein extracts from 3 ml of cells grown exponentially in iron-depleted or iron-replete media were prepared by the NaOH-trichloroacetic acid lysis technique (36). Aliquots (5 µl) were separated on an 8% sodium dodecyl sulfate-polyacrylamide gel electrophoresis gel and transferred to nitrocellulose membranes. The membranes were blocked with 3% bovine serum albumin (Sigma), 0.1% Tween 20 (Sigma) in Tris-buffered saline and probed at room temperature in the same blocking buffer. Anti-HA monoclonal antibodies (Santa Cruz Biotechnology) were diluted at 1:1,000, and anti-Pgk1 monoclonal antibodies (Roche Diagnostics) were diluted at 1:5,000. Horseradish peroxidase-conjugated anti-mouse immunoglobulin G (diluted 1:1,000) was used as the secondary antibody (Sigma) and was detected by enhanced chemiluminescence (ECL kit; Amersham).


   RESULTS
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Inhibition of the Aft2 regulon by Aft1. We investigated the effects of Aft1 and Aft2 on gene expression under iron-depleted conditions. Previous studies show that the transcription mediated by wild-type Aft2 can be detected in the absence of Aft1 (3, 29). We therefore examined the effect of Aft2 on gene expression in an aft1 mutant genetic context. DNA microarray hybridizations were performed with labeled transcripts extracted from the wild-type strain and single aft1 and double aft1 aft2 mutant cells grown in iron-depleted conditions. Comparisons of the gene expression in these genetic contexts allowed us to identify 332 genes whose expression decreased more than twofold in at least one of the 3 comparisons, wt/aft1, wt/aft1 aft2, or aft1/aft1 aft2 (experimental data sets are available at the Gene Expression Omnibus, accession number GSE1763 [http://www.ncbi.nlm.nih.gov/geo/]). To take advantage of both our "loss of function" approach and previous "gain of function" analyses, we combined the experimental data and focused on those 50 genes that were postulated as potential targets of AFT1-1up and/or AFT2-1up (29, 30, 33). Genes with similar transcription profiles were grouped by cluster analysis into 5 classes (classes A to E) (Fig. 1).



View larger version (33K):
[in this window]
[in a new window]
  		FIG. 1. Cluster analysis of DNA microarray hybridization with the wild-type, aft1, and aft1 aft2 strains grown under iron-depleted conditions. Each column displays the results from two experiments, and cells represent the averaged ratio of mRNAs (strain 1/strain 2). Transcripts more abundant in strain 1 are in red; transcripts more abundant in strain 2 are in green. The scale indicates the magnitude of the expression ratio. The AFT1-1up (1up) or AFT2-1up (2up) activation from previous studies (29, 30, 33) is indicated. Gene functions are described according to the Saccharomyces cerevisiae Genome Database. Genes selected for Northern blotting and ChIP analyses are indicated in boldface type. The number of MEME consensus sites within the 700 bp of the promoter of each gene are indicated.

 
Class A contained those genes whose mRNAs were at least twofold more abundant in the wild type than in either the aft1 or aft1 aft2 mutant. Most of these genes had been shown to be Aft1 target genes, encoding proteins involved in the plasma membrane iron transport (e.g., FET3, FRE1, FIT2 to FIT3, and ARN1 to ARN4). Class B contained genes whose mRNA amounts in the single aft1 mutant were similar to those of the wild-type strain, but whose mRNA amounts were lower in the aft1 aft2 double mutant. This transcription profile suggests that the roles of Aft1 and Aft2 overlap in the control of these genes. Class B, like class A, contained genes involved in iron metabolism, except for ZRT1, which encodes the high-affinity zinc transporter (44). Surprisingly, the mRNAs of class C and D genes were more abundant in the aft1 mutant than in the wild type, indicating that Aft1 has a negative influence on the transcription of these genes. The class C and D genes showed more mRNA in the aft1 mutant than in the double mutant (positive aft1/aft1 aft2 ratio), as did those of class B, suggesting that Aft2 activates their transcription in the absence of Aft1. As shown in Fig. 1, most of the class C and D genes are activated by AFT2-1up (29, 30). These genes had almost the same transcription profiles, except that the class C genes had a higher wild type/aft1 aft2 ratio than did those of class D. Unlike classes A and B, classes C and D contained few genes encoding proteins involved in iron homeostasis. These genes take no part in iron transport across the plasma membrane but are involved in vacuolar iron transport (SMF3, FTH1, and FET5) or mitochondrial iron transport and use (MRS4 and ISU1). Lastly, class E contained genes with neither positive wild type/aft1 nor positive aft1/aft1 aft2 ratios, unlike those of classes A to D. This suggests that Aft1 and Aft2 are not important for their transcription under our experimental conditions, and the effects of Aft1 and Aft2 on their transcription were not investigated further.

The MEME program (2) was used to identify potential regulatory elements in the promoters of the genes regulated by Aft1 and Aft2. The promoter region between the predicted ATG start codon and 700 bp upstream was chosen to identify the most probable motif within each class, A to E, of input promoters (Fig. 1 and 2). MEME successfully identified the canonical iron-responsive element (38) TGCACCC in the promoters of the A and B class genes (Fig. 2). Thus, 13 of the 19 genes (68%) analyzed contained at least one copy of this sequence in either orientation. Analysis of the whole genome identified 3% of all open reading frames having this sequence within their promoter. There was also generally an A 2 bases upstream of the TGCACCC sequence. In contrast, the sequence TGCACCC was present in only 6 of the 22 class C and D gene promoters (27%). The most probable motif identified within the promoters of class C and D genes was restricted to G/ACACCC, with 20 of the 22 genes (91%) containing at least one copy of this motif. This sequence was present in 24% of the promoters of the whole genome. About 80% of the class C genes contained the G/ACACCC sequence followed by an AT-rich region starting 3 bases downstream the motif. MEME identified the GCACCCT sequence as the most probable motif of the class E genes; it was present in 44% of the genes analyzed (4 of 9 genes) and in 3.2% of the promoters of the whole genome. This sequence was often preceded by a T, reminiscent of the known TGCACCC sequence.



