Skip to main content
NCBI home page
The EMBO Journal logoLink to The EMBO Journal
. 2007 Sep 27;26(20):4324–4334. doi: 10.1038/sj.emboj.7601859

Forkhead proteins control the outcome of transcription factor binding by antiactivation

Warren P Voth 1, Yaxin Yu 1, Shinya Takahata 1, Kelsi L Kretschmann 1, Jason D Lieb 2,3, Rebecca L Parker 1,*, Brett Milash 1, David J Stillman 1,a
PMCID: PMC2034669  PMID: 17898805

Abstract

Transcription factors with identical DNA-binding specificity often activate different genes in vivo. Yeast Ace2 and Swi5 are such activators, with targets we classify as Swi5-only, Ace2-only, or both. We define two unique regulatory modes. Ace2 and Swi5 both bind in vitro to Swi5-only genes such as HO, but only Swi5 binds and activates in vivo. In contrast, Ace2 and Swi5 both bind in vivo to Ace2-only genes, such as CTS1, but promoter-bound Swi5 fails to activate. We show that activation by Swi5 is prevented by the binding of the Forkhead factors Fkh1 and Fkh2, which recruit the Rpd3(Large) histone deacetylase complex to the CTS1 promoter. Global analysis shows that all Ace2-only genes are bound by both Ace2 and Swi5, and also by Fkh1/2. Genes normally activated by either Ace2 or Swi5 can be converted to Ace2-only genes by the insertion of Fkh-binding sites. Thus Fkh proteins, which function initially to activate SWI5 and ACE2, subsequently function as Swi5-specific antiactivators.

Keywords: cell cycle, ChIP–chip, DNA-binding, transcription factors

Introduction

Transcription factors with identical DNA-binding domains can activate different genes in vivo, and this constitutes an important unsolved puzzle in gene regulation. Proposed explanatory models include combinatorial control with other DNA-binding factors or subtle differences in binding affinity not detected in vitro. Here we present a different paradigm.

Ace2 and Swi5 are cell-cycle-regulated transcription factors in Saccharomyces cerevisiae that activate genes expressed in late M and early G1 phase (Simon et al, 2001). Ace2 and Swi5 have essentially identical DNA-binding domains and recognize the same sequences in vitro (Dohrmann et al, 1996), but they activate different genes in vivo (Dohrmann et al, 1992). For instance, the HO gene is only activated by Swi5, while CTS1 is only activated by Ace2. We show that Fkh1 and Fkh2 can act as selective antiactivators, blocking Swi5, but not Ace2, from activating transcription. Fkh1 and Fkh2 are G2/M phase-specific activators for a set of genes that includes Ace2 and Swi5. Fkh1 and Fkh2 also function as repressors at CTS1, but only against Swi5-dependent activation even though Swi5 and Ace2 bind equivalently to the promoter. This specific antiactivation requires recruitment of the Rpd3(Large) histone deacetylase (HDAC) complex. Our global analysis suggests that this mechanism operates in vivo at all yeast genes that are bound by these factors. Moreover, we show that Fkh antiactivation is transferable. Promoters that are naturally activated by either Swi5 or Ace2 can be converted to an Ace2-only activation program by insertion of Fkh-binding sites.

Results and discussion

Genetic identification of CTS1 NRE and Fkh regulatory factors

We identified a negative regulatory element (NRE) in the CTS1 promoter which, when deleted or mutated, allowed Swi5 to activate a CTS1-lacZ plasmid reporter (Dohrmann et al, 1996). This NRE was localized to 66 bps located between −404 and −470 bp upstream of the ATG of CTS1, with two Ace2-binding sites located nearby, between −520 and −545 bp (Figure 1A). The NRE represses activation independent of position, orientation, and promoter context (Dohrmann et al, 1996). We precisely deleted the minimal NRE region from the genomic CTS1 promoter, eliminating concerns about effects of a plasmid-based assay or the use of a lacZ reporter gene. This CTS1(nreΔ) allele was introduced into strains with mutations in ACE2 and SWI5, and expression of wild-type CTS1 and CTS1(nreΔ) were measured (Figure 1B). The native CTS1 gene is not expressed in the ace2 mutant (column 3), but strong expression of CTS1(nreΔ) is seen even in the absence of the normally required Ace2 activator (column 7). Furthermore, activation of CTS1(nreΔ) is Swi5-dependent, since expression is lost in the ace2 swi5 CTS1(nreΔ) strain (column 8). We also constructed a mutant CTS1(nre-m) promoter with sequence substitutions throughout the NRE, maintaining the spacing between the Swi5/Ace2-binding sites and the transcription start site, and this mutant promoter showed similar activation by Swi5 (Figure 1B, columns 11–12). Thus, the NRE element in the CTS1 promoter prevents bound Swi5 from activating CTS1 transcription. We note that the NRE deletion does not fully restore CTS1 expression in the absence of Ace2, suggesting that additional repressive mechanisms are still present in the CTS1(nreΔ) promoter.

Figure 1.

Figure 1

Open in a new tab

The NRE in the CTS1 promoter blocks activation by Swi5. (A) Map shows the Ace2/Swi5-binding sites at −546 and −526 from the ATG codon, the negative regulatory element (NRE) defined by deletion analysis as from −470 to −418 (Dohrmann et al, 1996), and the TATA element at −313. The transcription start site is at −238 (Dohrmann et al, 1992). The four Fkh-binding sites within the NRE are shown. The alignment shows the Fkh consensus site (Zhu et al, 2000), the sites from the CTS1 NRE and the ‘TG' mutation in NRE site 4 (Dohrmann et al, 1996). (B) Swi5 activates CTS1(nreΔ). RNAs were used for RT-real-time–PCR assays to measure CTS1 RNA levels. RNA values are normalized to the CTS1(WT) ACE2 SWI5 strain (defined as 100), and the error bars show the standard deviation of the triplicate PCR reactions. (C) Swi5 weakly activates CTS1-lacZ in an fkh1 fkh2 mutant by β-galactosidase activity from CTS1-lacZ. LacZ values are normalized to the ACE2 SWI5 FKH1 FKH2 strain (defined as 100), and the error bars show the standard deviation of triplicate cultures for lacZ assays. (D) Swi5 weakly activates CTS1 in an fkh1 fkh2 mutant by S1 nuclease protection assay. (E) Swi5 activates CTS1-lacZ in an fkh1 fkh2 mutant when Swi5 is expressed from the Fkh-independent MET17 promoter. LacZ values are normalized to the ACE2 SWI5 FKH1 FKH2 strain (defined as 100), and the error bars show the standard deviation of triplicate cultures for lacZ assays. (F) Fkh1 and Fkh2 repress CTS1 via the NRE element by S1 nuclease protection assay.

