Skip to main content
NCBI home page
Molecular Biology of the Cell logoLink to Molecular Biology of the Cell
. 2004 Sep;15(9):4191–4202. doi: 10.1091/mbc.E04-05-0412

Promoter-dependent Roles for the Srb10 Cyclin-dependent Kinase and the Hda1 Deacetylase in Tup1-mediated Repression in Saccharomyces cerevisiaeD⃞

Sarah R Green *, Alexander D Johnson *,†,‡
Editor: David Botstein
PMCID: PMC515351  PMID: 15240822

Abstract

The Tup1-Ssn6 complex has been well characterized as a Saccharomyces cerevisiae general transcriptional repressor with functionally conserved homologues in metazoans. These homologues are essential for cell differentiation and many other developmental processes. The mechanism of repression of all of these proteins remains poorly understood. Srb10 (a cyclin-dependent kinase associated with the Mediator complex) and Hda1 (a class I histone deacetylase) have each been implicated in Tup1-mediated repression. We present a statistically based genome-wide analysis that reveals that Hda1 partially represses roughly 30% of Tup1-repressed genes, whereas Srb10 kinase activity contributes to the repression of ∼15% of Tup1-repressed genes. These effects only partially overlap, suggesting that different Tup1-repression mechanisms predominate at different promoters. We also demonstrate a distinction between histone deacetylation and transcriptional repression. In an HDA1 deletion, many Tup1-repressed genes are hyperacetylated at lysine 18 of histone H3, yet are not derepressed, indicating deacetylation alone is not sufficient to repress most Tup1-controlled genes. In a strain lacking both Srb10 and Hda1 functions, more than half of the Tup1-repressed genes are still repressed, suggesting that Tup1-mediated repression occurs by multiple, partially overlapping mechanisms, at least one of which is unknown.

INTRODUCTION

The Tup1-Ssn6 complex is a general transcriptional repressor in Saccharomyces cerevisisae that controls a diverse set of genes generally characterized as being important for adaptation to nonstandard growth. Homologues of Tup1 have been identified in several other organisms (for example, unc-37 in Caenorhabditis elegans, Groucho in Drosophila, and TLE proteins in humans), and their repression functions are essential for embryonic development, cell differentiation, neurogenesis, and other developmental processes (Pflugrad et al., 1997; Fisher and Caudy, 1998; Levanon et al., 1998; Grbavec et al., 1999). Consequently, a better understanding of the mechanism of Tup1-mediated repression in yeast should illuminate this same process and its wide-ranging downstream consequences in other organisms. The Tup1-Ssn6 complex does not itself bind DNA but is recruited to target promoters through an association with sequence-specific DNA binding proteins; however, the crucial question of how transcriptional repression is established once this event occurs has not been clearly answered.

Two models for Tup1-mediated repression are supported by a number of earlier observations. One proposes that Tup1 produces a transcriptionally repressed chromatin state by recruiting histone deacetylases (HDACs). Hda1, a class I HDAC, has emerged as the most likely deacetylase to be acting with Tup1. Hda1 binds to Tup1 in vitro and an HDA1 deletion results in hyperacetylation of histones at several Tup1-controlled genes (Wu et al., 2001). Hyperacetylation of Tup1-repressed genes also is seen when Tup1 is deleted (Bone and Roth, 2001; Davie et al., 2002). Recently, a genomic analysis of hda1Δ-dependent hyperacetylation and Tup1-controlled genes expanded this correlation to a larger set of genes (Robyr et al., 2002). However, a clear link between loss of Hda1-mediated deacetylation activity and loss of transcriptional repression has been more difficult to establish.

A second model invokes a direct effect of Tup1 on the general transcriptional machinery. Tup1-mediated repression has been observed in an in vitro system using a naked DNA template (Herschbach et al., 1994; Redd et al., 1997). In addition, several components of the PolII transcriptional machinery (Rgr1, Sin4, Rox3, Hrs1, and Srbs8-11) have been identified in genetic screens for loss of Tup1-mediated repression (Sakai et al., 1990; Kuchin et al., 1995; Wahi and Johnson, 1995; Song et al., 1996; Carlson, 1997). A few also have been shown to physically interact with the Ssn6-Tup1 complex (Gromoller and Lehming, 2000; Papamichos-Chronakis et al., 2000). Defining their roles in Tup1-mediated repression has been difficult as many of them are essential and have wide-ranging effects on transcription in general. One component, Srb10, a nonessential cyclin-dependent kinase, is part of a distinct Mediator-associated complex (the Srb8-11 complex) that interacts with Tup1 (Myer and Young, 1998; Zaman et al., 2001; Borggrefe et al., 2002). Srb10 has been shown to negatively affect transcription, and its kinase function is necessary for full repression of a Tup1-controlled reporter construct (Holstege et al., 1998; Kuchin and Carlson, 1998; Song and Carlson, 1998; Lee et al., 2000). Furthermore, expression microarray experiments demonstrated some overlap between genes repressed by Srb10 and by Rox1, a DNA-binding protein that recruits the Tup1-Ssn6 complex and is responsible for repressing hypoxia-induced genes (Holstege et al., 1998; Becerra et al., 2002).

Previous work designed to dissect the mechanism of Tup1-mediated repression has concentrated mainly on the analysis of a few Tup1-repressed genes and reporter constructs. This piecemeal approach makes it difficult to determine the relative importance of the various mechanisms of Tup1-mediated repression at all Tup1-controlled genes. In particular, a genome-wide analysis that systematically investigates the contributions of both of the mechanisms represented by Hda1 and Srb10 functions has not been previously reported. This approach avoids the problems inherent in extrapolating a general mechanism of Tup1-mediated repression from the examination of only a few cases. Here, we describe the statistical analysis of gene expression microarrays of strains disrupted in all combinations of Tup1, Hda1, and Srb10 function. We have been able to divide the total set of Tup1-repressed genes into subclasses dependent on one, both, or neither of these mechanisms. As a result, it is clear that Srb10 and Hda1 are only two aspects of a complex, multilayer system for establishing Tup1-mediated gene repression.

MATERIALS AND METHODS

Yeast Strains

The S. cerevisiae strains used in this study were all generated from a parental strain of genotype MATα ura3-52, lys2-801amb, ade2-101och, leu2-Δ1, his3-Δ200, trp1-Δ1, which was a descendent of the original S288c strain. SGY201 (srb10D304) was made by transforming a full-length open reading frame (ORF) fragment containing the mutation into a strain in which the SRB10 locus has been replaced with URA3, leaving ≈200 bp of ORF homology on either side. Growth on 5-fluorootic acid selected for a strain in which the mutated SRB10 ORF had been integrated at the SRB10 genomic locus. SGY160 (hda1Δ), SGY84 (tup1Δ), and SGY203 (srb10D304hda1Δ) were constructed by transforming the parental strains with polymerase chain reaction (PCR) products of the TRP1 gene flanked by homologous sequences of the appropriate target gene. SGY167 (tup1Δhda1Δ) was made by crossing SGY160 and SGY83 (MATa tup1Δ) and SGY205 (srb10D304tup1Δhda1Δ) was made by repeating the protocol for mutating SRB10 described above but using SGY167 as the starting strain. The gene knockouts we made resulted in TRP1 strains, so we replaced the trp1-Δ1 mutation with TRP1 in SGY201 and the wild-type strain used in the microarrays (SGY92) so that all strains were matched for auxotrophies.