View larger version (57K):
[in this window]
[in a new window]
  		FIG. 2. Probability-based motif derived from MEME analysis of genes in classes A to E. The letter probability matrix of the motif is based on the elements within the 700-bp promoters of genes from each class. The scale indicates the probability of each possible base occurring at each position in the motif multiplied by 10 and rounded to the nearest integer. The most probable form of the motif (the MEME consensus) derived from the probability matrix is shown. The putative consensus binding sites for Aft1 and Aft2 in the DNA sequences upstream of the 50 genes up-regulated by AFT1-1up and/or AFT2-1up are shown. Boxed dark gray nucleotides are identical to the consensus sequence identified by MEME, and boxed light gray nucleotides are found in more than 50% of the analyzed sequences. Numbering corresponds to +1 at the putative translation start site.

 
We attempted to distinguish between the direct and indirect effects of Aft1 and Aft2 on the regulation of class A, B, C, and D genes by examining one iron-regulated gene from each class using ChIP experiments and Northern blotting analyses. The FET3, FTR1, SMF3, and MRS4 genes were chosen arbitrarily because of their known function in iron homeostasis. The ChIP experiments were performed in two genetic contexts (presence and absence of the paralogous protein) because Aft1 and Aft2 could interfere with each other.

Direct activation of FET3 transcription by Aft1 but not by Aft2. In vivo DNA footprint analyses have shown that Aft1 binds to the FET3 FeRE sequence and activates its transcription in an iron-dependent manner (37, 38). In contrast, the role of Aft2 in the transcriptional activation of FET3 is not clear. Aft2 binds to the same FET3 FeRE sequence as Aft1 in vitro (29), but Aft2-dependent regulation of FET3 in vivo has only been reported for specific conditions, such as overexpression of AFT2 in the absence of Aft1 (3) or expression of the constitutive allele AFT2-1up (29, 30). Northern blot analyses (Fig. 3A) confirmed that the transcription of FET3 required AFT1 but not AFT2 (3). Overexpression of AFT2 in aft1{Delta} aft2{Delta} increased the amount of FET3 mRNA, although this was still lower than that resulting from overexpression of AFT1. The amount of FET3 mRNA increased 1.5-fold in the absence of Aft2. ChIP analyses showed that Aft1 was strongly bound to the FET3 promoter in wild-type cells, whereas Aft2 was not (Fig. 3B). The occupancy of the FET3 promoter by Aft1 also increased 1.3-fold in the absence of Aft2. Conversely, low but reproducible amounts of Aft2 were bound to the FET3 promoter in the absence of Aft1, although this appeared to be insufficient to sustain observable FET3 mRNAs production (Fig. 3A and B). We further investigated the effect of iron on FET3 transcription and on the binding of Aft1/Aft2 to the FET3 promoter. Transcription of FET3 was repressed by adding iron (Fig. 3C), as previously reported (37). However, residual FET3 mRNAs were still detected in the wild-type and aft2{Delta} strains. ChIP assays indicated that iron decreased Aft1 binding to the FET3 promoter fivefold, but it was still four- to fivefold higher than the binding to nonrelevant DNA controls (Fig. 3D). The weak binding of Aft2 to the FET3 promoter in the absence of Aft1 was repressed by iron. Thus, Aft2 does not activate the transcription of FET3, although it can poorly bind to the FET3 promoter in the absence of Aft1, and FET3 is specifically activated by Aft1 under iron depletion.



View larger version (31K):
[in this window]
[in a new window]
  		FIG. 3. Effects of Aft1 and Aft2 on FET3 regulation. (A and C) Northern blot analysis of FET3 transcription. Cells were grown exponentially under iron-depleted conditions (–Fe) (A) or under iron-replete conditions (+Fe) (C). Identical amounts of total RNA extracted from the indicated strains were blotted onto a nylon membrane and hybridized with the indicated probes. The plasmids designated 2µAFT1 and 2µAFT2 are derivatives of plasmid pEG202 (see Materials and Methods). Experiments were repeated twice with similar results. Data from a single experiment are presented. Numbers represent the FET3 signal after normalization with the ACT1 signal. One hundred percent refers to the transcription of FET3 in the wild-type (wt) strain under iron-depleted conditions. (B and D) In vivo DNA binding of Aft1-HA and Aft2-HA to the FET3 promoter. (B) Strains containing Aft1-HA in a wild-type (AFT2, AFT1-HA) or aft2{Delta} (aft2{Delta}, AFT1-HA) context and strains containing Aft2-HA in a wild-type (AFT1, AFT2-HA) or aft1{Delta} (aft1{Delta}, AFT2-HA) context were grown exponentially under iron-depleted conditions. (D) Strains showing significant specific binding of Aft1-HA or Aft2-HA to the FET3 promoter were also grown under iron-replete conditions. ChIP experiments were performed as described in Materials and Methods. Samples (immunoprecipitated DNA, beads-alone control DNA, and total DNA) were amplified with the specified primer pairs. The heights of the bars represent the specific protein occupancy calculated as described in Materials and Methods. Error bars represent the standard deviations resulting from two independent experiments. The relative enrichment is defined as the protein occupancy of the FET3 promoter normalized to the average protein occupancy of the negative controls POL1 and RPO21.

 
Direct activation of FTR1 transcription by Aft2 in the absence of Aft1. FTR1 is an Aft1 target gene (38), and its transcription is reportedly activated by the AFT2-1up mutant allele (29), albeit to a lower degree than by AFT1-1up (30). Our DNA microarray results suggested that Aft1 and Aft2 were redundant in the activation of FTR1 transcription (Fig. 1). Northern blot analyses confirmed the DNA microarray data. The amounts of FTR1 mRNAs in the wild-type, aft1{Delta}, and aft2{Delta} strains were similar, while no transcript was detected in the aft1{Delta} aft2{Delta} double mutant (Fig. 4A). Overexpression of AFT2 induced the expression of FTR1 but to a lesser extent than overexpression of AFT1, in agreement with previous DNA microarray data obtained with the constitutive AFT1-1up/AFT2-1up alleles (29, 30, 33). ChIP experiments showed that Aft1 was bound to the FTR1 promoter in the wild-type and aft2{Delta} strains under iron-depleted conditions (Fig. 4B). The Aft1 occupancy of the FTR1 promoter was increased twofold in the absence of Aft2. Aft2 was also bound to the FTR1 promoter but only in the absence of Aft1. Aft1 and Aft2 occupied the FTR1 promoter similarly in the absence of their paralog, consistent with the similar amounts of FTR1 mRNAs found in the aft1{Delta} and aft2{Delta} mutants (Fig. 3A and B). These analyses indicate that Aft2 can compensate for the absence of Aft1 in the direct control of FTR1 transcription. We also investigated the effect of iron on the Aft1- and Aft2-dependent regulation of FTR1. The transcription of FTR1 decreased in iron-replete conditions, as previously reported (38). However, there was still residual transcription of FTR1 in the wild-type and aft2{Delta} strains, but not in the aft1{Delta} mutant (Fig. 4C). The degree of FTR1 promoter occupancy by Aft1 was two- to fourfold lower than under iron-depleted conditions, but it was still twofold higher than for DNA controls, whereas occupancy of the FTR1 promoter by Aft2 in the aft1{Delta} mutant was sevenfold lower, reaching the level of the DNA controls (Fig. 4D). Therefore, the binding of Aft2 to the FTR1 promoter is more sensitive to iron than is the binding of Aft1.