A genetic screen was carried out to identify mutations in genes that normally prevent Swi5 from activating CTS1 (Dohrmann et al, 1996). An ace2 strain with an integrated CTS1-lacZ reporter was mutagenized and suppressors were identified as blue colonies in the presence of the X-Gal chromogenic substrate. Genetic analysis and complementation cloning for one suppressor mutation, nce11, identified plasmids containing the FKH2 gene. Segregation analysis demonstrated that the nce11 mutations were allelic with an fkh2 disruption allele. FKH2 encodes a member of the winged helix superfamily of DNA-binding transcription factors. Fkh2 and its paralog Fkh1 are redundant activators that bind to the promoters of the CLB2 group of genes expressed in G2, including CLB2, SWI5, and ACE2 (Hollenhorst et al, 2000; Koranda et al, 2000; Kumar et al, 2000; Pic et al, 2000; Zhu et al, 2000). Strains with single mutations in FKH1 or FKH2 are predominantly normal in cell cycle progression, but fkh1 fkh2 double mutant strains exhibit strong defects consistent with reduced CLB2 expression. Although we only obtained an fkh2 mutant in our initial screen, we included FKH1 in subsequent analyses, based on its close homology and known functional overlap with FKH2.

Fkh1 and Fkh2 bind CTS1 and repress via the NRE element

Strains were constructed to assess the contributions of FKH1 and FKH2 in blocking Swi5 activation of CTS1. Deletion of either FKH1 or FKH2 weakly suppresses the ace2 defect in CTS1 expression, allowing Swi5 to activate CTS1 at 2–3 times the level observed in an ace2 mutant; similar effects were seen with both an integrated CTS1-lacZ reporter (Figure 1C) and the native CTS1 gene (Figure 1D). Suppression of the ace2 defect in CTS1 expression requires Swi5. Thus, mutations in either fkh1 or fkh2 have similar effects, allowing Swi5 to inappropriately activate expression of CTS1.

Transcription of the SWI5 gene is normally activated by Fkh1/2 (Hollenhorst et al, 2000; Koranda et al, 2000; Kumar et al, 2000; Pic et al, 2000; Zhu et al, 2000), so the nonadditive suppression of the ace2 transcriptional defect in the fkh1 fkh2 double mutant (Figure 1C and D) could be due to decreased SWI5 expression. To address this problem, a similar set of strains was constructed with SWI5 expressed from the MET17 promoter. MET17 expression under noninducing conditions is Fkh-independent and occurs at S/G2 (Spellman et al, 1998), when SWI5 is expressed. Swi5 produced from this METp∷SWI5 construct retains the post-translational signals for regulated nuclear localization and degradation within the cell cycle. Immunoblot quantitation shows levels of Swi5 expressed from the MET17 promoter are less than two-fold above native Swi5 (data not shown). Using this METp∷SWI5 allele, CTS1 expression in the absence of Ace2 increased four- to five-fold with either a fkh1 or fkh2 single mutation, and the ace2 fkh1 fkh2 mutant showed CTS1 expression levels about nine-fold over that of ace2 alone (Figure 1E). The additive increase in Swi5-dependent CTS1 expression in the fkh1 fkh2 double mutant indicates that the Fkh1 and Fkh2 factors are partially redundant for inhibiting CTS1 activation by Swi5, but both are required for full repression. In contrast, we do not see additivity in the fkh1 fkh2 double mutant when SWI5 is expressed from its native Fkh1/2-dependent promoter (Figure 1C and D). The additive effect in the fkh1 fkh2 double mutant when SWI5 is expressed from the Fkh-independent MET17 promoter (Figure 1E) indicates that the Fkh proteins are redundant. Additionally, combining the fkh1 and fkh2 mutations with the NRE deletion shows only a minor increase in suppression relative to the effect of the NRE deletion alone (Figure 1F). This comparative lack of additivity is consistent with Fkh1/2 acting through the NRE element at CTS1.

The CTS1 NRE region contains four matches to the consensus Fkh1-binding site (Zhu et al, 2000), with one site being a perfect match (Figure 1A). We previously identified a two-nucleotide substitution in the NRE region that allows Swi5 activation (Dohrmann et al, 1996), and these changes are within the perfect Fkh consensus site. Chromatin immunoprecipitation (ChIP) shows that Fkh1 and Fkh2 both bind to the CTS1 and CLB2 promoters, but no binding was seen to the CTS1(nreΔ) allele (Figure 2A and B) demonstrating that Fkh1 and Fkh2 bind specifically to the NRE region of the CTS1 promoter.

Figure 2.

Figure 2

Open in a new tab

Fkh1 and Fkh2 bind to the CTS1 NRE. (A) ChIP assay shows Fkh1-HA binding to CTS1 and CLB2. Deletion of the NRE eliminates Fkh1 binding. Error bars show the standard deviation of the ChIP PCR reactions performed in triplicate. (B) ChIP assay shows Fkh2-Myc binding to CTS1 and CLB2. Deletion of the NRE eliminates Fkh2 binding. Error bars show the standard deviation of the ChIP PCR reactions performed in triplicate. (C) ChIP assays from synchronized Ace2-Myc GAL-CDC20 cells show Ace2 binding to CTS1 but not HO. Flow cytometry profiles show the quality of the cell cycle synchrony. The vast majority of these haploid cells have an apparent 2C DNA content even at later time points, as cells release and progress into G1 without cell separation, which is only beginning at the 45 min time point. The strain has an swi5 mutation, and the modest defect in cell separation in an swi5 mutant (Dohrmann et al, 1992) results also in a small number of cells with apparent 4C DNA content. Progression from mitotic arrest through G1 and into S phase was also confirmed by synchronous bud formation at later time points (data not shown). (D) ChIP assays from synchronized Swi5-Myc GAL-CDC20 cells show Swi5 binding to HO and CTS1. (E) ChIP assay shows Fkh1-HA from log phase cells binding to the CLB2 promoter but not the CLB2 ORF.

Binding of the Rpd3(Large) complex to the NRE element

Recent work has suggested that Fkh2 recruits the Isw2 remodeling factor to promoters, and that Isw2 contributes to transcriptional repression (Sherriff et al, 2007). To investigate whether Isw2, or its paralog Isw1, were involved in blocking Swi5 activation at CTS1, we constructed ace2 isw2, ace2 isw1, and ace2 isw1 isw2 strains. The substantial reduction in CTS1 mRNA levels caused by an ace2 mutation is not suppressed by isw1 or isw2 mutations (data not shown), and we conclude that these chromatin regulators do not repress at CTS1. We next examined sin3 and rpd3 mutations, as Sin3 was identified as interacting with Fkh1 and Fkh2 in a global protein interaction screen (Ho et al, 2002), and Rpd3 is associated with Sin3 as components of HDAC complexes that repress transcription (Kurdistani and Grunstein, 2003). CTS1 mRNA levels are higher in ace2 sin3 and ace2 rpd3 strains than in an ace2 single mutant, indicating that Sin3/Rpd3 is involved in CTS1 repression (Figure 3A and B).