Microarrays and SAM Analysis

Microarrays of cDNA ORFs (∼6100 spots) were performed as described previously (http://derisilab.ucsf.edu/microarray/protocols.html). Briefly, mRNA was prepared from each strain and either labeled with Cy5 dye or mixed with the other mRNA samples and labeled with Cy3 dye to make a reference sample. Each microarray was hybridized with a mixture of a Cy5-labeled mRNA sample and a Cy3-labeled reference sample. After scanning on a GenePix4000A scanner, arrays were analyzed with GenePix 3.0 software. The data were normalized and filtered (sum of the median signal intensity ≥1000) by using NOMAD (http://derisilab5.ucsf.edu/NOMAD), and each spot's signal ratio (ratio of the median signal intensities) in the mutant strains was divided by its signal ratio in the wild-type control. Each of the microarrays was done four times (from independently grown cultures), except for the tup1Δ microarrays, which were done seven times (data available as Supplemental Tables 1 and 2). The set of repeats for each strain was then analyzed by SAM by using the One-Class Response and Row Average settings and the default Random Number Seed (1234567). Twenty-four permutations (the complete set for four repeats) were done for all mutant datasets, except tup1Δ upon which 5000 permutations were performed. The delta values for all datasets except hda1Δ were selected as the value that resulted in the lowest FDR calculated for the 90th percentile d-scores. The FDRs for the hda1Δ data were higher in general than those of the other mutants, most likely because of the relatively low levels of changes in expression in the dataset as a whole. The plot of the SAM analysis for the hda1Δ data suggested an alternative delta value selection. HDA1 was the only gene substantially down-regulated in the SAM Plot of the hda1Δ data, so we used its sole inclusion in the set of significantly down-regulated genes as the criterion for selecting the delta value. We chose the smallest delta value that still excluded any other gene from the set of down-regulated genes. This added 33 genes to the set of significant genes for the hda1Δ strain (and only increased the FDR by four percentage points) compared with the set of genes resulting from the delta value corresponding to the lowest FDR. The degree of overlap between the significant genes for the tup1Δ strain and the sets of significant genes for the hda1Δ strain by using these two delta values was essentially unchanged, so we elected to allow for a slightly higher FDR and to use the larger hda1Δ significant gene set.

Chromatin Immunoprecipitation (ChIP) and Quantitative PCR (Q-PCR)

Antibodies against acetylated lysine 18 of histone H3 were purchased from Upstate Biotechnology (Lake Placid, NY) (catalog no. 07-354). Cultures were cross-linked with formaldehyde for 5 min, and ChIPs were performed with slight modifications as described previously (Strahl-Bolsinger et al., 1997). Extract from 50 to 100 ml of culture at OD600 ∼1 was used for each immunoprecipitation (IP). Extracts were sonicated 10 times for 12 s by using a Branson sonifier 450 at 50% output power. ChIPs were quantitated by Q-PCR in a DNA Engine Opticon machine (MJ Research, Watertown, MA). PCR products were between 200 and 400 bp. Input ratios were calculated for each mutant strain versus wild-type α-cells to normalize the amount of total DNA added to each IP. The amount of immunoprecipitated DNA in each IP was normalized for input and then divided by the amount measured in an IP of the wild-type α-strain to produce a relative level of enrichment for each mutant. Q-PCR reactions were done at least twice for each gene from a particular ChIP experiment. Enrichment levels for each analyzed gene are averages of data from two to four independent ChIPs. Enrichment for a-specific genes was calculated separately for the two cell types, but for all other non-cell type controlled genes data for both a- and α-cells were averaged.

RESULTS

Deletion of TUP1 Derepresses a Large Group of Diverse Genes

One of the first articles describing the expression microarray technique included a set of data for a tup1Δ mutant (DeRisi et al., 1997). The wild-type control strain used in that study was later shown to have a duplicated chromosome XIII, which resulted in all the genes on chromosome XIII seeming to be slightly down-regulated in a tup1Δ strain (Hughes et al., 2000b). To correct for this strain abnormality and because microarray techniques and analysis have advanced since that first publication, we present a new set of expression microarray data for a tup1Δ mutant. The mutant constructed for this study is derived directly from the wild-type control strain. Our data represent seven duplicate experiments and have been analyzed using the significance analysis of microarrays (SAM) methodology (Tusher et al., 2001). SAM assigns each gene a d-score based on both its level of expression and reproducibility. All genes whose scores are higher than a selected threshold are then deemed significant. SAM also calculates a false discovery rate (FDR) for each threshold of significance. This value estimates the percentage of the genes with scores higher than a given threshold that are likely to be false positives. We decided to select a significance threshold that resulted in the lowest calculated FDR (based on the 90th percentile of d-scores) and, therefore, represented the highest confidence set of differentially expressed genes in the tup1Δ strain.

Three hundred and thirty-four genes passed this standard and are considered significantly derepressed in the tup1Δ mutant. Many of these genes also were reported in the initial published tup1Δ dataset (using a ≥2-fold increase in expression cut-off), but more than half were not. These newly described Tup1-repressed genes share a similar distribution among broad functional categories as the previously known Tup1-controlled genes (Table 1). For example, Table 2 lists more information about the newly reported genes found in the membrane transport, metabolism, cell wall, and stress response categories. The complete set of Tup1-repressed genes responds to very different signals and represents strategies for the cell to adjust to everything from simple changes in sugar availability to noxious environments. The statistical analysis of our multiple tup1Δ microarray experiments allowed us to determine a more comprehensive set of Tup1-controlled genes, which is vital for exposing the overall impact of different Tup1-repression mechanisms at regulated promoters.

Table 1.

GO classifications for Tup1-regulated gene

Not found in Derisi et al. set
Included in Derisi et al. set
Category No. % No. %
Unknown 79 41 56 39
Metabolism 34 18 21 15
Transport 23 12 25 18
Mating/meiosis 16 8 13 9
Cell wall 5 3 7 5
Stress response 9 5 9 6
Transcription 10 5 1 1
Kinase 3 2 2 1
Signal transduction 3 2 1 1
Miscellaneous 10 5 7 5
Total 192 142
Open in a new tab

Table 2.

Some previously unassigned Tup1-regulated genes

graphic file with name zmk00904281800t2.jpg

Open in a new tab

Overlap of tup1Δ, srb10D304, and hda1Δ Expression Profiles

As described in Introduction, a number of previous studies have demonstrated a link between Tup1-mediated repression and the functions of Srb10 and Hda1. We have applied the same techniques that we used to analyze the tup1Δ microarrays to determine the sets of genes derepressed when the functions of Srb10 and Hda1 are disrupted. We constructed an isogenic set of strains deleted for TUP1 or HDA1 or containing a mutant of Srb10 that lacks kinase activity (srb10D304) (Liao et al., 1995; Ansari et al., 2002). We also made a strain lacking both Hda1 and Srb10 functions to determine the combined effects of the two mutations on expression levels. Expression profiles of the mutant and wild-type strains (all grown to mid-log phase) were then compared in order to identify differentially regulated genes. We analyzed datasets representing at least four duplicates for each strain by using SAM and again adopted a significance threshold for each set of microarrays that corresponded to the lowest calculated FDR. The one exception was the HDA1 SAM analysis. Because the deletion of HDA1 affected many genes by only a small magnitude, we allowed a slightly higher FDR (see Materials and Methods).