View larger version (29K):
[in this window]
[in a new window]
  		FIG. 4. Effects of Aft1 and Aft2 on FTR1 regulation. (A and C) Northern blot analysis of FTR1 transcription. Cells grown exponentially under iron-depleted (A) or iron-replete (C) conditions were analyzed as described in the legend to Fig. 3. One hundred percent is the transcription of FET3 in the wild-type (wt) strain under iron-depleted conditions. (B and D) In vivo DNA binding of Aft1-HA and Aft2-HA to the FTR1 promoter. (B) Strains containing Aft1-HA in a wild-type (AFT2, AFT1-HA) or aft2{Delta} (aft2{Delta}, AFT1-HA) context and strains containing Aft2-HA in a wild type (AFT1, AFT2-HA) or aft1{Delta} (aft1{Delta}, AFT2-HA) context were grown exponentially under iron-depleted conditions. (D) Strains showing significant specific binding of Aft1-HA or Aft2-HA to the FTR1 promoter were also grown under iron-replete conditions. ChIP experiments and analyses were performed as described in Materials and Methods and the legend to Fig. 3.

 
Opposing roles for Aft1 and Aft2 in the control of SMF3 and MRS4. The transcription of SMF3 is activated by iron starvation in an Aft1/Aft2-dependent manner (23). Previous DNA microarray analyses indicated that this transcription is activated by AFT2-1up and, to a lesser extent, by AFT1-1up (29, 30, 33). However, SMF3 may be controlled by other transcription factors, unlike FET3 and FTR1, because its transcription is not abolished in the aft1{Delta} aft2{Delta} mutant (23). Our DNA microarray analysis indicated that the amount of SMF3 mRNAs in aft1{Delta} was greater than in the wild type, although it was lower than in the wild type in aft1{Delta} aft2{Delta} (Fig. 1). Northern blot analyses and ChIP experiments were performed to elucidate the antagonistic effect of the AFT1 deletion in wild-type versus aft2{Delta} genetic contexts. The amount of SMF3 mRNAs in the aft1{Delta} mutant was higher than in the wild type, consistent with the DNA microarray data, while it was slightly lower than in the wild type in the aft2{Delta} mutant (Fig. 5A). The effect of the AFT2 deletion was epistatic on that of the AFT1 deletion, since the amount of SMF3 mRNAs in the double aft1{Delta} aft2{Delta} mutant was lower than in the wild-type strain. The signal still detected in aft1{Delta} aft2{Delta} confirmed that other factors are involved in the activation of SMF3 transcription. Finally, overexpression of either AFT1 or AFT2 in the aft1{Delta} aft2{Delta} mutant clearly induced SMF3 expression. The occupancy of the SMF3 promoter by Aft1 was only 1.8-fold greater than in the DNA controls in aft2{Delta} and 2.7-fold greater than in the DNA controls in the wild-type strain (Fig. 5B). In contrast, a great deal of Aft2 (12 times more than in the DNA controls) was bound to the SMF3 promoter in the absence of Aft1. No Aft2 was bound to the SMF3 promoter in wild-type cells, as for FET3 and FTR1. This suggests that the increased SMF3 mRNA in aft1{Delta} was due to the direct binding of Aft2 and its activation of transcription. We checked this by investigating the Aft1-dependent and Aft2-dependent transcription of SMF3 in the presence of iron. The experiments performed in iron-replete conditions showed correlated decreases in both the amount of SMF3 mRNA (threefold) and the occupancy of the SMF3 promoter by Aft2 in the aft1{Delta} mutant (fivefold) (Fig. 5C and D). It also showed that the low occupancy of the SMF3 promoter by Aft1 was affected by iron. We conclude that Aft2 directly activates the transcription of SMF3 when Aft1 is absent.



View larger version (30K):
[in this window]
[in a new window]
  		FIG. 5. Effects of Aft1 and Aft2 on SMF3 regulation. (A and C) Northern blot analysis of SMF3 transcription. Strains were grown in iron-depleted (A) or iron-replete (C) medium and analyzed as described in the legend to Fig. 3. One hundred percent is the transcription of FET3 in the wild-type (wt) strain under iron-depleted conditions. (B and D) In vivo DNA binding of Aft1 and Aft2 to the SMF3 promoter. (B) Strains containing Aft1-HA in a wild-type (AFT2, AFT1-HA) or aft2{Delta} (aft2{Delta}, AFT1-HA) context and strains containing Aft2-HA in a wild-type (AFT1, AFT2-HA) or aft1{Delta} (aft1{Delta}, AFT2-HA) context were grown exponentially under iron-depleted conditions. (D) Strains showing significant specific binding of Aft1-HA or Aft2-HA to the SMF3 promoter were also grown under iron-replete conditions. ChIP experiments and analyses were performed as described in Materials and Methods and the legend to Fig. 3.