Figure 3.

Figure 3

Open in a new tab

Rpd3(Large) regulates CTS1 via the NRE. (A) RT–PCR assays show that sin3 and sds3 mutations allow CTS1 expression despite an ace2 mutation, while rco1 does not suppress. RNA values are normalized to the ace2 mutant (defined as 1); CTS1 RNA in the ace2 mutant is 2% of the ACE2+ strain. (B) RT–PCR assays show that CTS1 can be activated by Swi5 in an sin3 mutant, and that the effects of an sin3 mutation and deletion of the NRE promoter element are not additive. RNA values are normalized to the CTS1(WT) ACE2 SWI5 SIN3 strain (defined as 100), and the error bars show the standard deviation of the triplicate PCR reactions. (C) RT–PCR assays show that CTS1 can be activated by Swi5 in an sds3 mutant, and that the effects of an sds3 mutation and deletion of the NRE promoter element are not additive. RNA values are normalized to the CTS1(WT) ACE2 SWI5 SIN3 strain (defined as 100), and the error bars show the standard deviation of the triplicate PCR reactions. (D) The Sin3-HA, Rpd3-Myc, and Sds3-Myc components of Rpd3(Large) bind to CTS1 in a ChIP assay, while the Rco1-Myc component of Rpd3(Small) does not. Error bars show the standard deviation of the ChIP PCR reactions performed in triplicate. (E) ChIP assay shows Sin3-HA binding to CTS1, CLB2, and INO1. Deletion of the NRE reduces Sin3 binding. Error bars show the standard deviation of the ChIP PCR reactions performed in triplicate. (F) ChIP assay shows Sds3-HA binding to CTS1, CLB2, and INO1. Deletion of the NRE reduces Sds3 binding. Error bars show the standard deviation of the ChIP PCR reactions performed in triplicate. (G) ChIP assay shows Sds3-HA binding to EGT2 and CTS1. Insertion of the NRE element from CTS1 into the EGT2(+NRE) promoter increases Sds3 binding. Error bars show the standard deviation of the ChIP PCR reactions performed in triplicate.

At least two HDAC complexes contain Rpd3 (Carrozza et al, 2005; Keogh et al, 2005). The Rpd3(Large) complex is present primarily at promoters while Rpd3(Small) acts primarily at transcribed open reading frames (ORFs). While both complexes contain Sin3 and Rpd3, there are unique subunits such as Sds3 present only in Rpd3(Large), and Rco1 present only in Rpd3(Small). An sds3 mutation allows CTS1 expression in the absence of Ace2, while an rco1 mutation does not (Figure 3A and C), and thus Rpd3(Large) represses CTS1. The CTS1(nreΔ) allele also allows CTS1 activation by Swi5, but there is no additive effect from combining this promoter mutation with either sin3 or sds3 in the ace2 SWI5 strains. This lack of additivity suggests that Rpd3(Large) represses CTS1 expression via the NRE element. As seen for the NRE deletion, neither sin3 nor sds3 mutations fully restore CTS1 expression in the absence of Ace2, suggesting additional mechanisms of repression. Finally, while sin3 and sds3 mutations allow CTS1 expression in the absence of Ace2, CTS1 expression is largely lost in the ace2 swi5 strains, demonstrating that Swi5 is responsible for CTS1 transcription in the absence of Ace2.

ChIP experiments were performed to examine binding of Sin3/Rpd3 complexes to the CTS1 promoter. Sin3, Rpd3, and Sds3 all bind to CTS1, indicating binding by Rpd3(Large); in contrast, Rco1, the Rpd3(Small)-specific subunit, does not bind (Figure 3D). We next compared binding of Rpd3(Large) to the wild-type CTS1 promoter and the CTS1(nreΔ) promoter deletion. Deleting the NRE element results in a substantial reduction of binding of Sin3 (Figure 3E) and Sds3 (Figure 3F), demonstrating that the NRE is required for recruiting Rpd3 (Large) to CTS1. As a control, we measured binding to the INO1 promoter, known to bind Sin3/Rpd3 (Kadosh and Struhl, 1997). Binding to INO1 was unaffected in the CTS1(nreΔ) strain, demonstrating specificity. Interestingly, Rpd3(Large) binds to the CLB2 promoter as well, which is also bound by Fkh1 and Fkh2. An ace2 mutation does not affect Rpd3(Large) binding to CTS1 (data not shown). We next examined histone acetylation directly at CTS1 in wild-type and rpd3 strains. The rpd3 mutation results in decreased histone H3 at CTS1, and increased levels of acetylated H3 and acetylated H4 (Figure 4A). Normalizing the acetylated histone ChIP signals to the levels of H3 at the promoter shows that the rpd3 mutation results in a marked increase in histone acetylation at the CTS1 promoter (Figure 4B). We also determined whether Rpd3(Large) binding to promoters was dependent upon Fkh proteins, measuring Sin3-HA binding in wild-type, fkh1 and fkh2 single mutants, and in the fkh1 fkh2 double mutant (Figure 4C). The Sin3 ChIP signal at the CLB2 and CTS1 promoters is abolished in the fkh1 fkh2 double mutant, but largely unaffected in the fkh1 or fkh2 single mutants. We conclude that either Fkh1 or Fkh2 is competent to recruit Rpd3(Large) to the CLB2 and CTS1 promoters. In contrast, Rpd3(L) binding to the INO1 promoter is unaffected in the fkh1 fkh2 double mutant. In summary, these experiments demonstrate that the Rpd3(Large) HDAC complex is recruited to the CTS1 promoter by Fkh1 and Fkh2, Rpd3(Large) and functions to specifically reduce CTS1 activation by Swi5.

Figure 4.

Figure 4

Open in a new tab

rpd3 affects histone acetylation and fkh mutation affect Rpd3(L) binding to CTS1. (A) Lysates from wild-type or rpd3 mutant cells were immunoprecipitated with either no antibody, anti-H3, anti-H3(K14-Ac), or anti-H4(-Ac). PCR reactions were performed with CTS1 and control primers, and the ratio of these ChIP values were normalized to ratios for PCR reactions with the two primer sets using input DNA. Error bars show the standard deviation of the ChIP PCR reactions performed in triplicate. (B) Using the data in part (A), the ChIP ratios for H3(K14-Ac) and H4(-Ac) were divided by the ChIP ratios for histone H3. (C) ChIP assays of Sin3-HA binding in wild-type, fkh1, fkh2, and fkh1 fkh2 strains show that either Fkh1 or Fkh2 can recruit Sin3-HA to CLB2 and CTS1, while fkh mutations do not affect Sin3-HA binding to the INO1 promoter.

