Hap1 regulates aerobic genes during both aerobic and hypoxic growth conditions. Expression microarray experiments revealed 197 genes that were significantly downregulated in wild-type cells upon a shift to hypoxic growth. The hap1Δ/HAP1 expression ratio of these 197 genes is shown for aerobic and hypoxic cells. The promoters of 34 of these genes (black diamonds) were significantly bound by Hap1 under aerobic conditions. Points labeled with gene names changed more than twofold in hap1Δ compared to HAP1. Statistical analyses are described in the Materials and Methods section.

Hap1 is both an aerobic activator of CYC1 and a hypoxic repressor of ERG genes. (A) ERG5 and CYC1 mRNA levels were measured by Northern analysis in a HAP1 strain (FY2609) and a hap1Δ strain (FY2611) grown in hypoxic conditions for the indicated times. Shown is a typical Northern blot with the averages of the results of multiple Northern analyses (as described in Materials and Methods) as follows: lane 1, 5.4; lane 2, 1.7; lane 3, 2.0; lane 4, 1.0; lane 5, 0.9; lane 6, 4.5; lane 7, 8.7; lane 8, 11.8; lane 9, 15.9; lane 10, 19.8. hr, hour. (B) Northern analyses were performed to measure ERG2 and ERG11 mRNA levels in HAP1 (FY2609) and hap1Δ (FY2611) strains grown aerobically (+) and hypoxically (−) for 8 h. The averages of the quantitations from Northern analyses for ERG2 are as follows: lane 1, 1.8; lane 2, 1.0; lane 3, 1.3; lane 4, 4.8. The averages of the quantitations from Northern analyses for ERG11 are lane 1, 2.9; lane 2, 1.0; lane 3, 2.7; lane 4, 7.3.

Hap1 binds to the promoters of ERG and CYC1 genes under aerobic and hypoxic conditions. Hap1 binding was measured by ChIP (see Materials and Methods), using an α-myc antibody, and by quantitating the following gene sequences by real-time PCR: ERG2 (−491 to −371), ERG5 (−323 to −189), ERG11 (−607 to −490), and CYC1 (−397 to −268). Cells expressing Hap1 fused to a myc epitope tag (myc-HAP1, strain FY2612 or FY2613) or untagged HAP1 (HAP1⁺, strain FY2609 or FY2610) were grown under aerobic and hypoxic conditions for 4 h. Shown are the means and standard errors of at least three independent experiments.

Hap1 represses ERG5 through recruitment of the Tup1 corepressor. (A) ERG5 and CYC1 expression were monitored by Northern blotting of wild-type (FY2609), hap1Δ (FY2611), and tup1Δ (FY2673) cells grown hypoxically (−O₂) or aerobically (+O₂) for 4 h. The averages of the quantitations from Northern analyses are as follows: lane 1, 9.4; lane 2, 3.8; lane 3, 6.5; lane 4, 1.0; lane 5, 13.2; lane 6, 7.3. (B) Tup1 binding to the ERG5 promoter (−323 to −189) was monitored by ChIP experiments in aerobic or hypoxic conditions, using an untagged (wild-type, FY2609), TUP1-myc (FY2674), or TUP1-myc hap1Δ (FY2675) strain. Shown are the means and standard errors for three independent experiments.

Two Hap1 binding sites contribute to ERG5 regulation. (A) At the top is a diagram of the structure of the ERG5 promoter, with numbers denoting locations relative to the ERG5 ATG. The two consensus TATA sequences and putative Hap1 binding sites are shown. Shown below is the alignment of the consensus Hap1 binding sites within the ERG5 promoter of the related yeast species: S. cerevisiae (Scer), S. paradoxus (Spar), S. kudravzevii (Skud), S. bayanus (Sbay), and S. mikitae (Smik). For site 2, there is no consensus site in S. bayanus. Also shown is the alignment for UAS1 from the CYC1 promoter. The numbers denote the first nucleotide of each of the sequences relative to ATG. Nucleotides shown in bold are consensus Hap1 positions, while the nucleotides in shaded boxes denote conserved positions. Site 1 is the reverse complement. (B) ERG5 mRNA levels were measured by Northern blotting in strains containing a wild-type ERG5 promoter (ERG5, FY2612), ERG5 Hap1 site 2 mutant (erg5-2, FY2624), ERG5 Hap1 site 1 mutant (erg5-1, FY2626), ERG5 with both sites mutant (erg5-1,2, FY2628), and hap1Δ (FY2611). Strains that are not hap1Δ contain the myc-HAP1 allele. Cells were grown hypoxically (−) or aerobically (+) for 4 h. Shown is one of the Northern blots from multiple experiments. The averages of the quantitations from multiple Northern analyses are as follows: lane 1, 9.4; lane 2, 1.0; lane 3, 6.1; lane 4, 1.6; lane 5, 6.4; lane 6, 2.9; lane 7, 3.6; lane 8, 2.8; lane 9, 3.8; lane 10, 13.2. (C) Hap1 binding to the ERG5 promoter (−323 to −189) was monitored by ChIP in the following strains: untagged HAP1 (ERG5, HAP1⁺, FY2610), myc-HAP1 (FY2612 or FY2613), myc-HAP1 erg5-1 (FY2625), myc-HAP1 erg5-2 (FY2623), and myc-HAP1 erg5-1,2 (FY2627). Shown are the means and standard errors for three independent experiments. Note that the data for untagged HAP1 and myc-HAP1 are also shown in Fig. 3.