 
DNA microarray analyses have shown that the transcription of MRS4 is activated by AFT2-1up (29) and AFT1-1up but to a lower degree (30). However, in contrast to the positive action of Aft1 in the control of MRS4, others have reported that MRS4 transcription is greater in the aft1{Delta} mutant than in the wild-type strain (7). Our DNA microarray data (Fig. 1) and Northern blot analyses (Fig. 6A) confirmed the latter result. Moreover, the introduction of a plasmid carrying AFT1 into the aft1{Delta} mutant led to a decrease in the MRS4 mRNAs (Fig. 6A). The amount of MRS4 mRNA was not affected by sole deletion of AFT2, but the increased amount of MRS4 mRNA in the aft1{Delta} mutant was suppressed by deleting AFT2 as well. The remaining signal in the aft1{Delta} aft2{Delta} mutant indicated that other factors are involved in activating MRS4 transcription. These results suggest that the increased transcription of MRS4 in the aft1{Delta} mutant is due to Aft2, as for SMF3. Considerable amounts of Aft2 were consistently bound to the MRS4 promoter in the absence of Aft1 (27.5-fold more than to the DNA controls) (Fig. 6B). Little Aft2 was bound to the MRS4 promoter in wild-type cells (two times more than bound to the DNA controls). No Aft1 was bound to the MRS4 promoter in the wild-type or aft2{Delta} strains, unlike Aft2. The amounts of MRS4 mRNAs were greatly reduced in the absence of Aft1 under iron-replete conditions (Fig. 6C). In agreement with this decreased MRS4 mRNA amount, the binding of Aft2 to the MRS4 promoter was 15 times less than under iron limitation (Fig. 6D). Thus, Aft2 directly activates the transcription of MRS4 in iron-depleted conditions when Aft1 is absent. Control experiments indicated that there was no binding of Aft2 in a wild-type strain grown with iron, regardless of the promoter studied (FET3, FTR1, SMF3, or MRS4) (data not shown).



View larger version (38K):
[in this window]
[in a new window]
  		FIG. 6. Effects of Aft1 and Aft2 on MRS4 regulation. (A and C) Northern blot analysis of MRS4 transcription. Cells were grown exponentially under iron-depleted (A) or iron-replete (C) conditions and analyzed as described in the legend to Fig. 3. One hundred percent is the transcription of FET3 in the wild-type (wt) strain under iron-depleted conditions. (B and D) In vivo DNA binding of Aft1 and Aft2 to the MRS4 promoter. (B) Strains containing Aft1-HA in a wild-type (AFT2, AFT1-HA) or aft2{Delta} (aft2{Delta}, AFT1-HA) context and strains containing Aft2-HA in a wild-type (AFT1, AFT2-HA) or aft1{Delta} (aft1{Delta}, AFT2-HA) context were grown exponentially under iron-depleted conditions. (D) Only the strain containing Aft2-HA in an aft1{Delta} context (aft1{Delta}, AFT2-HA) was grown under iron-replete conditions. ChIP experiments and analyses were performed as described in Materials and Methods and the legend to Fig. 3.

 
Mutational analysis of the FET3 regulatory sequence. The above results indicate that Aft1 and Aft2 activate different target genes. The transcription of FET3 was specifically activated by Aft1, while those of SFM3 and MRS4 were specifically activated by Aft2. Furthermore, computer analyses (Fig. 2) identified a slight difference between the consensus binding sites of genes activated by either Aft1 (TGCACCC) or Aft2 (G/ACACCC). We modified the iron regulatory sequence in the FET3 promoter to resemble that of the SFM3 promoter by site-directed mutagenesis so as to investigate the functional importance of the difference in these consensus binding sites. Dinucleotides 5' and/or 3' of the core sequence GCACCC of the pFC-W FET3 promoter LacZ fusion were changed as shown in Fig. 7A. The pFC-W plasmid and the plasmids pFC-M1, pFC-M2, and pFC-M3 with mutated promoters were used to transform the wild-type, aft1{Delta}, aft2{Delta}, and aft1{Delta} aft2{Delta} strains. These plasmids were also used to transform the aft1{Delta} aft2{Delta} strain harboring a high-copy-number plasmid which overexpressed either AFT1 or AFT2. We evaluated the Aft1 and Aft2 transcriptional activity from the mutated promoters by comparing the ß-galactosidase activities in these strains. The ß-galactosidase activities obtained with the pFC-W plasmid in the different genetic contexts (Fig. 7B) confirmed previous data showing that the transcription of FET3 is predominantly Aft1 dependent (3). The slightly higher ß-galactosidase activity in aft2{Delta} than in the wild-type strain is in agreement with the increased FET3 mRNA (Fig. 3A). The ß-galactosidase activity obtained with pFC-M1 was 3.5-fold lower than that obtained with the pFC-W plasmid in the wild-type strain, 2.5-fold lower than that obtained with pFC-M2, and 6.6-fold lower than that obtained with pFC-M3. Similar results were obtained in the aft2{Delta} strain. These results indicate that Aft1 is a poor activator for the mutated promoters and that Aft2 is not involved in the residual activation when Aft1 is present. No significant ß-galactosidase activity was detected in the aft1{Delta} aft2{Delta} mutant transformed with the pFC-W, pFC-M1, pFC-M2, or pFC-M3 plasmid, indicating that the ß-galactosidase activity mediated by these plasmids was strictly Aft1/2 dependent. The ß-galactosidase activity measured in the aft1{Delta} mutant, attributed to Aft2, was 3.5-fold higher than that of the aft1{Delta} aft2{Delta} mutant with pFC-W. The Aft2-dependent activation was more efficient in the pFC-M1 context (2.3-fold greater than pFC-W); in contrast, this Aft2-dependent activation was lower (1.7-fold decrease) with pFC-M2 than with pFC-W and remained unchanged with pFC-M3. The ß-galactosidase activities measured with pFC-W, pFC-M1, pFC-M2, and pFC-M3 were decreased three- to fourfold by adding 100 µM iron in the wild type, aft2{Delta}, and aft1{Delta} strains (data not shown). The differences in Aft1- and Aft2-dependent activation were further confirmed with aft1{Delta} aft2{Delta} strains overexpressing either the AFT1 or AFT2 gene (Fig. 7B). Overexpression of AFT2 increased the transcription from the pFC-M1 promoter 2.4-fold and from the pFC-M3 promoter 2.6-fold, while it decreased the transcription from pFC-M2 1.9-fold compared to pFC-W. The transcription from pFC-M1, pFC-M2, and pFC-M3 was lower than with pFC-W when AFT1 was overexpressed.