Different in vivo DNA binding by Ace2 and Swi5

Although in vitro binding experiments show that Ace2 and Swi5 recognize the same sequences in promoters, they can activate transcription of different genes in vivo (Dohrmann et al, 1996). Ace2, but not Swi5, activates CTS1 expression, and Swi5, but not Ace2, activates HO. Ace2 and Swi5 are cell-cycle-regulated transcription factors, and cells with a GAL∷CDC20 allele were synchronized within the cell cycle by removing galactose to arrest in mitosis, followed by re-addition of galactose (Bhoite et al, 2001). Flow cytometry and analysis of cycle-regulated mRNAs show a high degree of synchrony (Figure 2C and data not shown). ChIP experiments were performed with synchronized cells showing Swi5-Myc binding to the HO and CTS1 promoters (Figure 2D). Swi5-Myc binds to both HO and CTS1, although it does not activate CTS1. Thus, the NRE in CTS1 does not prevent Swi5 from binding, but acts to prevent promoter-bound Swi5 from activating transcription.

ChIP with Ace2-Myc shows binding to CTS1 but not to HO (Figure 2C); Ace2 does not activate HO although it binds in vitro (Dohrmann et al, 1996). Swi5 does not block Ace2 from binding to HO, as no binding of Ace2 to HO is seen in either an SWI5 or an swi5 strain. Thus, there are distinct mechanisms of promoter restriction for Swi5 and Ace2 at HO and CTS1. Swi5 binds to CTS1, but does not activate transcription. In contrast, Ace2 is unable to activate HO transcription simply because it does not bind to the HO promoter. During the cell cycle, Swi5 binds DNA before Ace2 (Figure 2C and D), and Swi5 enters the nucleus before Ace2 (Sbia et al, 2007). Work is in progress to understand why Ace2 is unable to bind to and activate expression of the HO gene, although Ace2 binds well to HO in vitro.

Genome-wide analysis of regulation and binding by Swi5 and Ace2

To determine whether these distinct promoter restriction mechanisms are general for all Ace2 and/or Swi5 targets throughout the genome, global expression, chromatin binding, and binding motif correlations were undertaken. First, genome-wide experiments were performed to determine whether these Ace2- and Swi5-binding patterns are seen at other promoters. Expression microarrays were used to define their targets of regulation and revealed that the RNA levels for 66 genes was reduced more than two-fold in either the ace2 single mutant or the swi5 single mutant or the ace2 swi5 double mutant (Supplementary data). To more stringently identify genes directly regulated by Ace2 and Swi5, we eliminated genes that are not cell cycle regulated (Cho et al, 1998; Spellman et al, 1998; Pramila et al, 2006). Among the 23 remaining genes, eight were activated only by Ace2, six by Swi5 only, and nine required either Ace2 or Swi5 for activation (Figure 5A). Expression of CDC6 and PCL2 was reduced less than two-fold, but these genes were included in our analysis as their expression was previously shown to be activated by Ace2 or Swi5 (Piatti et al, 1995; Aerne et al, 1998; McBride et al, 1999; Doolin et al, 2001). RT–PCR experiments confirm the dependence of these genes on Ace2 and/or Swi5 (Supplementary Figure S1).

Figure 5.

Figure 5

Open in a new tab

Expression and promoter occupancy of Ace2- and Swi5-dependent genes. (A) Relative RNA levels in ace2, swi5, and ace2 swi5 strains as a log2 ratio (mutant/WT). RNA levels with greater than a two-fold reduction are shown in red. (B) The presence of binding sites for Ace2/Swi5 or Fkh1/2 in promoters, from MacIsaac et al (2006). (C) The degree of Fkh1 or Fkh2 binding by ChIP–chip are the binding ratios from the data supporting Simon et al (2001), with binding levels greater than two standard deviations from the mean shown in red. (D) Ace2 and Swi5 binding as a heat map showing the results for an untagged control strain, and for the Ace2-Myc and Swi5-Myc strains at the CDC20 arrest (t=0) and time points after release. The heatmap represents the results of 33 microarray experiments (Materials and methods) with the number of repeats for each time point as shown in Supplementary Table 3. Percentile ranks were calculated for each experiment and then averaged for each time point to generate the values shown. A color scale represents the degree of binding expressed as percentile ranks of ChIP enrichment.

To determine Ace2 and Swi5 genomic-binding locations, Ace2-Myc and Swi5-Myc strains were synchronized at M with GAL∷CDC20, and samples taken at timed intervals after the release were processed for ChIP and hybridized to whole-genome DNA microarrays. The complete experiment was performed three times (Supplementary data). Some promoters were bound by both Swi5 and Ace2, and some by Swi5 only; but none of the selected promoters were bound by Ace2 only (Figure 5D). The analysis revealed that Ace2 and Swi5 both bind to all genes activated only by Ace2, but only Swi5 binds to genes activated only by Swi5. This is the same pattern observed for CTS1 and HO (Figure 2C and D). Cells synchronized with GAL∷CDC20 and ChIPs quantitated with real-time PCR confirm the ChIP–chip results for selected representative genes (Figure 6A).

Figure 6.

Figure 6

Open in a new tab

Binding of Swi5 and Ace2 to target genes. (A) Ace2-Myc and Swi5-Myc binding in synchronized GAL-CDC20 cells were measured by real-time PCR after ChIP. ChIP-binding ratios are expressed as a ratio to an internal control. (B) Diagrams show the positions of Ace2- or Swi5-binding sites (yellow arrows) or Fkh-binding sites (blue arrows) on genes characterized as (1) activated by Ace2 only, (2) activated by either Ace2 or Swi5, or (3) activated either by Swi5 only.

Upstream Fkh sites define genes that are activated only by Ace2

At CTS1, Fkh binding is required to prevent Swi5 from activating transcription. To assess whether Fkh-binding sites are present at other genes activated by Ace2 but not by Swi5 (Ace2-only genes), we analyzed data from the expression microarray and ChIP–chip experiments (Figure 5A and D) in relation to the genome-wide distribution of DNA sequence motifs bound by Ace2/Swi5 and/or Fkh1/2 (MacIsaac et al, 2006) (Figure 5B). Ace2-only genes contain both Ace2/Swi5 and Fkh sites (red), while Swi5-only genes contain Ace2/Swi5 sites but lack Fkh sites (yellow) (Figure 7A). Examining genes whose expression is reduced in an ace2 mutant, we find that both Ace2 (Figure 7B) and Swi5 (Figure 7C) bind, and all of these genes contain Fkh-binding sites. In contrast, Swi5 (Figure 7D) but not Ace2 (Figure 7E) binds to Swi5-only genes, and none of these contain Fkh sites. The Ace2-only genes all contain multiple Fkh sites, but there is no common pattern of binding site spacing or orientation in the three classes of promoters (Figure 6B). ChIP–chip experiments (Simon et al, 2001) show Fkh1 and Fkh2 bind to the Ace2-only genes, but not to either class of gene activated by Swi5 (Figure 5C).

