Direct activation of genes involved in intracellular iron use by the yeast iron-responsive transcription factor Aft2 without its paralog Aft1

Abstract

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.

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 AFT1-1^up (1up) or AFT2-1^up (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.

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 AFT1-1^up and/or AFT2-1^up 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.

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 aft2Δ (aft2Δ, AFT1-HA) context and strains containing Aft2-HA in a wild-type (AFT1, AFT2-HA) or aft1Δ (aft1Δ, 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.

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 aft2Δ (aft2Δ, AFT1-HA) context and strains containing Aft2-HA in a wild type (AFT1, AFT2-HA) or aft1Δ (aft1Δ, 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.

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 aft2Δ (aft2Δ, AFT1-HA) context and strains containing Aft2-HA in a wild-type (AFT1, AFT2-HA) or aft1Δ (aft1Δ, 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.

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 aft2Δ (aft2Δ, AFT1-HA) context and strains containing Aft2-HA in a wild-type (AFT1, AFT2-HA) or aft1Δ (aft1Δ, AFT2-HA) context were grown exponentially under iron-depleted conditions. (D) Only the strain containing Aft2-HA in an aft1Δ context (aft1Δ, 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.

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 (aft1Δ), Y11090 (aft2Δ), and SCMC01 (aft1Δ aft2Δ), 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 β-galactosidase activities are shown in the lower panel.

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 aft2Δ strains (A) and HA-tagged Aft2 in the wild-type and aft1Δ 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%.

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 aft2Δ strain (black bars) and promoter occupancy by Aft2 in the aft1Δ 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.