View larger version (18K):
[in this window]
[in a new window]
  		FIG. 7. Mutational analysis of the FET3 regulatory sequence. (A) Plasmid pFC-W contains the –263/–234 of the upstream region of the FET3 gene that has been inserted into the CYC1 promoter and fused to the lacZ gene (38). Plasmid pFC-M1 is identical to pFC-W, except that the dinucleotide GT flanking the 5' FET3 core sequence GCACCC was replaced by the dinucleotide CC that flanks the 5' SMF3 GCACCC sequence (Fig. 2). Plasmid pFC-M2 is identical to pFC-W, except that the dinucleotide AT flanking the 3' FET3 GCACCC sequence was replaced with the dinucleotide TG flanking the 3' SMF3 GCACCC sequence. Plasmid pFC-M3 contains both substitutions (GT to CC and AT to TG) of the pFC-M1 and pFC-M2 plasmids. The 5' FET3 core sequence GCACCC is shown in boldface type. Nucleotides that deviate from the FET3 sequence are underlined. (B) The strains BY4742 (wild type [wt]), Y14438 (aft1{Delta}), Y11090 (aft2{Delta}), and SCMC01 (aft1{Delta} aft2{Delta}), harboring the plasmids pFC-W, pFC-M1, pFC-M2, and pFC-M3, with or without overexpression of AFT1 or AFT2 (plasmids pEG202-AFT1 and pEG202-AFT2, see Materials and Methods) were grown for 18 h in iron-limiting medium containing 1 µM iron and then diluted to an OD600 of 0.3 in the same medium without iron and grown to an OD600 of 1.0. Error bars represent the standard deviations (less than 10%) for assays performed on three independent transformants. Numerical values of ß-galactosidase activities are shown in the lower panel.

 
Iron-dependent increases in Aft1 and Aft2 in the absence of paralog. The above results show that the occupancy of promoters by Aft1 and Aft2 depends on the presence or the absence of the paralog and on the iron concentration in the culture medium. We measured the total Aft1-HA and Aft2-HA levels in wild-type cells and in either aft2{Delta} or aft1{Delta} cells grown in iron-depleted and in iron-replete conditions by Western blotting with anti-HA antibody to determine whether these effects resulted from changes in the amounts of Aft1 and Aft2 proteins. The concentration of Aft1 was higher in the absence of Aft2 than in the wild type under iron depletion (Fig. 8A); this is in agreement with the increased Aft1-mediated activation of FET3 and FTR1 transcription in the aft2{Delta} strain (Fig. 3 and 4). The amount of Aft1 in wild-type cells was not greatly affected by adding iron, consistent with previous analyses (39). Nevertheless, it was decreased at least twofold by iron in the absence of Aft2. Analysis of the amounts of Aft2 showed a weak band corresponding to Aft2, fainter than that of Aft1 in wild-type cells (Fig. 8B). The amount of Aft2 was 3.3-fold higher in the absence of Aft1 under iron-depleted conditions, in agreement with the stimulation of the Aft2-activated transcription alone under these conditions (Fig. 3 to 6). The amount of Aft2 in wild-type cells was not affected by adding iron, while it was decreased by iron in the absence of the paralog, as was that of Aft1. The Western blot analyses therefore show that the amounts of Aft1 and Aft2 are increased in the absence of the paralog under iron depletion and that iron represses the amounts Aft1 and Aft2 in these genetic contexts.



View larger version (25K):
[in this window]
[in a new window]
  		FIG. 8. Analysis of the relative abundances of Aft1 and Aft2. Equal amounts of total protein extract from cells grown exponentially in iron-depleted (–Fe) or iron-replete (+Fe) medium were analyzed by Western blotting with an anti-HA antibody to detect HA-tagged Aft1 in the wild-type (wt) and aft2{Delta} strains (A) and HA-tagged Aft2 in the wild-type and aft1{Delta} strains (B). Aft1 and Aft2 were assayed after the same exposure times. Results from two independent protein extracts are shown in bar graphs after normalization to the Pgk1 signal. The amounts of Aft1 and Aft2 in the wild-type context under iron-depleted conditions were arbitrarily defined as 100%.

 

   DISCUSSION
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
The yeast paralogs Aft1 and Aft2 are iron-responsive transcription activators that have overlapping, but not redundant, functions in iron homeostasis. Aft1 is the major regulator of the iron uptake systems, while the role of Aft2 is not yet clear. We assessed the role of Aft2 by analyzing the profiles of gene expression in the wild-type yeast strain and in mutant strains deficient for Aft1 or for both Aft1 and Aft2. The cells were grown in iron-depleted medium, which promotes Aft1- and Aft2-dependent activation of the genes involved in iron metabolism. We used our DNA microarray data to perform a cluster analysis on the 50 genes that were previously identified as target genes of the AFT1-1up and/or AFT2-1up allele (29, 30, 33). This allowed us to uncover several classes of genes that are differentially regulated by Aft1 and Aft2. The class A to D genes exhibited transcription profiles that were consistent with a role of either Aft1 (class A), Aft2 (classes C and D), or both (class B) in their activation. Only the small class E genes exhibited no transcription decrease in each of the 3 comparisons: wt/aft1, wt/aft1 aft2, and aft1/af1 aft2. This discrepancy between our "loss of function" analyses and the previous "gain of function" results could be caused by the different genetic contexts or different experimental conditions.