Figure 7.

Figure 7

Open in a new tab

Genome-wide analysis of expression, binding sites, and factor binding. Promoters with Ace2/Swi5-binding sites and Fkh-binding sites are red, those with Ace2/Swi5 but no Fkh sites are yellow, those with only Fkh sites are blue, and those with neither site are gray. (A) log2(ace2/WT) plotted versus log2(swi5/WT) microarray data, showing Ace2-only genes do contain Fkh sites and Swi5-only genes do not. (B) log2(ace2/WT) microarray data plotted versus Ace2 ChIP–chip occupancy data, showing Ace2 binding to Ace2-only genes. (C) log2(ace2/WT) microarray data plotted versus Swi5 ChIP–chip occupancy data, showing Swi5 binding to Ace2-only genes. (D) log2(swi5/WT) microarray data plotted versus Swi5 ChIP–chip occupancy data, showing Swi5 binding to Swi5-only genes. (E) log2(swi5/WT) microarray data plotted versus Ace2 ChIP–chip occupancy data, showing Ace2 does not bind to Swi5-only genes.

Introduction of Fkh sites into promoters reduces activation by Swi5 but not Ace2

To test whether Fkh sites are sufficient to reduce activation by Swi5, we inserted the CTS1 NRE, containing Fkh-binding sites into the native promoters of the EGT2 and SIC1 genes. EGT2 and SIC1 are activated by both Ace2 and Swi5 (Knapp et al, 1996; Kovacech et al, 1996; Toyn et al, 1997), and are not normally bound by Fkh1 or Fkh2 (Simon et al, 2001). Expression of EGT2 and SIC1 is modestly reduced in ace2 or swi5 single mutants, but more substantially reduced in the ace2 swi5 double mutant (Figure 8). In the ace2 swi5 strain, EGT2 levels are reduced to undetectable levels, while SIC1 mRNA is reduced to 30% of wild type. Presumably, there are other basal activators of SIC1 expression in addition to Ace2 and Swi5. In contrast, CTS1 mRNA is absent in the ace2 SWI5 strain.

Figure 8.

Figure 8

Open in a new tab

The NRE element reduces activation by Swi5. RT–PCR assays show CTS1 is only activated by Ace2, while EGT2 and SIC1 is activated by either Ace2 or Swi5. Insertion of the CTS1 NRE into EGT2 and SIC1 (EGT2(+NRE) and SIC1(+NRE)) reduces SWI5-dependent activation, while insertion of a CTS1 NRE fragment with sequence substitutions through the four four Fkh sites (EGT2(+NRE-m) and SIC1(+NRE-m)) has no effect on SWI5-dependent activation. The RNA values for each set of strains are normalized to the ACE2 SWI5 strain (defined as 100), and the error bars show the standard deviation of the triplicate PCR reactions.

We made two types of insertions into the EGT2 and SIC1 promoters. Both were made at the endogenous chromosomal locations with no loss of native sequences. The EGT2(+NRE) and SIC1(+NRE) promoters have a region from the CTS1 promoter with the entire NRE element inserted, while the EGT2(+NRE-m) and SIC1(+NRE-m) promoters contain a similar sized fragment from CTS1 except for sequence substitutions throughout the four Fkh sites (see Figure 1). Expression of EGT2(+NRE) and SIC1(+NRE) is significantly reduced in the ace2 strain, where Swi5 is the remaining major activator (Figure 8). Expression of SIC1(+NRE) is reduced to nearly the level seen in the ace2 swi5 strain, providing additional evidence that the NRE blocks activation by Swi5 (Figure 8).

Data from EGT2(+NRE) suggests that the NRE is only partially able to block Swi5 activation at EGT2. This may be due to the number and dispersed locations of Swi5/Ace2-binding motifs throughout the promoter (see Figure 6B). For both promoters, however, the NRE reproducibly confers at least a two-fold reduction only in Swi5-dependent activation, with no effect on Ace2-dependent activation (Figure 8). This results in an approximately three-fold bias for Ace2 relative to Swi5 for these promoters, which are normally equivalently sensitive to either activator. The EGT2(+NRE-m) and SIC1(+NRE-m) promoters with a CTS1 promoter fragment lacking the Fkh sites are expressed at levels similar to that of wild-type promoters, indicating that the observed effects are not due to alterations in spacing or architecture within these promoters. These experiments show that Fkh sites are sufficient to prevent Swi5 from activating specific target genes.

We also examined Sds3 binding to the EGT2(+NRE) promoter. As shown in Figure 3G, there is increased Sds3 binding to EGT2(+NRE) compared to wild-type EGT2. This demonstrates that the NRE element with Fkh-binding sites is sufficient to recruit the Rpd3(Large) complex to promoters.

DNA binding by Fkh1 and Fkh2 varies during the cell cycle but this is unlikely to be required for the antiactivation specificity

Fkh1 and Fkh2 binding at CLB2 group promoters activates transcription in G2, yet binding at Ace2-activated targets results in selective repression in G2/M. Additionally, binding of Swi5 and Ace2 to their G1 targets occurs at different times, with Swi5 preceding Ace2. Thus, we hypothesized that differences in Fkh1 and Fkh2 occupancy at CTS1 and CLB2 with respect to the cell cycle might determine their roles as either activators versus selective antiactivators. We therefore directly analyzed the Fkh1 and Fkh2 proteins during the cell cycle by ChIP and by western immunoblotting (Supplementary Figure S2). During the cell cycle, ChIP experiments show significant variation in binding to CTS1 and CLB2 over time, with maxima during the M and early G1 period. Additionally, western blots show subtle differences in abundance and mobility of Fkh proteins during the cell cycle. However, the cell cycle times of these changes do not explain why Fkh1 and Fkh2 differently repress and activate CTS1 and CLB2. Thus, it appears unlikely that the mechanism of Swi5-specific antiactivation involves restricting Fkh occupancy of CTS1 to a limited period of the cell cycle.

Activation of CLB2 involves Fkh2 recruiting the Ndd1 activator at certain cell cycle times (Koranda et al, 2000). At other times of the cell cycle, Fkh1 and Fkh2, in the absence of Ndd1, actually repress CLB2 transcription (Hollenhorst et al, 2000), an idea supported by the observation that fkh1 mutations suppress the lethality caused by deletion of the NDD1 gene (Koranda et al, 2000). CLB2 activation during G2 requires the presence of Mcm1 at the promoter (Althoefer et al, 1995a). At CTS1, in the absence of Mcm1 interaction, but where there are multiple Fkh-binding sites, Fkh1 and Fkh2 binding may continuously reinforce each other, resulting in a sharp peak of Fkh2 binding to CTS1 at G2/M and a subsequent Fkh1 peak in M/G1, resulting in sufficient occupancy by both factors for full repression. Thus, promoter context and neighboring DNA-binding proteins may determine the activity of Fkh1 and Fkh2.

