GBD-Adr1 binds to GAL genes and recruits RNA Pol II in the presence of Bmh1 activity. (A) GBD ChIP. YLL908 (BMH1 bmh2Δ) and YLL1087 (bmh1-ts bmh2Δ) were transformed with a plasmid expressing either GBD (pOBD2) or GBD-Adr1 (154–424) (pGBDA4) under the control of the ADH1 promoter. Triplicate cultures were grown in selective medium plus 2% glucose to maintain the TRP1 GBD plasmids. Samples were processed for a GBD ChIP, and GBD binding to the GAL1-10 and GAL7 promoters relative to the telomere (TEL) was determined by qPCR. (B) RT-qPCR. Triplicate samples for RT-qPCR were collected from the cultures in panel A. GAL1 and GAL7 mRNA levels relative to ACT1 mRNA were determined. (C) RNA Pol II ChIP. Duplicate cultures of the strains listed in panel A were grown in selective medium plus 2% glucose. Samples were collected for RNA Pol II ChIP, and the levels of RNA Pol II binding to the GAL1-10 and GAL7 promoters relative to the telomere were determined by qPCR. The values represent the means and SDs of three biological replicates.

Bmh proteins and HDACs repress transcription of Adr1-dependent genes in glucose. (A) Heat map of clustered bmh1-ts bmh2Δ (bmh [YLL1087]), hda1Δ rpd3Δ (hdacΔ [CTY-TY44]), and bmh1-ts bmh2Δ hda1Δ rpd3Δ (bmh hdacΔ [KBY3]) microarray data. Genes were clustered using a decision tree based on the direction of 2-fold differences in expression for each mutant compared to the BMH1 BMH2 HDA1 RPD3 wild-type strain (W303-1a). (B) Genes exhibiting the strongest bmh-hdacΔ synergism from cluster 8. The top 52 of 72 cluster 8 genes are listed in order of the magnitude of bmh-hdacΔ synergism. Expression values listed for each mutant are the log₂ ratio to the wild-type strain. The synergism value was calculated as (bmh hdacΔ/BMH HDAC)/[(bmh/BMH) + (hdacΔ/HDAC)]. Values greater than 1 indicate a higher-than-additive increase in expression of the bmh hdacΔ mutant compared to the sum of the individual bmh and hdacΔ mutants. Genes highlighted in yellow are Adr1 dependent (), and the genes in bold have promoters bound by Adr1 (). (C) Enrichment of Adr1, Cat8, Snf1, and glucose-regulated genes in gene clusters. The P value represents the enrichment of the indicated gene sets in each cluster over the yeast transcriptome which was considered to be the genes represented on the Affymetrix Yeast 2.0 arrays. Bold values indicate significant enrichment based on the Fisher's exact test P value. The source of Adr1 targets was YEASTRACT (). The source of Cat8 targets was described in reference . Snf1-dependent and glucose-repressible genes were those identified by Young et al. (). (D) RT-qPCR. Duplicate cultures of YLL908 (WT bmh2Δ), YLL1087 (bmh1-ts bmh2Δ), CTY-TY44 (hda1Δ rpd3Δ), KBY8 (bmh2Δ hda1Δ rpd3Δ), and KBY3 (bmh1-ts bmh2Δ hda1Δ rpd3Δ) were grown in YP plus 5% glucose, and repressed samples were collected for RT-qPCR. Cells from the repressed cultures were pelleted and resuspended in YP plus 0.05% glucose and grown for 6 h, and derepressed samples were collected for RT-qPCR. The levels of ADH2 mRNA relative to ACT1 mRNA were measured for repressed and derepressed samples. The values represent the means and SDs of three biological replicates.