The detailed study by Northern blotting and ChIP experiments of the prototype genes FET3, FTR1, SMF3, and MRS4, representing the four A, B, C, and D class genes, respectively, has provided new insights into the direct and indirect actions of Aft1 and Aft2 in the regulation of genes involved in iron homeostasis. The degrees of promoter occupancy by Aft1 and Aft2 were correlated with the subsequent transcription of the corresponding genes. A striking feature here is that Aft1 and Aft2 mostly activate distinct iron-regulated genes in vivo by selective binding to the promoter, while they are able to bind to the same sequence in vitro (29). Aft1 binds well to FET3 and FTR1 promoters and poorly to the SMF3 promoter, but Aft1 is not bound to the MRS4 promoter; conversely, Aft2 binds poorly to the FET3 promoter and well to the FTR1, SMF3, and MRS4 promoters (summarized in Fig. 9A). This raises the question of how Aft1 and Aft2 identify the appropriate promoters. A search for specific cis-acting sequences in the promoter regions of class A and B genes activated by Aft1 identified the canonical TGCACCC sequence of the defined FeRE element (38), indicating the importance of this sequence in the Aft1-mediated activation. In contrast, we observed that the consensus sequence in the promoter regions of genes specifically activated by Aft2 (classes C and D) was not the TGCACCC sequence, but the shorter G/ACACCC sequence (Fig. 2). Hence, the TGCACCC sequence appears to be important for Aft1-mediated activation, but not for Aft2-mediated activation. This was confirmed by transcriptional analysis of a LacZ reporter gene cloned under the control of variants of the FET3 GTGCACCCAT iron-responsive element. Changing the 5' GT to CC dramatically decreased Aft1-dependent activation. Changing the 3' AT to TG also affected the Aft1-mediated activation by decreasing the transcription of LacZ, but to a lesser extent than the previous mutation. Introducing both changes in the 5' and 3' of the FET3 iron-responsive element led to a cumulated loss of Aft1-mediated transcription. By contrast with Aft1 reactivity, the 5' variant FET3 CCGCACCCAT element supported increased Aft2-mediated activation. Although the activation measured with the natural amount of Aft2 was low, it was significant, and overexpression of AFT2 confirmed that this 2-bp change is critical for increased activation. Changing only the 3' end of the site (AT to TG) decreased the Aft2-mediated activation. The two mutations led to an overall increased activation when AFT2 was overexpressed. Our results on Aft1-mediated activation agree well with previous DNA binding competition experiments demonstrating that the in vitro-translated Aft1 protein interacted better with the TGCACCCA sequence than with the sequences GGCACCCA or TGCACCCC (38). The new data we provided on Aft2-mediated activation are also in accordance with in vivo analyses of lacZ reporter fusion constructs showing (i) that iron regulation of the Aft2-activated gene SMF3 depends on the –361CGCACCC sequence and not on the –430TGCACCC sequence (22) and (ii) that the AFT2-1up allele activates the transcription of MRS4 through the –238GGCACCC sequence (30). Taken together, our computer analysis of the iron-responsive elements of the Aft1- and Aft2-regulated genes and transcriptional analysis of the FET3 promoter LacZ fusion provide strong support for differently defined Aft1 and Aft2 DNA binding sites; Aft1 appears to be more selective in recognizing the 5' context of the GCACCC sequence than is Aft2. However, the presence of the TGCACCC sequence in the promoter region of a gene is not sufficient for its activation by Aft1 because some class C, D, and E genes contain the TGCACCC sequence in their promoter regions and are not activated by Aft1 (Fig. 1 to 2). Thus, Aft1 (and Aft2) may recognize the promoter through combination with other trans-acting factors in addition to the specific regulatory cis-acting sequence. Recent studies have shown that the HMG box chromatin-associated architectural factor Nhp6 associates with Aft1 in vivo to facilitate its recruitment to the promoter region of certain of the Aft1-activated genes (8).



View larger version (13K):
[in this window]
[in a new window]
  		FIG. 9. Diagram of Aft1 and Aft2 DNA binding activities and protein amounts under iron-depleted conditions. (A) The bars represent the promoter occupancy by Aft1 in the aft2{Delta} strain (black bars) and promoter occupancy by Aft2 in the aft1{Delta} strain (white bars). The heights of the bars are proportional to each calculated relative enrichment from Fig. 3 (FET3), 4 (FTR1), 5 (SMF3), and 6 (MRS4). (B) The amounts of Aft1 and Aft2 proteins are represented by spheres with areas proportional to the protein measured as described in the legend to Fig. 7. Aft1 is in black, and Aft2 is in white. wt, wild type.

 
Our results show that Aft1 specifically activates FET3 transcription and that Aft2 specifically activates SMF3/MRS4 transcription (Fig. 3 to 6). In contrast, the constitutive allele AFT1-1up activates the transcription of MRS4 (30) and AFT2-1up activates that of FET3 (29, 30). The discrepancy between these results suggests that strains carrying the AFT1/AFT2 wild-type alleles are required to unravel the specificity of gene targeting by Aft1 and Aft2 and that the expression of one hyperactive AFT1-1up/AFT2-1up allele may lead to aberrant activation of genes specifically controlled by the paralog. More importantly, the specificity of gene activation by Aft1 and Aft2 described in this work correlates with a specificity of gene function. Aft1 specifically activates the transcription of genes involved in cell surface iron uptake systems (FET3, FRE1, ARN1 to ARN4, and FIT2 to FIT3) (Fig. 1, class A), while Aft2 specifically activates the transcription of genes involved in vacuolar and mitochondrial iron subcompartmentation and use (SMF3, FRE6, FTH1, MRS4, FET5, and ISU1) (Fig. 1, classes C and D). Thus, with two paralogous transcription factors displaying a functional specialization in the control of iron homeostasis, the cell may adapt to environmental iron changes with greater flexibility.

Phenotype analyses have shown that the sole AFT2 deletion confers no iron-specific phenotype, whereas it reveals misregulation of intracellular iron use and oxidative stress-related phenotypes in the absence of Aft1 (3). Consistently, we have now shown that the Aft2-mediated activation of transcription is revealed under iron-depleted conditions and in the absence of Aft1. This suggests that the Aft2 activity is triggered by exacerbated iron-limiting conditions caused by the cumulative effects of environmental iron depletion and a lack of Aft1-dependent iron uptake systems. Thus, in response to severe iron limitation, the activation by Aft2 of the transcription of genes involved in vacuolar and mitochondrial iron transport may lead to a reorganization of the intracellular iron distribution. This is further supported by recent data indicating that the Aft2 target gene MRS4 is involved in a mitochondrial-vacuolar iron-signaling pathway (19). A hierarchical model implicating Aft1 and Aft2 in a graded response to iron limitation fits well with the greater sensitivity of Aft2 to iron: a given iron concentration in the culture medium may completely abolish the binding of Aft2 to DNA but only decrease that of Aft1. Further investigation is now required to clarify the fine-tuning of Aft2 triggering in response to iron limitation.

The absence of one of the Aft1/Aft2 paralogs under iron deprivation conditions leads to an increase in the binding of the resident paralog to its specific promoters and subsequent gene activation (Fig. 3 to 6). These effects are correlated with a change in the abundance of paralog protein in whole cells (Fig. 8 and 9B). The extent to which the absence of either Aft1 or Aft2 increases the amount of the remaining paralog protein varies: the Aft2 concentration increases more in the absence of Aft1 than does that of Aft1 in the absence of Aft2. The reciprocal negative influence of Aft1 and Aft2 may reflect a compensatory mechanism to counterbalance a failure in processes regulated by one factor by stimulating those of the paralog. This would allow the cell to tightly coordinate the Aft1-mediated regulation of extracellular iron transport and the Aft2-mediated regulation of iron intracellular use.