Fkh proteins do not bind to the CLB2 ORF

Because both Ace2 and Swi5 are equivalently able to program Pol II transcription in Fkh-independent contexts, we hypothesized that the mechanism of discrimination may involve FKH-dependent differences in the ability of Ace2 and Swi5 to program a step subsequent to Pol II recruitment. Morillon et al (2003) reported that Fkh1, in addition to binding to the CLB2 promoter, also binds to the CLB2-coding region. The authors postulated that Fkh proteins influence transcriptional elongation and termination, and regulate phosphorylation of the Pol II CTD. CLB2 is cell cycle regulated, expressed soon after the release during G2. One prediction based on Morillon et al (2003) is that the Fkh proteins would be bound to the promoter before expression, and then appear to travel down the ORF after transcription begins. A variation would be for Fkh1 to be bound to the CLB2 ORF throughout the cell cycle, irrespective of whether the gene is transcribed or not.

To test these models, we examined the kinetics of Fkh1 binding to this ORF, using chromatin extracts from synchronized cells expressing Fkh1-HA. However, our ChIP assays provided no evidence for Fkh1 binding to the CLB2 ORF at a level above background during the cell cycle (Supplementary Figure S3A). We also looked for Fkh1 binding to the CLB2 ORF in log phase cells, as nonsynchronized cells had been used by Morillon et al (2003). Once again, we did not see any Fkh1 binding to the CLB2 ORF, although we did see strong binding to the CLB2 promoter (Figure 2E) in both sets of samples, as well as strongly periodic binding to the CTS1 promoter.

Based on the suggestion that Fkh proteins regulate pol II CTD phosphorylation (Morillon et al, 2003), we considered a model where Swi5 binding leads to pol II recruitment to CTS1, but the Fkh proteins prevent pol II from initiating transcription. However, ChIP shows equivalent pol II binding to CTS1 in WT versus swi5 strains, and pol II binding is reduced to equivalent levels in ace2 and ace2 swi5 strains (Supplementary Figure S3B). This result rules out a model where Swi5 recruits pol II to CTS1, but the Fkh proteins block the pol II from initiating transcription. Thus, the selective antiactivation mechanism impinges on the transcriptional activation program at an earlier step.

Conclusions

We have analyzed divergent regulation by paralogous transcription factors with identical DNA-binding domains. Ace2 is prevented from binding to genes activated solely by Swi5. In contrast, Swi5 binds to, but cannot activate, genes like CTS1 due to the presence of Fkh1 and Fkh2, which function as selective antiactivators. Antiactivation has been observed in eukaryotes before at the yeast GAL genes (Traven et al, 2006). Here, the Gal80 protein blocks the activation potential of the Gal4 activator, and antiactivation requires physical interaction between the two proteins. In the presence of galactose, the Gal3 protein overcomes the antiactivation. In contrast, the antiactivation effected by Fkh1/2 onto Swi5 involves two separate DNA-binding proteins that bind to separate elements on the same promoter and the recruitment of a HDAC. Further work is needed to understand how the Fkh proteins act as selective antiactivators, blocking activation by Swi5 but not by Ace2, and the unique attributes possessed by Ace2 that allow it to overcome this antiactivation.

The Fkh1 and Fkh2 proteins have been previously characterized as activators of the G2/M CLB2 group of genes, although they may have negative roles at these genes in other cell cycle phases (Hollenhorst et al, 2000; Kumar et al, 2000; Zhu et al, 2000). The CLB2 group promoters all contain a single Mcm1-binding site in close apposition to a single Fkh consensus site (Althoefer et al, 1995b; Pic et al, 2000; Boros et al, 2003). Here, we show that the Ace2-activated/Fkh-anti-activated group of M/G1 promoters contains multiple Fkh consensus sites, with no discernible Mcm1-binding sequences; additionally, no Mcm1 binding has been detected at these promoters (data not shown, Simon et al, 2001). Moreover, Mcm1-Fkh recruit the cell-cycle-regulated transcriptional activator Ndd1 to CLB2-group promoters (Koranda et al, 2000), but no Ndd1 recruitment to Ace2-regulated promoters has been detected (Simon et al, 2001). These differences may explain in part the contrasting earlier function of Fkh proteins at G2/M versus the Ace2 and Swi5-bound genes expressed in late M and G1 phases. We propose that Ace2-activated M/G1 target genes are selectively regulated through a novel antiactivation mechanism, in part comprising HDAC recruitment by the Fkh1/Fkh2 paralogs, determining which target genes are activated by Ace2 as opposed to Swi5 during the ordered cascade of the cell cycle transcriptional program (Simon et al, 2001; Pramila et al, 2006). It is possible that the acetylation of Swi5 is important for its function as an activator, and the HDACs block this activation; further work will be needed to test this hypothesis.

Materials and methods

All yeast strains used are listed in Supplementary Table 1 and are isogenic in the W303 background (Thomas and Rothstein, 1989). Standard genetic methods were used for strain construction (Rothstein, 1991; Sherman, 1991), and details on construction of specific alleles are presented in the Supplementary data. Cells were grown at 30°C in YEPD medium (Sherman, 1991). Cell cycle synchronization was performed either by α-factor arrest and release as described (Mitra et al, 2006), or by galactose withdrawal and re-addition with a GAL∷CDC20 strain (Bhoite et al, 2001). A high degree of synchrony was demonstrated by flow cytometry analysis, budding indices, and analysis of cycle-regulated mRNAs (data not shown).

RNA levels were determined with S1 nuclease protection assays as described (Voth et al, 2005) or by RT–PCR. For RT–PCR, DNA-free total RNA was purified from mid-log cultures by hot-acid phenol extraction followed by precipitation through CsCl by ultracentrifugation (Ausubel et al, 1987). Gene expression levels were quantitated by randomly primed cDNA synthesis with M-MLV Reverse Transcriptase (Invitrogen), followed by real-time quantitative PCR. For microarrays, RNA from ace2, swi5, and ace2 swi5 strains were reverse transcribed and each was competitively hybridized with RNA from wild-type strains on Agilent oligonucleotide arrays. Swi5 levels in immunoblots were quantitated using an Odyssey system (Li-Cor).