We chose to analyze our data with SAM because the levels of expression changes we observed in the different mutant strains varied significantly, and there was no reasonable way to apply a uniform requirement of a fold-change in expression to each set of data. For example, a fairly standard cutoff of ≥2-fold change in expression worked well for identifying significant genes in the tup1Δ dataset, but it was not a practical measure of significance for the hda1Δ strain in which the expression of many genes increased only ∼1.5-fold. Rather than subjectively setting an expression threshold that was unique to each dataset, we chose the lowest FDR calculated by SAM as a universal standard for all datasets. Using SAM allowed us to apply a consistently stringent significance criterion that did not require all datasets to have similar ranges of expression changes.

Table 3 lists the total number of genes considered significant for each mutant and the FDR corresponding to that significance threshold. The largest portion of the significant genes for each disruption is up-regulated (derepressed) compared with a wild-type strain, consistent with the previously described roles of Tup1, Hda1, and Srb10 in transcriptional repression. To determine the roles Srb10 and Hda1 play in Tup1-mediated repression specifically, we compared the set of derepressed genes for the tup1Δ strain to those of each of the other three mutant strains. We focused only on derepressed genes because the down-regulation of genes in the tup1Δ strain is likely to be due to indirect effects. A substantial fraction (73%) of the genes derepressed upon deletion of HDA1 are also derepressed in the tup1Δ microarrays, suggesting that a primary transcriptional regulatory function of Hda1 is to repress Tup1-controlled genes (Figure 1A). However, less than one-third of Tup1-controlled genes are significantly derepressed in the hda1Δ strain, indicating there must be at least one Hda1-independent mechanism of Tup1-mediated repression. The overlap between the sets of derepressed genes identified in the srb10D304 and tup1Δ microarrays was smaller than that observed between the hda1Δ and tup1Δ datasets but still significant. Thirty-three percent of the significantly derepressed genes in the srb10D304 strain overlap with those derepressed in the tup1Δ dataset (Figure 1B). Clearly, Srb10 participates in many other modes of transcriptional repression that are independent of Tup1. For example, Srb10 has been shown to directly down-regulate the activity of several transcriptional activators, which could account for many of the genes derepressed in the srb10D304 microarrays (Chi et al., 2001). There is relatively little overlap between the hda1Δ and srb10D304 datasets (∼16-20%), which suggests they are parts of two separate mechanisms of Tup1-repression, both of which are required at only a relatively small number of genes.

Table 3.

Genes deemed significant by SAM

No. of significant genes
Total Up-regulated Down-regulated FDRa (%)
tup1 Δ 354 334 20 0.12
hda1 Δ 133 132 1 11.4
srb10D304 217 166 51 0.24
srb10D304hda1 Δ 327 277 50 0.13
Open in a new tab
a

False Discovery Rate calculated for 90th percentile d-scores, expressed as a percentage of the total number of significant genes.

Figure 1.

Figure 1.

Open in a new tab

Pairwise comparison of derepressed genes in tup1Δ, hda1Δ, srb10D304, and srb10D304hda1Δ strains. The area of the circles in the Venn diagrams are proportional to the number of significantly derepressed genes for each strain and depict the overlap of the genes for the corresponding pairs of strains: tup1Δ vs. (A) hda1Δ, (B) srb10D304, and (C) srb10D304hda1Δ. Numbers below strain names are the total number of significantly derepressed genes in that strain, and numbers within the Venn diagram reflect the number of genes that fall into that category. Cluster diagrams show representative genes (selected for those with data for each replicate of each strain) falling into the corresponding overlaps of datasets. Red represents an increase in gene expression and green represents a decrease in gene expression compared with a wild-type strain. Lists of the significantly derepressed genes in each mutant strain are available as Supplementary Table 3.

As expected the expression profile of the srb10D304hda1Δ double mutant exhibits a degree of overlap with Tup1-repressed genes (47%) that falls between those of each of the single mutants (Figure 1C). Interestingly, there are 32 Tup1-controlled genes that are only significantly derepressed when both SRB10 and HDA1 are disrupted, demonstrating that each of these mechanisms can compensate for the loss of the other at some promoters. Furthermore, 22 genes are derepressed in either mutant strain, providing evidence that for some genes Hda1 and Srb10 are both required for full repression. Each mutation disrupted the repression of many Tup1-controlled genes, but rarely was the level of derepression in the mutant strains equal to the full derepression measured in a tup1Δ strain (Figure 1). This is consistent with the premise that Tup1-mediated repression occurs through several mechanisms, including, but not limited to, those disrupted by the HDA1 and SRB10 mutations.

We performed the same microarray and SAM analysis described above on the double and triple mutants hda1Δtup1Δ, srb10D304tup1Δ, and srb10D304hda1Δtup1Δ. The expression profiles of Tup1-repressed genes in these strains closely resemble that of the tup1Δ strain (unpublished data). We conclude that both Hda1 and Srb10 are working through Tup1 to cause repression at Tup1-controlled genes and do not represent independent mechanisms of repression acting on Tup1-repressed genes. We also used microarrays to compare the effects on transcription of the srb10D304 mutation and an SRB10 deletion. We saw no significant difference in the expression patterns of these two mutants (unpublished data) and conclude that the kinase activity of Srb10 accounts for the transcriptional effects observed in this study.

DNA-Binding Proteins Do Not Dictate Mechanism of Tup1-mediated Repression

The microarray experiments presented above allow us to divide the larger set of Tup1-controlled genes into five subclasses based on the influence of the other mutations on gene expression. These subclasses are described as follows: 1) genes derepressed in the tup1Δ and srb10D304 strains; 2) genes derepressed in the tup1Δ and srb10D304hda1Δ strains but not the single hda1Δ and srb10D304 mutant strains; 3) genes derepressed in the tup1Δ, hda1Δ, srb10D304, and srb10D304hda1Δ strains; (4) genes derepressed in the tup1Δ and hda1Δ strains; and 5) genes derepressed in the tup1Δ strain but none of the other mutant strains (Figure 2). These subclasses of Tup1-controlled genes represent the first genome-wide evidence that there are different sets of repression mechanisms acting at different genes.

Figure 2.

Figure 2.