Bmh inactivation enhances transcription without significantly increasing Adr1 promoter binding. (A) Adr1-Myc ChIP. Triplicate cultures of PPY6 (BMH1 bmh2Δ ADR1-MYC) and PPY13 (bmh1-ts bmh2Δ ADR1-MYC) were grown in YP plus 5% glucose (repressed) or YP plus 0.05% glucose (derepressed). Samples were processed for Adr1-Myc ChIP, and Adr1 binding to the ADH2, ACS1, and POX1 promoters relative to the telomere was determined by qPCR. The data are expressed as binding (ChIP/input) at ADH2, ACS1, and POX1 relative to binding (ChIP/input) at the telomere. (B) RT-qPCR. mRNA was extracted from the strains used for panel A after growth under repressed (5% glucose) and derepressed (0.05% glucose) conditions. RT-qPCR was performed to measure the levels of mRNA, and the levels of ADH2, ACS1, and POX1 mRNA were normalized to ACT1 mRNA. For panels A and B, the error bars represent the means of three biological replicates assayed in triplicate.

Adr1 is required for transcriptional activation of ADH2 in the bmh hdac strain in glucose. (A) β-Galactosidase assay. YLL908 (WT bmh2Δ), YLL1087 (bmh1-ts bmh2Δ), CTY-TY44 (hda1Δ rpd3Δ), and KBY3 (bmh1-ts bmh2Δ hda1Δ rpd3Δ) were transformed with one of the ADH2-lacZ reporters (pLGADH2-lacZ or pHDY10). Three or more transformants of each strain were grown in selective medium plus 5% glucose to maintain the URA3 reporter plasmids, and repressed samples were collected for β-galactosidase assays. Cells from the repressed YLL908 culture were pelleted and resuspended in selective medium plus 0.05% glucose and grown overnight for derepression. Derepressed (DR) samples were collected for β-galactosidase assays. The level of reporter activity is reported in Miller units and represents the mean of at least three transformants. (B) RT-qPCR. Triplicate cultures of the adr1Δ strains KBY26 (adr1Δ), KBY30 (adr1Δ bmh1-ts bmh2Δ), KBY31 (adr1Δ hda1Δ rpd3Δ), and KBY34 (adr1Δ bmh1-ts bmh2Δ hda1Δ rpd3Δ) and the cat8Δ strains KBY15 (cat8Δ), KBY20 (cat8Δ bmh1-ts bmh2Δ), KBY18 (cat8Δ hda1Δ rpd3Δ), and KBY22 (cat8Δ bmh1-ts bmh2Δ hda1Δ rpd3Δ) were grown in YP plus 5% glucose. Repressed samples were collected, and ADH2 mRNA levels were assayed by RT-qPCR and plotted relative to ACT1 mRNA. The values represent the means and SDs of three biological replicates. (C) RT-qPCR. The adr1Δ strains KBY26 (adr1Δ), KBY30 (adr1Δ bmh1-ts bmh2Δ), KBY31 (adr1Δ hda1Δ rpd3Δ), and KBY34 (adr1Δ bmh1-ts bmh2Δ hda1Δ rpd3Δ) were transformed with a plasmid expressing either WT Adr1 (pKD16) or Adr1^c (S230A) (pKD14) from the ADR1 promoter. Triplicate cultures were grown in selective medium plus 5% glucose to maintain the TRP1 plasmid. Repressed samples were collected for RT-qPCR, and the levels of ADH2 mRNA relative to ACT1 mRNA were determined. The values represent the means and SDs of three biological replicates.

Bmh is bound to Adr1-dependent promoters in hdac strains under both repressed and derepressed conditions. Triplicate cultures of KBY88 (ADR1-FLAG BMH1-MYC bmh2Δ hda1Δ rpd3Δ) were grown in YP plus 5% glucose (repressed), and then the cells were pelleted and resuspended in YP plus 0.05% glucose and derepressed for 4.5 h. Repressed and derepressed samples were collected for sequential ChIP. The values represent the means of three biological replicates. (A) Adr1-Flag–Bmh1-Myc sequential ChIP. Adr1-Flag was immunoprecipitated from the ChIP extracts, followed by Bmh1-Myc. Bmh1 binding to the ADH2 and ACS1 promoters relative to the telomere was determined by qPCR. (B) RNA Pol II–Bmh1-Myc sequential ChIP. RNA Pol II was immunoprecipitated from the ChIP extracts, followed by Bmh1-Myc. Bmh1 binding to the ADH2 and ACS1 promoters relative to the telomere was determined by qPCR.