All ChIPs (except for ChIP–chips) were performed as described (Bhoite et al, 2001) using 9E11 (Abcam) or 4A6 (Upstate) monoclonal antibody to the Myc epitope, and 12CA5 antibody to the HA epitope (University of Utah Bioprocessing Resource), anti-histone H3 (07–690, Upstate), anti-histone H3(Ac-Lys14) (07–353, Upstate), anti-acetyl-histone H4 (06–598, Upstate), and antibody coated magnetic beads (Rabbit and Pan Mouse IgG beads, Dynal Biotech). ChIP assays were analyzed as described, either using multiplex PCR (Bhoite et al, 2001) or by real-time PCR (Eriksson et al, 2004). Oligonucleotides used for ChIP, RT–PCR, or S1 nuclease protection are listed in Supplementary Table 2. Immunoblots of proteins transferred from low-bis-polyacrylamide gel electrophoresis (Whalen and Steward, 1993) were probed with anti-HA, anti-Myc, and anti-PGK 22C5-D8 (Molecular Probes) antibodies, and visualized with an Odyssey Infrared system (LI-COR Biosciences Lincoln, NE). Quantitation of ChIP was calculated as described (Eriksson et al, 2004), except that ChIP signals were additionally normalized to the total DNA amount from input chromatin for each sample. Standard deviations for normalized PCR replicates were calculated using Equation (7) of van Kempen and van Vliet (2000).

For the ChIP–chip experiments, eight separate synchrony and immunoprecipitation experiments were performed (four for Ace2 and four for Swi5). A total of 33 ChIP–chip hybridizations were performed, and are listed in Supplementary Table 3. The ChIP–chip experiments were performed as essentially as described (Lieb et al, 2001), with complete details provided in the Supplementary data.

Supplementary Material

Supplementary data

7601859s1.pdf (567.1KB, pdf)

Acknowledgments

We thank Bruce Futcher for providing information about Fkh genes in advance of publication, Mike Holland for advice on RT–PCR, Gustav Ammerer and Rick Young for providing strains, Deirdre Tatomer for help with microarray data extraction, and Tim Formosa, David Grunwald, and Dean Tantin for comments on the manuscript. RP was supported by NIH training grant T32 GM08537. This work was supported by a grant from the National Institutes of Health awarded to DJS.