Open in a new tab

Distribution of Mig1- and Rox1-repressed genes across subclasses of Tup1-repressed genes. Four independent microarrays are shown for each strain; displayed are all significantly up-regulated genes in the tup1Δ microarray that had data for all 16 arrays represented; genes were clustered by their d-scores calculated by SAM. Five subclasses of Tup1-repressed genes are displayed representing genes derepressed in 1) tup1Δ and srb10D304, 2) tup1Δ and srb10D304hda1Δ, 3) tup1Δ, hda1Δ, srb10D304, and srb10D304hda1Δ, 4) tup1Δ and hda1Δ, and 5) tup1Δ only. Mig1- and Rox1-controlled genes for which there is some evidence of direct regulation are identified by their gene names. Red and green colors represent an increase and a decrease in expression respectively compared with a wild-type strain.

One possible characteristic of Tup1-repressed promoters that could dictate the repression mechanism(s) in use at that gene is the identity of the DNA-binding protein that recruits the Tup1-Ssn6 complex to the promoter. Tup1 does not itself bind to DNA, but instead is recruited to promoters by corepressors specific for the various classes of Tup1-repressed genes. Several of these sequence-specific DNA-binding proteins have been identified and, for a few, a set of direct target genes has been defined. Two well-characterized Tup1 corepressors are Mig1 and Rox1, the sequence-specific DNA-binding proteins controlling glucose-repressed and hypoxia-induced genes, respectively (Lowry et al., 1990; Zitomer and Lowry, 1992; Balasubramanian et al., 1993; Amillet et al., 1995; Treitel and Carlson, 1995; Ozcan and Johnston, 1996; Deckert et al., 1998; Lutfiyya et al., 1998; Johnston, 1999; Lee et al., 2002). If the sequence-specific DNA-binding protein were responsible for determining which Tup1-repression mechanism(s) acts at a particular gene, then all of the genes controlled by that DNA-binding protein might be expected to fall into the same subclass of Tup1-repressed genes. To test this possibility, we mapped well-documented Mig1- and Rox1-controlled genes onto a cluster diagram of Tup1-repressed genes that reflects the five subclasses (Figure 2). Both the Mig1- and the Rox1-controlled genes are found throughout the cluster and across multiple subgroups. We conclude that genes repressed by the same DNA-binding protein can have different requirements for Hda1 and Srb10, and therefore the identity of the DNA-binding protein is not likely to determine the mechanism(s) of repression used at a particular promoter.

Chromosomal Position Bias of Tup1-controlled Genes

We wanted to know whether the mechanism(s) of Tup1-repression acting at a particular gene could be dictated by the position of that gene along its chromosome. For example, previous work has shown that the hyperacetylation seen upon deletion of HDA1 is concentrated in regions of the genome within 25 kb of a chromosome end (subtelomeric regions) (Robyr et al., 2002). Our expression microarrays show this same subtelomeric bias for the genes that are derepressed upon deletion of HDA1 (Figure 3A). Approximately one-third of all genes derepressed in the hda1Δ mutant lie within these subtelomeric regions, whereas only ∼6% of all genes are contained within this same region. We see a similar bias for the set of genes derepressed in a tup1Δ strain. Thirty percent of Tup1-repressed genes identified in our microarray experiments are subtelomeric, 5 times higher than the 6% predicted for a random chromosomal distribution. The genes derepressed in an srb10D304 strain, however, do not exhibit this subtelomeric bias and, in fact, have the same positional distribution as the total genome (Figure 3A).

Figure 3.

Figure 3.

Open in a new tab

Overlap of repressed subtelomeric genes. (A) Table shows the total number of significantly derepressed genes both within and outside of subtelomeric regions (≤25kb from a chromosome end) for each mutant strain. (B) Venn diagrams depicting the overlap between only the subtelomeric genes that are significantly derepressed in tup1Δ and each mutant strain. (C) Table shows the percentage of the total set of Tup1-repressed genes either within or outside of subtelomeric regions that are shared by the corresponding mutant strain.

The overlaps we observed among our microarrays increase significantly when only considering the subtelomeric genes (Figure 3). For example, ∼90% of the subtelomeric genes affected by Hda1 or Srb10 are also Tup1-repressed genes. In other words, the transcriptional functions of Hda1 and Srb10 in subtelomeric regions seem more dedicated to Tup1-mediated repression than they are at internal chromosome positions. However, this increase in the overlap between the gene sets probably reflects the density of Tup1-repressed genes in subtelomeric regions rather than any mechanistic bias. Each mutant strain showed roughly the same percentage of overlap with the total set of Tup1-repressed genes whether considering only subtelomeric genes or all genes (Figure 3C).

Loss of Deacetylation by Hda1 Is Not Sufficient for Loss of Tup1-mediated Repression

Previous work described a correlation between the set of promoters that are hyperacetylated upon the deletion of HDA1 and the set of genes repressed by Tup1. Hda1 preferentially deacetylates histones H3 and H2B (on positions K9, K14, K18, K23, K27 of H3 and positions K11 and K16 on H2B), and these same residues are hyperacetylated at a Tup1-repressed gene (ENA1) when TUP1 is deleted (Wu et al., 2001). ChIP microarrays also have demonstrated that the pattern of genes derepressed upon TUP1 deletion significantly overlaps with acetylation patterns resulting from an HDA1 deletion (Robyr et al., 2002). These data, in addition to evidence of a physical interaction between Tup1 and Hda1, suggest that Tup1 recruits Hda1 to promoters to repress transcription (Wu et al., 2001). The substantial overlap between the tup1Δ and hda1Δ microarray datasets is also consistent with the idea that at least one function of Hda1 is to repress transcription at Tup1-controlled promoters.

To further examine the relationship between hyperacetylation and derepression, we compared the acetylation of lysine 18 of histone H3 (H3-K18) in mutant and wild-type strains by ChIP. We selected a representative set of Tup1-repressed promoters, shown by ChIP to be directly controlled by Tup1 (unpublished data), that included examples from each of the five subclasses defined by our microarrays. MFA1 and MAL12 represent genes derepressed upon deletion of Tup1, but whose expression is not significantly affected by either an HDA1 or SRB10 disruption. HXT16 is derepressed in only the tup1Δ and hda1Δ strains, whereas HSP12 is derepressed in only the tup1Δ and srb10D304 strains. Finally, CYC7 and SPI1 are derepressed in all three mutant strains (Figure 4B).

Figure 4.

Figure 4.

Open in a new tab

Acetylation of histone H3 at Tup1-repressed genes. Chromatin IPs were carried out for acetylated K18 of histone H3. The y-axis represents the fold increase in acetylation in a mutant strain compared with a wild-type strain. Bars represent the average of at least six repeats in A and 12 repeats in C; data for C includes measurements for both mating types. Error bars reflect the SE calculation for the averaged data. Table in B describes the effect of each of the mutations on the expression of the genes tested by ChIP.

In the tup1Δ strain, all of the promoters we examined were transcriptionally derepressed and hyperacetylated at H3-K18 compared with a wild-type strain (Figure 4, A and C). We next examined the effect of an HDA1 deletion on H3-K18 acetylation at Tup1-controlled genes. In the experiments shown in Figure 4, all of the Tup1-controlled promoters we tested are hyperacetylated at H3-K18 in an hda1Δ strain compared with a wild-type strain. However, this hyperacetylation does not correlate with the derepression of these genes in an hda1Δ strain (Figure 4B). The observation that some promoters (i.e., MFA1, HSP12, and MAL12) can be hyperacetylated at H3-K18 but still remain transcriptionally repressed by Tup1 suggests that an additional Tup1-mediated repression mechanism is at work at these promoters. It cannot be the mechanism defined by Srb10 because only a few of these genes are derepressed when its kinase function is disrupted. Thus, it seems that Hda1 is functioning at most (if not all) promoters directly repressed by Tup1 as part of a multicomponent repression mechanism that can maintain significant transcriptional repression even when one arm of the machinery has been disrupted.