The modulation of protein amounts may involve transcriptional and/or posttranscriptional regulation. Aft1 binds to its own promoter (16). We found –614TGCACCC and –658GGCACCC sequences in the AFT1 promoter. This suggests that Aft1 and Aft2 are directly involved in the regulation of AFT1 transcription. In contrast, no CACCC core element of the FeRE sequence was found in the AFT2 promoter. Any change in its transcription in response to AFT1 deletion would thus occur through other cis- and trans-regulatory elements. Posttranslational effects may also be involved. Our data agree with recent work on mammals showing that the amount of the iron-regulatory protein IRP2 is increased when the paralog gene encoding IRP1 is deleted. Since iron regulates IRP2 by mediating its proteasomal degradation, these experiments suggest that IRP1 is involved in this step of regulation (22). We show that the negative effect of Aft1 and Aft2 on the amount of the paralog is iron dependent. How iron is involved in controlling the balance between the Aft1 and Aft2 proteins is still unknown, and answering this question is critical for a better understanding of the functions of these iron-responsive paralogous transcription factors in the yeast cell. Iron regulates the function of Aft1 by modulating its subcellular distribution (39) but is likely to be involved at other steps of Aft1 control. Nothing is yet known about the regulation of Aft2 function by iron. As a first step toward clarifying this critical point, experiments are in progress to determine the level at which iron regulates Aft2 abundance in the absence of Aft1.


   ACKNOWLEDGMENTS
 
We thank Joel Pothier, Laurent Kuras, Arnaud Teichert, and Renata Santos for helpful discussions and all the members of Rosine Haguenauer-Tsapis's laboratory for advice and technical support. Frederic Devaux critically read the manuscript. The English text was edited by Owen Parkes.

This work was supported by grants from the Ministère de la Recherche (Programme de Recherches Fondamentales en Microbiologie, Maladies Infectieuses et Parasitaires), the Association pour la Recherche sur le Cancer (ARC no. 5439), and the Centre National pour la Recherche Scientifique (grant from the Programme de Toxicologie Nucléaire).


   FOOTNOTES
 
* Corresponding author. Mailing address: Laboratoire d'Ingénierie des Protéines et Contrôle Métabolique, Département de Biologie des Génomes, Institut Jacques-Monod, UMR 7592 CNRS-Universités Paris 6 and 7, 2 Place Jussieu, F-75251 Paris cedex 05, France. Phone: 33 1 44 27 47 41. Fax: 33 1 44 27 57 16. E-mail: blaiseau{at}ijm.jussieu.fr. Back


   REFERENCES
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 
1. Askwith, C., D. Eide, A. Van Ho, P. S. Bernard, L. Li, S. Davis-Kaplan, D. M. Sipe, and J. Kaplan. 1994. The FET3 gene of S. cerevisiae encodes a multicopper oxidase required for ferrous iron uptake. Cell 76:403-410.[CrossRef][Medline]

2. Bailey, T. L., and C. Elkan. 1994. Fitting a mixture model by expectation maximization to discover motifs in biopolymers. Proc. Int. Conf. Intell. Syst. Mol. Biol. 2:28-36.[Medline]

3. Blaiseau, P. L., E. Lesuisse, and J. M. Camadro. 2001. Aft2p, a novel iron-regulated transcription activator that modulates, with Aft1p, intracellular iron use and resistance to oxidative stress in yeast. J. Biol. Chem. 276:34221-34226.[Abstract/Free Full Text]

4. Cobb, J. A., L. Bjergbaek, K. Shimada, C. Frei, and S. M. Gasser. 2003. DNA polymerase stabilization at stalled replication forks requires Mec1 and the RecQ helicase Sgs1. EMBO J. 22:4325-4336.[CrossRef][Medline]

5. Dancis, A., R. D. Klausner, A. G. Hinnebusch, and J. G. Barriocanal. 1990. Genetic evidence that ferric reductase is required for iron uptake in Saccharomyces cerevisiae. Mol. Cell. Biol. 10:2294-2301.[Abstract/Free Full Text]

6. Eisen, M. B., P. T. Spellman, P. O. Brown, and D. Botstein. 1998. Cluster analysis and display of genome-wide expression patterns. Proc. Natl. Acad. Sci. USA 95:14863-14868.[Abstract/Free Full Text]

7. Foury, F., and T. Roganti. 2002. Deletion of the mitochondrial carrier genes MRS3 and MRS4 suppresses mitochondrial iron accumulation in a yeast frataxin-deficient strain. J. Biol. Chem. 277:24475-24483.[Abstract/Free Full Text]

8. Fragiadakis, G. S., D. Tzamarias, and D. Alexandraki. 2004. Nhp6 facilitates Aft1 binding and Ssn6 recruitment, both essential for FRE2 transcriptional activation. EMBO J. 23:333-342.[CrossRef][Medline]

9. Garland, S. A., K. Hoff, L. E. Vickery, and V. C. Culotta. 1999. Saccharomyces cerevisiae ISU1 and ISU2: members of a well-conserved gene family for iron-sulfur cluster assembly. J. Mol. Biol. 294:897-907.[CrossRef][Medline]

10. Georgatsou, E., and D. Alexandraki. 1999. Regulated expression of the Saccharomyces cerevisiae Fre1p/Fre2p Fe/Cu reductase related genes. Yeast 15:573-584.[CrossRef][Medline]

11. Guarente, L. 1983. Yeast promoters and lacZ fusions designed to study expression of cloned genes in yeast. Methods Enzymol. 101:181-191.[Medline]

12. Hantke, K. 2001. Iron and metal regulation in bacteria. Curr. Opin. Microbiol. 4:172-177.[CrossRef][Medline]

13. Hentze, M. W., M. U. Muckenthaler, and N. C. Andrews. 2004. Balancing acts: molecular control of mammalian iron metabolism. Cell 117:285-297.[CrossRef][Medline]

14. Kohrer, K., and H. Domdey. 1991. Preparation of high molecular weight RNA. Methods Enzymol. 194:398-405.[Medline]

15. Kuras, L., and K. Struhl. 1999. Binding of TBP to promoters in vivo is stimulated by activators and requires Pol II holoenzyme. Nature 399:609-613.[CrossRef][Medline]