Snf1, but not Cat8, is required for derepression in the absence of Bmh activity. (A) Tetrad analysis. CHY35a (MATa reg1Δ) and KBY48 (MATα bmh1-ts bmh2Δ) were mated; diploids were selected, purified, and sporulated; and tetrads were dissected on YPD. Growth is shown after 5 days at 30°C. Twelve tetrads are shown vertically, with each set separated horizontally. The genotype for each spore is shown to the right of the tetrads and was determined based on the markers present in each tetrad. (B) RT-qPCR. Triplicate KBY83 (SNF1^as bmh1-ts bmh2Δ hda1Δ rpd3Δ) cultures were grown in YP plus 5% glucose (repressed), and then the cells were pelleted and resuspended in YP plus 0.05% glucose for derepression. At the start of derepression, Snf1^as activity was inhibited using 5 μM 2NM-PP1, and an equal volume of DMSO was used for a control. Samples were collected from the repressed culture (time zero) and 30, 60, 120, 240, and 360 min following derepression (DR) and Snf1^as inhibition for RT-qPCR. The level of ADH2 mRNA relative to ACT1 mRNA was determined for each time point. The values represent the means of three biological replicates. (C) RT-qPCR. Triplicate cultures of WT, YLL1087 (bmh1-ts bmh2Δ), KBY15 (cat8Δ), and KBY20 (cat8Δ bmh1-ts bmh2Δ) were grown in YP plus 5% glucose, and then the cells were pelleted and resuspended in YP plus 0.05% glucose and derepressed for 4 h. Derepressed samples were collected for RT-qPCR, and the levels of ADH2, ACS1, and FBP1 mRNA relative to ACT1 mRNA were determined. Values were plotted as a percentage of the expression in the WT strain for each of the genes. The values represent the means of three biological replicates. (D) RT-qPCR. CKY13 (adr1Δ CAT8) and CKY23 (adr1Δ cat8Δ) were transformed with a plasmid expressing either WT Adr1 (pKD16) or Adr1^c (S230A) (pKD14) from the ADR1 promoter. Triplicate cultures were grown in YP plus 5% glucose, and then the cells were pelleted and resuspended in YP plus 0.05% glucose and derepressed for 4 h. Derepressed samples were collected for RT-qPCR, and the levels of ADH2, ACS1, and POX1 mRNA relative to ACT1 mRNA were determined. Values were plotted as a percentage of the expression in the corresponding WT CAT8 strain for each of the genes and represent the means of three biological replicates assayed in duplicate.

Model illustrating the activation of a glucose-repressed gene in the presence and the absence of Bmh. In low glucose, transcriptional activators, Adr1 (purple) and Cat8 (green), activate transcription. They bind the upstream activating sequences, UAS1 and UAS2, when the histones (gray) are hyperacetylated by the HATs (red). Subsequently, Adr1 and Cat8 each recruit a subset of coactivators and RNA Pol II (blue). Bmh (orange) binds to the regulatory domain of Adr1 at the promoter under derepressed conditions and weakly inhibits transcription. When Bmh is deleted, Adr1 is sufficient to activate transcription in the absence of Cat8. We propose that Bmh inhibits the cryptic activation domain (cAD) of Adr1, which leaves only the major activation domain (TADIII) available for recruiting coactivators and RNA Pol II. Therefore, Cat8 must provide the second activation domain to recruit the additional coactivators for optimal transcription. In the absence of Bmh, both the cAD and TADIII are available to activate transcription and are sufficient to recruit all the coactivators and RNA Pol II in the absence of Cat8.