We tested whether Hda1 is the only deacetylase responsible for the hyperacetylation we observed at Tup1-repressed promoters. ChIP experiments measuring acetylated H3-K18 in a tup1Δhda1Δ double deletion strain showed no increase in the level of hyperacetylation at these promoters in the double mutant versus a single tup1Δ mutant strain. This confirms that the H3-K18 hyperacetylation resulting from a TUP1 deletion is dependent on Hda1 and that Hda1 is acting in concert with Tup1 at these promoters to produce repression.

Finally, we analyzed a gene that is indirectly controlled by Tup1, FIG1, to demonstrate that hyperacetylation at H3-K18 is a result of the loss of Tup1 (and consequently Hda1) rather than an increase in transcription (unpublished data; Erdman et al., 1998). The FIG1 promoter is not hyperacetylated at H3-K18 in either the tup1Δ or hda1Δ strains compared with a wild-type strain despite the fact that its expression is induced in the tup1Δ mutant (Figure 4A). Therefore, we conclude the hyperacetylation at H3-K18 that we observe at other genes is likely due to a specific loss of deacetylase activity rather than an indirect result of increased transcription.

DISCUSSION

The Tup1-Ssn6 complex is a general repressor of transcription in S. cerevisiae that is recruited to hundreds of promoters through association with sequence-specific DNA-binding proteins. Several repetitions of expression microarrays and statistical analysis of the results have allowed us to compile a more complete list of genes repressed by the Tup1-Ssn6 complex. This analysis takes advantage of the improvements in microarray technology and analysis since the original publication of the tup1Δ mutant study and has added nearly 200 genes (Table 1) to that earlier list of Tup1-controlled genes. The newly assigned Tup1-controlled genes fall into roughly the same functional categories as those of the previous list of Tup1-controlled genes, and therefore we believe this represents an expansion of the set of Tup1-repressed genes rather than the identification of new networks of genes. Whereas uncharacterized genes still represent the largest portion of Tup1-repressed genes, we also saw an abundance of genes involved in cellular metabolism, membrane transport, and cell wall organization. Within the category of transport, for example, are transporters of everything from glycerol and sugars to water and ferrochromes. The fact that genes coding for proteins of such diverse specificities and sensitivities are all repressed by Tup1 reflect its role in mediating the cell's adaptive response to the external environment.

In this article, we address several aspects of the mechanism of Tup1-mediated repression subsequent to promoter recruitment by disrupting known components of the repression machinery. Careful, statistically based microarray analysis has also allowed us to compare the effects of a Tup1 deletion to those observed upon disruption of two previously described Tup1-repression mechanisms. First, the expression microarrays clearly reflect each protein's overall role in transcription. Tup1's role as a transcriptional repressor is made obvious by the fact that >94% of the genes whose expression is significantly altered in a tup1Δ strain are up-regulated. Similarly, Hda1 seems to act primarily as a transcriptional repressor. In fact, no genes (save HDA1 itself) are significantly down-regulated in the hda1Δ strain (Table 3). In contrast to Hda1 and Tup1, Srb10 positively affects the expression of a substantial number of genes in addition to its role as a transcriptional repressor.

Second, we can examine the extent of the functional overlap between Tup1 and the two mechanisms represented by Hda1 and Srb10. The Hda1-repressed genes are almost entirely included within the set of genes repressed by Tup1. Conversely, deletion of HDA1 significantly affects the expression of less than one-third of Tup1-repressed genes. These data suggest that Hda1's main role in transcriptional regulation is to function with Tup1, whereas Tup1-mediated repression at most genes is not solely dependent on Hda1 function. Approximately 16% of Tup1-repressed genes are significantly derepressed by inactivation of the Srb10 kinase, and similarly most of the Srb10-repressed genes are not affected by a TUP1 deletion. These results indicate Srb10 plays roles in both Tup1-dependent and Tup1-independent transcriptional repression. Our microarray comparisons identified five subclasses of Tup1-controlled genes defined by their dependence (or lack thereof) on Hda1 and Srb10 function (Figure 2). Some Tup1-repressed genes are responsive to either of the mechanisms involving Srb10 and Hda1, but there are also genes (∼10% of the total set of Tup1-repressed genes) that are only derepressed when both mechanisms are disrupted, indicating both of these mechanisms are at work at these promoters.

The substantial overlap observed between the genes repressed by Hda1 and those repressed by Tup1 raises several issues about the role of deacetylation in repression. Two sets of Hda1-affected genes can be defined. Previous work identified a set of genes whose promoters become hyperacetylated at H3-K18 upon deletion of HDA1 (Robyr et al., 2002). The experiments in our work describe a set of genes that are derepressed upon deletion of HDA1, which constitutes only a subset of the hyperacetylated genes. Seventy-three percent of the Hda1-repressed genes also are repressed by Tup1, whereas only a minority of the genes deacetylated by Hda1 also are regulated by Tup1. This difference suggests that although there could be other roles for Hda1-mediated H3-K18 deacetylation in addition to transcriptional repression, at promoters for which Hda1-mediated deacetylation is a requirement for repression, Tup1 is responsible for that repression.

This distinction between Hda1-mediated deacetylation and Hda1-mediated repression is further supported by our ChIP experiments. We saw no correlation between hda1Δ-dependent transcriptional derepression and hda1Δ-dependent H3-K18 hyperacetylation when examining the acetylation of H3-K18 at various genes in the five subclasses of Tup1-repressed genes. Although deletion of TUP1 always resulted in both hyperacetylation and increased expression, deletion of HDA1 always caused hyperacetylation but did not always lead to increased transcription. It seems likely that Hda1-mediated deacetylation is one of several complementary mechanisms working to repress transcription at Tup1-controlled promoters (Figure 5). The concept of multiple factors converging at Tup1 to produce repression has been proposed previously and seems to fit well with the regulatory requirements of a repressor of diverse gene sets (Carlson, 1997; Lee et al., 2000; Papamichos-Chronakis et al., 2000; Smith and Johnson, 2000; Schreiber and Bernstein, 2002).

Figure 5.

Figure 5.

Open in a new tab

Model of multiple Tup1-mediated repression mechanisms. This model depicts only the dependency of repression at a particular gene on Hda1 and Srb10 function and does not imply anything about the presence of either factor at Tup1-repressed promoters. It is possible that Hda1 and Srb10 are still present at genes that are not significantly derepressed when either Hda1 or Srb10 function is disrupted. It remains to be proven whether Factor X is indeed one or more distinct components or an as yet undocumented role of Tup1 itself.