References

  1. Aerne BL, Johnson AL, Toyn JH, Johnston LH (1998) Swi5 controls a novel wave of cyclin synthesis in late mitosis. Mol Biol Cell 9: 945–956 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Althoefer H, Schleiffer A, Wassmann K, Nordheim A, Ammerer G (1995a) Mcm1 is required to coordinate G2-specific transcription in Saccharomyces cerevisiae. Mol Cell Biol 15: 5917–5928 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Althoefer H, Schleiffer A, Wassmann K, Nordheim A, Ammerer G (1995b) Mcm1 is required to coordinate G2-specific transcription in Saccharomyces cerevisiae. Mol Cell Biol 15: 5917–5928 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Ausubel FM, Brent R, Kingston RE, Moore DE, Seidman JG, Smith JA, Struhl K (1987) Current Protocols in Molecular Biology. New York: Wiley and Sons [Google Scholar]
  5. Bhoite LT, Yu Y, Stillman DJ (2001) The Swi5 activator recruits the Mediator complex to the HO promoter without RNA polymerase II. Genes Dev 15: 2457–2469 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Boros J, Lim FL, Darieva Z, Pic-Taylor A, Harman R, Morgan BA, Sharrocks AD (2003) Molecular determinants of the cell-cycle regulated Mcm1p-Fkh2p transcription factor complex. Nucleic Acids Res 31: 2279–2288 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Carrozza MJ, Li B, Florens L, Suganuma T, Swanson SK, Lee KK, Shia WJ, Anderson S, Yates J, Washburn MP, Workman JL (2005) Histone H3 methylation by Set2 directs deacetylation of coding regions by Rpd3S to suppress spurious intragenic transcription. Cell 123: 581–592 [DOI] [PubMed] [Google Scholar]
  8. Cho RJ, Campbell MJ, Winzeler EA, Steinmetz L, Conway A, Wodicka L, Wolfsberg TG, Gabrielian AE, Landsman D, Lockhart DJ, Davis RW (1998) A genome-wide transcriptional analysis of the mitotic cell cycle. Mol Cell 2: 65–73 [DOI] [PubMed] [Google Scholar]
  9. Dohrmann PR, Butler G, Tamai K, Dorland S, Greene JR, Thiele DJ, Stillman DJ (1992) Parallel pathways of gene regulation: the homologous regulators SWI5 and ACE2 differentially control transcription of HO and chitinase. Genes Dev 6: 93–104 [DOI] [PubMed] [Google Scholar]
  10. Dohrmann PR, Voth WP, Stillman DJ (1996) Role of negative regulation in promoter specificity of the homologous transcriptional activators Ace2p and Swi5p. Mol Cell Biol 16: 1746–1758 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Doolin MT, Johnson AL, Johnston LH, Butler G (2001) Overlapping and distinct roles of the duplicated yeast transcription factors Ace2p and Swi5p. Mol Microbiol 40: 422–432 [DOI] [PubMed] [Google Scholar]
  12. Eriksson P, Biswas D, Yu Y, Stewart JM, Stillman DJ (2004) TATA-binding protein mutants that are lethal in the absence of the Nhp6 high-mobility-group protein. Mol Cell Biol 24: 6419–6429 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Ho Y, Gruhler A, Heilbut A, Bader GD, Moore L, Adams SL, Millar A, Taylor P, Bennett K, Boutilier K, Yang L, Wolting C, Donaldson I, Schandorff S, Shewnarane J, Vo M, Taggart J, Goudreault M, Muskat B, Alfarano C et al. (2002) Systematic identification of protein complexes in Saccharomyces cerevisiae by mass spectrometry. Nature 415: 180–183 [DOI] [PubMed] [Google Scholar]
  14. Hollenhorst PC, Bose ME, Mielke MR, Muller U, Fox CA (2000) Forkhead genes in transcriptional silencing, cell morphology and the cell cycle. Overlapping and distinct functions for FKH1 and FKH2 in Saccharomyces cerevisiae. Genetics 154: 1533–1548 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Hollenhorst PC, Pietz G, Fox CA (2001) Mechanisms controlling differential promoter-occupancy by the yeast forkhead proteins Fkh1p and Fkh2p: implications for regulating the cell cycle and differentiation. Genes Dev 15: 2445–2456 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Kadosh D, Struhl K (1997) Repression by Ume6 involves recruitment of a complex containing Sin3 corepressor and Rpd3 histone deacetylase to target promoters. Cell 89: 365–371 [DOI] [PubMed] [Google Scholar]
  17. Keogh MC, Kurdistani SK, Morris SA, Ahn SH, Podolny V, Collins SR, Schuldiner M, Chin K, Punna T, Thompson NJ, Boone C, Emili A, Weissman JS, Hughes TR, Strahl BD, Grunstein M, Greenblatt JF, Buratowski S, Krogan NJ (2005) Cotranscriptional set2 methylation of histone H3 lysine 36 recruits a repressive Rpd3 complex. Cell 123: 593–605 [DOI] [PubMed] [Google Scholar]
  18. Knapp D, Bhoite L, Stillman DJ, Nasmyth K (1996) The transcription factor Swi5 regulates expression of the cyclin kinase inhibitor p40-SIC1. Mol Cell Biol 16: 5701–5707 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Koranda M, Schleiffer A, Endler L, Ammerer G (2000) Forkhead-like transcription factors recruit Ndd1 to the chromatin of G2/M-specific promoters. Nature 406: 94–98 [DOI] [PubMed] [Google Scholar]
  20. Kovacech B, Nasmyth K, Schuster T (1996) EGT2 gene transcription is induced predominantly by Swi5 in early G1. Mol Cell Biol 16: 3264–3274 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Kumar R, Reynolds DM, Shevchenko A, Goldstone SD, Dalton S (2000) Forkhead transcription factors, Fkh1p and Fkh2p, collaborate with Mcm1p to control transcription required for M-phase. Curr Biol 10: 896–906 [DOI] [PubMed] [Google Scholar]
  22. Kurdistani SK, Grunstein M (2003) Histone acetylation and deacetylation in yeast. Nat Rev Mol Cell Biol 4: 276–284 [DOI] [PubMed] [Google Scholar]
  23. Lieb JD, Liu X, Botstein D, Brown PO (2001) Promoter-specific binding of Rap1 revealed by genome-wide maps of protein–DNA association. Nat Genet 28: 327–334 [DOI] [PubMed] [Google Scholar]
  24. MacIsaac KD, Wang T, Gordon DB, Gifford DK, Stormo GD, Fraenkel E (2006) An improved map of conserved regulatory sites for Saccharomyces cerevisiae. BMC Bioinform 7: 113. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. McBride HJ, Yu Y, Stillman DJ (1999) Distinct regions of the Swi5 and Ace2 transcription factors are required for specific gene activation. J Biol Chem 274: 21029–21036 [DOI] [PubMed] [Google Scholar]
  26. Mitra D, Parnell EJ, Landon JW, Yu Y, Stillman DJ (2006) SWI/SNF binding to the HO promoter requires histone acetylation and stimulates TATA-binding protein recruitment. Mol Cell Biol 26: 4095–4110 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Morillon A, O'Sullivan J, Azad A, Proudfoot N, Mellor J (2003) Regulation of elongating RNA polymerase II by forkhead transcription factors in yeast. Science 300: 492–495 [DOI] [PubMed] [Google Scholar]
  28. Piatti S, Lengauer C, Nasmyth K (1995) Cdc6 is an unstable protein whose de novo synthesis in G1 is important for the onset of S phase and for preventing a ‘reductional' anaphase in the budding yeast Saccharomyces cerevisiae. EMBO J 14: 3788–3799 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Pic A, Lim FL, Ross SJ, Veal EA, Johnson AL, Sultan MR, West AG, Johnston LH, Sharrocks AD, Morgan BA (2000) The forkhead protein Fkh2 is a component of the yeast cell cycle transcription factor SFF. EMBO J 19: 3750–3761 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Pramila T, Wu W, Miles S, Noble WS, Breeden LL (2006) The forkhead transcription factor Hcm1 regulates chromosome segregation genes and fills the S-phase gap in the transcriptional circuitry of the cell cycle. Genes Dev 20: 2266–2278 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Rothstein R (1991) Targeting, disruption, replacement, and allele rescue: integrative DNA transformation in yeast. Meth Enzymol 194: 281–302 [DOI] [PubMed] [Google Scholar]
  32. Sbia M, Parnell EJ, Yu Y, Kretschmann KL, Voth WP, McBride HJ, Stillman DJ (2007) Regulation of Ace2 DNA-binding by the Cbk1 kinase (submitted)
  33. Sherman F (1991) Getting started with yeast. Meth Enzymol 194: 1–21 [DOI] [PubMed] [Google Scholar]
  34. Sherriff JA, Kent NA, Mellor J (2007) The Isw2 chromatin-remodeling ATPase cooperates with the Fkh2 transcription factor to repress transcription of the B-type cyclin gene CLB2. Mol Cell Biol 27: 2848–2860 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Simon I, Barnett J, Hannett N, Harbison CT, Rinaldi NJ, Volkert TL, Wyrick JJ, Zeitlinger J, Gifford DK, Jaakkola TS, Young RA (2001) Serial regulation of transcriptional regulators in the yeast cell cycle. Cell 106: 697–708 [DOI] [PubMed] [Google Scholar]
  36. Spellman PT, Sherlock G, Zhang MQ, Iyer VR, Anders K, Eisen MB, Brown PO, Botstein D, Futcher B (1998) Comprehensive identification of cell cycle-regulated genes of the yeast Saccharomyces cerevisiae by microarray hybridization. Mol Biol Cell 9: 3273–3297 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Thomas BJ, Rothstein R (1989) Elevated recombination rates in transcriptionally active DNA. Cell 56: 619–630 [DOI] [PubMed] [Google Scholar]
  38. Toyn JH, Johnson AL, Donovan JD, Toone WM, Johnston LH (1997) The Swi5 transcription factor of Saccharomyces cerevisiae has a role in exit from mitosis through induction of the cdk-inhibitor Sic1 in telophase. Genetics 145: 85–96 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Traven A, Jelicic B, Sopta M (2006) Yeast Gal4: a transcriptional paradigm revisited. EMBO Rep 7: 496–499 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. van Kempen GM, van Vliet LJ (2000) Mean and variance of ratio estimators used in fluorescence ratio imaging. Cytometry 39: 300–305 [DOI] [PubMed] [Google Scholar]
  41. Voth WP, Olsen AE, Sbia M, Freedman KH, Stillman DJ (2005) ACE2, CBK1, and BUD4 in budding and cell separation. Eukaryot Cell 4: 1018–1028 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Whalen AM, Steward R (1993) Dissociation of the dorsal-cactus complex and phosphorylation of the dorsal protein correlate with the nuclear localization of dorsal. J Cell Biol 123: 523–534 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Zhu G, Spellman PT, Volpe T, Brown PO, Botstein D, Davis TN, Futcher B (2000) Two yeast forkhead genes regulate the cell cycle and pseudohyphal growth. Nature 406: 90–94 [DOI] [PubMed] [Google Scholar]

Associated Data

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

Supplementary Materials

Supplementary data

7601859s1.pdf (567.1KB, pdf)

Articles from The EMBO Journal are provided here courtesy of Nature Publishing Group

RESOURCES