Additional ERG5 promoter elements are important for repression. (A) Structure of the ERG5 promoter denoting regions 3 to 6 that were mutated. (B) ERG5 mRNA levels were measured by Northern blotting in strains containing the following ERG5 promoter mutations: ERG5 (FY2609), erg5-3 (FY2618), erg5-4 (FY2619), erg5-5 (FY2620), or erg5-6 (FY2621). Cells were grown aerobically (+) or hypoxically (−) for 4 h. Shown is one of the Northern blots from multiple experiments. The averages of the quantitations are as follows: lane 1, 9.4; lane 2, 1.0; lane 3, 6.2; lane 4, 0.8; lane 5, 4.5; lane 6, 9.2; lane 7, 4.4; lane 8, 2.5; lane 9, 9.7; lane 10, 0.8. (C) Conservation of region 4 in the sequenced sensu stricto yeast species S. cerevisiae (Scer), S. paradoxus (Spar), S. kudravzevii (Skud), S. bayanus (Sbay), and S. mikitae (Smik). Putative Hap1 half-sites in the ERG5 promoter are denoted in bold and are underlined, while the nucleotides in the shaded boxes denote conserved nucleotides. Regions 7 to 14 are the same regions noted and analyzed for Fig. 7. Five conserved half-sites were found in region 4, more than the 0.2 CGG or CGC sites expected by chance (P < 10⁻²⁶, as determined by the chi-square test for independence). The background rate was determined by counting the number of conserved sites (on either strand) within 5,000 bp of the promoter sequence (specifically within −100 to −600 bp of ATG) of 10 randomly chosen genes. (D) Hap1 binding to the ERG5 promoter (−482 to −367) was monitored by ChIP experiments with the following strains: untagged HAP1 (HAP1⁺ ERG5, strain FY2610), myc-HAP1 (strain FY2612), and myc-HAP1 erg5-4 (strain FY2644). Shown are the means and standard errors of three independent experiments. The lower level of Hap1 binding to the ERG5 promoter compared to the results shown in Fig. 3 and 5C may be due to strain or primer differences; the primers used for the experiments in Fig. 3 and 5C could not be used here because one of the primers hybridized to the mutated region 4.

Multiple sequences within ERG5 region 4 are required for repression. (A) Structure of the ERG5 promoter denoting regions 4 and 7 to 14 that were mutated. (B) Effect of 50-bp mutations in region 4 on ERG5 expression. Northern blotting was performed for strains containing the following ERG5 promoter alleles: ERG5 (FY2612), erg5-4 (FY2644), erg5-7 (FY2645), erg5-8 (FY2646), and erg5-9 (FY2647). The averages of the quantitations from multiple Northern blotting experiments are as follows: lane 1, 9.4; lane 2, 1.0; lane 3, 3.7; lane 4, 8.1; lane 5, 6.0; lane 6, 4.1; lane 7, 6.2; lane 8, 4.3; lane 9, 3.9; lane 10, 2.6. (C) Effect of 20-bp mutations on ERG5 expression. Northern blotting was performed for strains containing the following ERG5 promoter alleles: ERG5 (FY2612), erg5-4 (FY2644), erg5-10 (FY2639), erg5-11 (FY2640), erg5-12 (FY2641), erg5-13 (FY2642), and erg5-14 (FY2643). Cells were grown aerobically (+) or hypoxically (−) for 4 h. The averages of the quantitations from Northern analyses are as follows: lane 1, 9.4; lane 2, 1.0; lane 3, 3.7; lane 4, 8.1; lane 5, 4.9; lane 6, 1.7; lane 7, 5.6; lane 8, 2.1; lane 9, 5.1; lane 10, 3.3; lane 11, 6.1; lane 12, 2.9; lane 13, 9.2; lane 14, 1.7.

ERG5 and CYC1 are differentially regulated at intermediate heme levels. (A) ERG5 expression was monitored by Northern blotting in HAP1⁺ (FY2609) and hap1Δ (FY2611) cells grown hypoxically (−O₂) or aerobically (+O₂) for 8 h in the presence (+) or absence (−) of 50 μg/ml heme. The averages of quantitations are as follows: lane 1, 7.5; lane 2, 1.0; lane 3, 15.5; lane 4, 6.1; lane 5, 25.2; lane 6, 20.6. (B) Northern blotting was performed to measure ERG5 and CYC1 mRNA levels in hem1Δ (FY2637) and hem1Δ hap1Δ (FY2638) mutants. Cells were grown in media containing the indicated concentrations of δ-ala for 4 h. (C) ERG2, ERG5, ERG11, and CYC1 mRNA levels in hem1Δ cells were quantitated as described in the Materials and Methods section. Shown are the means and standard errors for three independent experiments.