We attempted to identify traits of Tup1-controlled genes that might dictate which mechanisms are important for transcriptional repression at that gene. First, we examined whether control by a certain sequence-specific DNA-binding protein correlated with one mechanism or another. Genes controlled by a common DNA-binding protein are found in all subclasses of Tup1-repressed genes, and therefore do not seem to be subject to a particular repression mechanism simply because of their shared regulation. The position of the genes in the genome also had little predictive value for placing a gene in a particular subclass. However, we did observe a general bias in the occurrence of Tup1-repressed genes in these subtelomeric regions. It is possible that proximity to a telomere somehow facilitates the establishment or maintenance of Tup1-repression; however, most Tup1-controlled genes are found at internal chromosomal positions and repression is maintained there as well. Recent work describing the synteny among closely related Saccharomyces species suggests an intriguing explanation for the propensity of Tup1-repressed genes to be found in subtelomeric regions. Kellis et al. (2003) note that the remarkably conserved synteny between these genomes breaks down close to the telomeres, which seem to be areas of rapid genomic evolution. The authors also point out the occurrence of several large gene families in these regions and even mention some that are repressed by Tup1 (the HXT, FLO, PAU, and THI families). It is possible that Tup1's subtelomeric bias began as a few Tup1-repressed genes found within this region of genomic flexibility that then expanded into evolutionarily advantageous gene families while maintaining their Tup1-controlled gene expression.

Another possibility for a characteristic of Tup1-controlled promoters that could influence which repression mechanisms are important at a particular gene is the composition of the general transcriptional machinery regulating expression at that promoter. For example, Basehoar et al. (2004) and Huisinga and Pugh (2004) identified a set of genes (∼10% of the genome) whose regulation is dominated by the SAGA complex rather than the TFIID complex and showed that Tup1-controlled genes disproportionately fall into this category. However, all five subclasses of Tup1-repressed genes we describe in this article exhibit this same propensity for SAGA-dominated transcriptional regulation, so although inclusion in this group does seem to be a characteristic of Tup1-repressed genes, it does not seem to dictate the influence of a particular repression mechanism.

Another model, proposed by Edmondson et al. (1996), is that Tup1 interacts directly with histones to repress transcription. However, recent work has demonstrated effective repression by a Tup1 protein lacking most of the described histone-binding domain (Zhang et al., 2002). In addition, our own initial microarray analysis of a strain containing this internally deleted Tup1 allele (Δ129-282aa) revealed no affect on Tup1-mediated repression (unpublished data). Similarly, microarrays of a strain with the histone H3 tail deleted (Δ1-28) show that the tails are important for the repression of a large set of genes but that only a minority of these are repressed by Tup1 (Sabet et al., 2003).

Another recent report linked repression of a-specific genes, a Tup1-repressed set of genes, to a chromatin remodeling complex, Isw2-Itc1 (Ruiz et al., 2003). However, derepression in an Itc1 deletion is not complete, and microarrays of an isw2Δ strain do not demonstrate this loss of repression for the larger set of Tup1-controlled genes (Hughes et al., 2000a). It seems that this mechanism does not apply to the broader set of all Tup1-repressed genes. Finally, it is possible that the presence at a promoter of the Tup1-Ssn6 complex, which measures ∼450 kDa, could itself be sufficient to generate a significant degree of repression simply by interfering with transcriptional initiation conditions (Varanasi et al., 1996; Redd et al., 1997). This could be the mechanism responsible for the ability of the Tup1-Ssn6 complex to repress transcription initiation on a naked DNA template and could constitute the missing repression mechanism. However, recent work shows Tup1 remains bound to some promoters even under inducing conditions, seemingly discounting this idea, although it is not know from these studies whether the complex remains intact (Papamichos-Chronakis et al., 2002; Proft and Struhl, 2002; Mennella et al., 2003).

We believe the work presented here substantially clarifies the picture of Tup1-mediated transcriptional repression. Previous work concentrated on the analysis of a few carefully selected promoters and sometimes resulted in conflicting conclusions. Two basic mechanisms arose from this work, one based on the deacetylation of histones and the other involving the mediator, although the degree of intersection between the models was not established. We have provided a comprehensive, genome-wide picture of how these two mechanisms affect all Tup1-mediated repression. Much like the emerging picture of transcriptional activation, transcriptional repression seems to be more complicated than previously appreciated. Of the 334 Tup1-repressed genes identified in this study, few were fully derepressed (as measured by microarray) by the simultaneous disruptions of HDA1 and SRB10. Moreover, again by microarray analysis, full levels of repression for more than one-half of Tup1-controlled genes are maintained even when these two repression mechanisms are disrupted. Our experiments point to a model of Tup1-mediated repression that is the result of several, functionally overlapping mechanisms whose relative importance for overall repression varies at different genes.

Supplementary Material

[Supplemental Tables]
mbc_E04-05-0412_index.html (1.1KB, html)

Acknowledgments

We thank Adam Carroll and the University of California, San Francisco, Core Facility for Genomics and Proteomics, for valuable advice and assistance with microarray construction. We thank Joachim Li for generously providing strains and Virgil Rhodius for constructive advice and assistance with SAM analysis. We thank Hiten Madhani and Anita Sil for helpful comments on the manuscript and are grateful to Anita Sil for helpful discussions about all aspects of this work. S.G. was supported by a Howard Hughes Predoctoral Fellowship. A.J. is supported by a grant from the National Institutes of Health (GM-37049).

Article published online ahead of print. Mol. Biol. Cell 10.1091/mbc.E04-05-0412. Article and publication date are available at www.molbiolcell.org/cgi/doi/10.1091/mbc.E04-05-0412.

D⃞

Online version of this article contains supporting material. Online version is available at www.molbiolcell.org.