16. Lee, T. I., N. J. Rinaldi, F. Robert, D. T. Odom, Z. Bar-Joseph, G. K. Gerber, N. M. Hannett, C. T. Harbison, C. M. Thompson, I. Simon, J. Zeitlinger, E. G. Jennings, H. L. Murray, D. B. Gordon, B. Ren, J. J. Wyrick, J. B. Tagne, T. L. Volkert, E. Fraenkel, D. K. Gifford, and R. A. Young. 2002. Transcriptional regulatory networks in Saccharomyces cerevisiae. Science 298:799-804.[Abstract/Free Full Text]

17. Lesuisse, E., P. L. Blaiseau, A. Dancis, and J. M. Camadro. 2001. Siderophore uptake and use by the yeast Saccharomyces cerevisiae. Microbiology 147:289-298.[Abstract/Free Full Text]

18. Lesuisse, E., M. Simon-Casteras, and P. Labbe. 1998. Siderophore-mediated iron uptake in Saccharomyces cerevisiae: the SIT1 gene encodes a ferrioxamine B permease that belongs to the major facilitator superfamily. Microbiology 144:3455-3462.[Abstract]

19. Li, L., and J. Kaplan. 2004. A mitochondrial-vacuolar signaling pathway in yeast that affects iron and copper metabolism. J. Biol. Chem. 279:33653-33661.[Abstract/Free Full Text]

20. Longtine, M. S., A. McKenzie III, D. J. Demarini, N. G. Shah, A. Wach, A. Brachat, P. Philippsen, and J. R. Pringle. 1998. Additional modules for versatile and economical PCR-based gene deletion and modification in Saccharomyces cerevisiae. Yeast 14:953-961.[CrossRef][Medline]

21. Martins, L. J., L. T. Jensen, J. R. Simon, G. L. Keller, D. R. Winge, and J. R. Simons. 1998. Metalloregulation of FRE1 and FRE2 homologs in Saccharomyces cerevisiae. J. Biol. Chem. 273:23716-23721.[Abstract/Free Full Text]

22. Meyron-Holtz, E. G., M. C. Ghosh, K. Iwai, T. LaVaute, X. Brazzolotto, U. V. Berger, W. Land, H. Ollivierre-Wilson, A. Grinberg, P. Love, and T. A. Rouault. 2004. Genetic ablations of iron regulatory proteins 1 and 2 reveal why iron regulatory protein 2 dominates iron homeostasis. EMBO J. 23:386-395.[CrossRef][Medline]

23. Portnoy, M. E., L. T. Jensen, and V. C. Culotta. 2002. The distinct methods by which manganese and iron regulate the Nramp transporters in yeast. Biochem. J. 362:119-124.[CrossRef][Medline]

24. Portnoy, M. E., X. F. Liu, and V. C. Culotta. 2000. Saccharomyces cerevisiae expresses three functionally distinct homologues of the nramp family of metal transporters. Mol. Cell. Biol. 20:7893-7902.[Abstract/Free Full Text]

25. Protchenko, O., T. Ferea, J. Rashford, J. Tiedeman, P. O. Brown, D. Botstein, and C. C. Philpott. 2001. Three cell wall mannoproteins facilitate the uptake of iron in Saccharomyces cerevisiae. J. Biol. Chem. 276:49244-49250.[Abstract/Free Full Text]

26. Protchenko, O., and C. C. Philpott. 2003. Regulation of intracellular heme levels by HMX1, a homologue of heme oxygenase, in Saccharomyces cerevisiae. J. Biol. Chem. 278:36582-36587.[Abstract/Free Full Text]

27. Puig, S., E. Askeland, and D. J. Thiele. 2005. Coordinated remodeling of cellular metabolism during iron deficiency through targeted mRNA degradation. Cell 120:99-110.[CrossRef][Medline]

28. Rutherford, J. C., and A. J. Bird. 2004. Metal-responsive transcription factors that regulate iron, zinc, and copper homeostasis in eukaryotic cells. Eukaryot. Cell 3:1-13.[Free Full Text]

29. Rutherford, J. C., S. Jaron, E. Ray, P. O. Brown, and D. R. Winge. 2001. A second iron-regulatory system in yeast independent of Aft1p. Proc. Natl. Acad. Sci. USA 98:14322-14327.[Abstract/Free Full Text]

30. Rutherford, J. C., S. Jaron, and D. R. Winge. 2003. Aft1p and Aft2p mediate iron-responsive gene expression in yeast through related promoter elements. J. Biol. Chem. 278:27636-27643.[Abstract/Free Full Text]

31. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.

32. Schilke, B., C. Voisine, H. Beinert, and E. Craig. 1999. Evidence for a conserved system for iron metabolism in the mitochondria of Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. USA 96:10206-10211.[Abstract/Free Full Text]

33. Shakoury-Elizeh, M., J. Tiedeman, J. Rashford, T. Ferea, J. Demeter, E. Garcia, R. Rolfes, P. O. Brown, D. Botstein, and C. C. Philpott. 2004. Transcriptional remodeling in response to iron deprivation in Saccharomyces cerevisiae. Mol. Biol. Cell 15:1233-1243.[Abstract/Free Full Text]

34. Stearman, R., D. S. Yuan, Y. Yamaguchi-Iwai, R. D. Klausner, and A. Dancis. 1996. A permease-oxidase complex involved in high-affinity iron uptake in yeast. Science 271:1552-1557.[Abstract]

35. Urbanowski, J. L., and R. C. Piper. 1999. The iron transporter Fth1p forms a complex with the Fet5 iron oxidase and resides on the vacuolar membrane. J. Biol. Chem. 274:38061-38070.[Abstract/Free Full Text]

36. Volland, C., D. Urban-Grimal, G. Geraud, and R. Haguenauer-Tsapis. 1994. Endocytosis and degradation of the yeast uracil permease under adverse conditions. J. Biol. Chem. 269:9833-9841.[Abstract/Free Full Text]

37. Yamaguchi-Iwai, Y., A. Dancis, and R. D. Klausner. 1995. AFT1: a mediator of iron regulated transcriptional control in Saccharomyces cerevisiae. EMBO J. 14:1231-1239.[Medline]

38. Yamaguchi-Iwai, Y., R. Stearman, A. Dancis, and R. D. Klausner. 1996. Iron-regulated DNA binding by the AFT1 protein controls the iron regulon in yeast. EMBO J. 15:3377-3384.[Medline]

39. Yamaguchi-Iwai, Y., R. Ueta, A. Fukunaka, and R. Sasaki. 2002. Subcellular localization of Aft1 transcription factor responds to