References

  1. Amillet, J.M., Buisson, N., and Labbe-Bois, R. (1995). Positive and negative elements involved in the differential regulation by heme and oxygen of the HEM13 gene (coproporphyrinogen oxidase) in Saccharomyces cerevisiae. Curr. Genet. 28, 503-511. [DOI] [PubMed] [Google Scholar]
  2. Ansari, A.Z., Koh, S.S., Zaman, Z., Bongards, C., Lehming, N., Young, R.A., and Ptashne, M. (2002). Transcriptional activating regions target a cyclin-dependent kinase. Proc. Natl. Acad. Sci. USA 99, 14706-14709. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Balasubramanian, B., Lowry, C.V., and Zitomer, R.S. (1993). The Rox1 repressor of the Saccharomyces cerevisiae hypoxic genes is a specific DNA-binding protein with a high-mobility-group motif. Mol. Cell. Biol. 13, 6071-6078. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Basehoar, A.D., Zanton, S.J., and Pugh, B.F. (2004). Identification and distinct regulation of yeast TATA box-containing genes. Cell 116, 699-709. [DOI] [PubMed] [Google Scholar]
  5. Becerra, M., Lombardia-Ferreira, L.J., Hauser, N.C., Hoheisel, J.D., Tizon, B., and Cerdan, M.E. (2002). The yeast transcriptome in aerobic and hypoxic conditions: effects of hap1, rox1, rox3 and srb10 deletions. Mol. Microbiol. 43, 545-555. [DOI] [PubMed] [Google Scholar]
  6. Bone, J.R., and Roth, S.Y. (2001). Recruitment of the yeast Tup1p-Ssn6p repressor is associated with localized decreases in histone acetylation. J. Biol. Chem. 276, 1808-1813. [DOI] [PubMed] [Google Scholar]
  7. Borggrefe, T., Davis, R., Erdjument-Bromage, H., Tempst, P., and Kornberg, R.D. (2002). A complex of the Srb8, -9, -10, and -11 transcriptional regulatory proteins from yeast. J. Biol. Chem. 277, 44202-44207. [DOI] [PubMed] [Google Scholar]
  8. Brachmann, C.B., Davies, A., Cost, G.J., Caputo, E., Li, J., Hieter, P., and Boeke, J.D. (1998). Designer deletion strains derived from Saccharomyces cerevisiae S288C: a useful set of strains and plasmids for PCR-mediated gene disruption and other applications. Yeast 14, 115-132. [DOI] [PubMed] [Google Scholar]
  9. Carlson, M. (1997). Genetics of transcriptional regulation in yeast: connections to the RNA polymerase II CTD. Annu. Rev. Cell Dev. Biol. 13, 1-23. [DOI] [PubMed] [Google Scholar]
  10. Chi, Y., Huddleston, M.J., Zhang, X., Young, R.A., Annan, R.S., Carr, S.A., and Deshaies, R.J. (2001). Negative regulation of Gcn4 and Msn2 transcription factors by Srb10 cyclin-dependent kinase. Genes Dev. 15, 1078-1092. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Davie, J.K., Trumbly, R.J., and Dent, S.Y. (2002). Histone-dependent association of Tup1-Ssn6 with repressed genes in vivo. Mol. Cell. Biol. 22, 693-703. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Deckert, J., Torres, A.M., Hwang, S.M., Kastaniotis, A.J., and Zitomer, R.S. (1998). The anatomy of a hypoxic operator in Saccharomyces cerevisiae. Genetics 150, 1429-1441. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. DeRisi, J.L., Iyer, V.R., and Brown, P.O. (1997). Exploring the metabolic and genetic control of gene expression on a genomic scale. Science 278, 680-686. [DOI] [PubMed] [Google Scholar]
  14. Edmondson, D.G., Smith, M.M., and Roth, S.Y. (1996). Repression domain of the yeast global repressor Tup1 interacts directly with histones H3 and H4. Genes Dev. 10, 1247-1259. [DOI] [PubMed] [Google Scholar]
  15. Erdman, S., Lin, L., Malczynski, M., and Snyder, M. (1998). Pheromone-regulated genes required for yeast mating differentiation. J. Cell Biol. 140, 461-483. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Fisher, A.L., and Caudy, M. (1998). Groucho proteins: transcriptional corepressors for specific subsets of DNA-binding transcription factors in vertebrates and invertebrates. Genes Dev. 12, 1931-1940. [DOI] [PubMed] [Google Scholar]
  17. Grbavec, D., Lo, R., Liu, Y., Greenfield, A., and Stifani, S. (1999). Groucho/transducin-like enhancer of split (TLE) family members interact with the yeast transcriptional co-repressor SSN6 and mammalian SSN6-related proteins: implications for evolutionary conservation of transcription repression mechanisms. Biochem. J. 337, 13-17. [PMC free article] [PubMed] [Google Scholar]
  18. Gromoller, A., and Lehming, N. (2000). Srb7p is a physical and physiological target of Tup1p. EMBO J. 19, 6845-6852. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Herschbach, B.M., Arnaud, M.B., and Johnson, A.D. (1994). Transcriptional repression directed by the yeast alpha 2 protein in vitro. Nature 370, 309-311. [DOI] [PubMed] [Google Scholar]
  20. Holstege, F.C., Jennings, E.G., Wyrick, J.J., Lee, T.I., Hengartner, C.J., Green, M.R., Golub, T.R., Lander, E.S., and Young, R.A. (1998). Dissecting the regulatory circuitry of a eukaryotic genome. Cell 95, 717-728. [DOI] [PubMed] [Google Scholar]
  21. Hughes, T.R., et al. (2000a). Functional discovery via a compendium of expression profiles. Cell 102, 109-126. [DOI] [PubMed] [Google Scholar]
  22. Hughes, T.R., et al. (2000b). Widespread aneuploidy revealed by DNA microarray expression profiling. Nat. Genet. 25, 333-337. [DOI] [PubMed] [Google Scholar]
  23. Huisinga, K.L., and Pugh, B.F. (2004). A genome-wide housekeeping role for TFIID and a highly regulated stress-related role for SAGA in Saccharomyces cerevisiae. Mol. Cell 13, 573-585. [DOI] [PubMed] [Google Scholar]
  24. Johnston, M. (1999). Feasting, fasting and fermenting. Glucose sensing in yeast and other cells. Trends Genet. 15, 29-33. [DOI] [PubMed] [Google Scholar]
  25. Kellis, M., Patterson, N., Endrizzi, M., Birren, B., and Lander, E.S. (2003). Sequencing and comparison of yeast species to identify genes and regulatory elements. Nature 423, 241-254. [DOI] [PubMed] [Google Scholar]
  26. Kuchin, S., and Carlson, M. (1998). Functional relationships of Srb10-Srb11 kinase, carboxy-terminal domain kinase CTDK-I, and transcriptional corepressor Ssn6-Tup1. Mol. Cell. Biol. 18, 1163-1171. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Kuchin, S., Yeghiayan, P., and Carlson, M. (1995). Cyclin-dependent protein kinase and cyclin homologs SSN3 and SSN8 contribute to transcriptional control in yeast. Proc. Natl. Acad. Sci. USA 92, 4006-4010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Lee, M., Chatterjee, S., and Struhl, K. (2000). Genetic analysis of the role of Pol II holoenzyme components in repression by the Cyc8-Tup1 corepressor in yeast. Genetics 155, 1535-1542. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Lee, T.I., et al. (2002). Transcriptional regulatory networks in Saccharomyces cerevisiae. Science 298, 799-804. [DOI] [PubMed] [Google Scholar]
  30. Levanon, D., Goldstein, R.E., Bernstein, Y., Tang, H., Goldenberg, D., Stifani, S., Paroush, Z., and Groner, Y. (1998). Transcriptional repression by AML1 and LEF-1 is mediated by the TLE/Groucho corepressors. Proc. Natl. Acad. Sci. USA 95, 11590-11595. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Liao, S.M., Zhang, J., Jeffery, D.A., Koleske, A.J., Thompson, C.M., Chao, D.M., Viljoen, M., van Vuuren, H.J., and Young, R.A. (1995). A kinase-cyclin pair in the RNA polymerase II holoenzyme. Nature 374, 193-196. [DOI] [PubMed] [Google Scholar]
  32. Lowry, C.V., Cerdan, M.E., and Zitomer, R.S. (1990). A hypoxic consensus operator and a constitutive activation region regulate the ANB1 gene of Saccharomyces cerevisiae. Mol. Cell. Biol. 10, 5921-5926. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Lutfiyya, L.L., Iyer, V.R., DeRisi, J., DeVit, M.J., Brown, P.O., and Johnston, M. (1998). Characterization of three related glucose repressors and genes they regulate in Saccharomyces cerevisiae. Genetics 150, 1377-1391. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Mennella, T.A., Klinkenberg, L.G., and Zitomer, R.S. (2003). Recruitment of Tup1-Ssn6 by yeast hypoxic genes and chromatin-independent exclusion of TATA binding protein. Eukaryot. Cell 2, 1288-1303. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Myer, V.E., and Young, R.A. (1998). RNA polymerase II holoenzymes and subcomplexes. J. Biol. Chem. 273, 27757-27760. [DOI] [PubMed] [Google Scholar]
  36. Ozcan, S., and Johnston, M. (1996). Two different repressors collaborate to restrict expression of the yeast glucose transporter genes HXT2 and HXT4 to low levels of glucose. Mol. Cell. Biol. 16, 5536-5545. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Papamichos-Chronakis, M., Conlan, R.S., Gounalaki, N., Copf, T., and Tzamarias, D. (2000). Hrs1/Med3 is a Cyc8-Tup1 corepressor target in the RNA polymerase II holoenzyme. J. Biol. Chem. 275, 8397-8403. [DOI] [PubMed] [Google Scholar]
  38. Papamichos-Chronakis, M., Petrakis, T., Ktistaki, E., Topalidou, I., and Tzamarias, D. (2002). Cti6, a PHD domain protein, bridges the Cyc8-Tup1 corepressor and the SAGA coactivator to overcome repression at GAL1. Mol. Cell 9, 1297-1305. [DOI] [PubMed] [Google Scholar]
  39. Pflugrad, A., Meir, J.Y., Barnes, T.M., and Miller, D.M., 3rd. (1997). The Groucho-like transcription factor UNC-37 functions with the neural specificity gene unc-4 to govern motor neuron identity in C. elegans. Development 124, 1699-1709. [DOI] [PubMed] [Google Scholar]
  40. Proft, M., and Struhl, K. (2002). Hog1 kinase converts the Sko1-Cyc8-Tup1 repressor complex into an activator that recruits SAGA and SWI/SNF in response to osmotic stress. Mol. Cell 9, 1307-1317. [DOI] [PubMed] [Google Scholar]
  41. Redd, M.J., Arnaud, M.B., and Johnson, A.D. (1997). A complex composed of tup1 and ssn6 represses transcription in vitro. J. Biol. Chem. 272, 11193-11197. [DOI] [PubMed] [Google Scholar]
  42. Robyr, D., Suka, Y., Xenarios, I., Kurdistani, S.K., Wang, A., Suka, N., and Grunstein, M. (2002). Microarray deacetylation maps determine genome-wide functions for yeast histone deacetylases. Cell 109, 437-446. [DOI] [PubMed] [Google Scholar]
  43. Ruiz, C., Escribano, V., Morgado, E., Molina, M., and Mazon, M.J. (2003). Cell-type-dependent repression of yeast a-specific genes requires Itc1p, a subunit of the Isw2p-Itc1p chromatin remodelling complex. Microbiology 149, 341-351. [DOI] [PubMed] [Google Scholar]
  44. Sabet, N., Tong, F., Madigan, J.P., Volo, S., Smith, M.M., and Morse, R.H. (2003). Global and specific transcriptional repression by the histone H3 amino terminus in yeast. Proc. Natl. Acad. Sci. USA 100, 4084-4089. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Sakai, A., Shimizu, Y., Kondou, S., Chibazakura, T., and Hishinuma, F. (1990). Structure and molecular analysis of RGR1, a gene required for glucose repression of Saccharomyces cerevisiae. Mol. Cell. Biol. 10, 4130-4138. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Schreiber, S.L., and Bernstein, B.E. (2002). Signaling network model of chromatin. Cell 111, 771-778. [DOI] [PubMed] [Google Scholar]
  47. Smith, R.L., and Johnson, A.D. (2000). Turning genes off by Ssn6-Tup 1, a conserved system of transcriptional repression in eukaryotes. Trends Biochem. Sci. 25, 325-330. [DOI] [PubMed] [Google Scholar]
  48. Song, W., and Carlson, M. (1998). Srb/mediator proteins interact functionally and physically with transcriptional repressor Sfl1. EMBO J. 17, 5757-5765. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Song, W., Treich, I., Qian, N., Kuchin, S., and Carlson, M. (1996). SSN genes that affect transcriptional repression in Saccharomyces cerevisiae encode SIN4, ROX3, and SRB proteins associated with RNA polymerase II. Mol. Cell. Biol. 16, 115-120. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Strahl-Bolsinger, S., Hecht, A., Luo, K., and Grunstein, M. (1997). SIR2 and SIR4 interactions differ in core and extended telomeric heterochromatin in yeast. Genes Dev. 11, 83-93. [DOI] [PubMed] [Google Scholar]
  51. Treitel, M.A., and Carlson, M. (1995). Repression by SSN6-TUP1 is directed by MIG1, a repressor/activator protein. Proc. Natl. Acad. Sci. USA 92, 3132-3136. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Tusher, V.G., Tibshirani, R., and Chu, G. (2001). Significance analysis of microarrays applied to the ionizing radiation response. Proc. Natl. Acad. Sci. USA 98, 5116-5121. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Varanasi, U.S., Klis, M., Mikesell, P.B., and Trumbly, R.J. (1996). The Cyc8 (Ssn6)-Tup1 corepressor complex is composed of one Cyc8 and four Tup1 subunits. Mol. Cell. Biol. 16, 6707-6714. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Wahi, M., and Johnson, A.D. (1995). Identification of genes required for alpha 2 repression in Saccharomyces cerevisiae. Genetics 140, 79-90. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Wu, J., Suka, N., Carlson, M., and Grunstein, M. (2001). TUP1 utilizes histone H3/H2B-specific HDA1 deacetylase to repress gene activity in yeast. Mol. Cell 7, 117-126. [DOI] [PubMed] [Google Scholar]
  56. Zaman, Z., Ansari, A.Z., Koh, S.S., Young, R., and Ptashne, M. (2001). Interaction of a transcriptional repressor with the RNA polymerase II holoenzyme plays a crucial role in repression. Proc. Natl. Acad. Sci. USA 98, 2550-2554. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Zhang, Z., Varanasi, U., and Trumbly, R.J. (2002). Functional dissection of the global repressor Tup1 in yeast: dominant role of the C-terminal repression domain. Genetics 161, 957-969. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Zitomer, R.S., and Lowry, C.V. (1992). Regulation of gene expression by oxygen in Saccharomyces cerevisiae. Microbiol. Rev. 56, 1-11. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

[Supplemental Tables]
mbc_E04-05-0412_index.html (1.1KB, html)
mbc_E04-05-0412_supp_table_1.xls (1.4MB, xls)
mbc_E04-05-0412_supp_table_2.xls (1.9MB, xls)
mbc_E04-05-0412_supp_table_3.xls (62KB, xls)

Articles from Molecular Biology of the Cell are provided here courtesy of American Society for Cell Biology

RESOURCES