• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of molcellbPermissionsJournals.ASM.orgJournalMCB ArticleJournal InfoAuthorsReviewers
Mol Cell Biol. Oct 2005; 25(20): 9127–9137.
PMCID: PMC1265784

Global Regulation by the Yeast Spt10 Protein Is Mediated through Chromatin Structure and the Histone Upstream Activating Sequence Elements

Abstract

The yeast SPT10 gene encodes a putative histone acetyltransferase (HAT) implicated as a global transcription regulator acting through basal promoters. Here we address the mechanism of this global regulation. Although microarray analysis confirmed that Spt10p is a global regulator, Spt10p was not detected at any of the most strongly affected genes in vivo. In contrast, the presence of Spt10p at the core histone gene promoters in vivo was confirmed. Since Spt10p activates the core histone genes, a shortage of histones could occur in spt10Δ cells, resulting in defective chromatin structure and a consequent activation of basal promoters. Consistent with this hypothesis, the spt10Δ phenotype can be rescued by extra copies of the histone genes and chromatin is poorly assembled in spt10Δ cells, as shown by irregular nucleosome spacing and reduced negative supercoiling of the endogenous 2μm plasmid. Furthermore, Spt10p binds specifically and highly cooperatively to pairs of upstream activating sequence elements in the core histone promoters [consensus sequence, (G/A)TTCCN6TTCNC], consistent with a direct role in histone gene regulation. No other high-affinity sites are predicted in the yeast genome. Thus, Spt10p is a sequence-specific activator of the histone genes, possessing a DNA-binding domain fused to a likely HAT domain.

Chromatin structure plays an essential role in gene regulation. The structural unit of chromatin is the nucleosome, which is composed of 147 bp of DNA wrapped in a negative superhelix around a central octamer of core histones (composed of two molecules each of H2A, H2B, H3, and H4) (25). Nucleosomes are separated by linker DNA, forming a “beads on a string” structure. A fifth histone, H1, binds to both the nucleosome and the linker DNA to drive the coiling of the nucleosomal filament to form the 30-nm fiber.

The nucleosome presents a problem for regulatory proteins seeking access to DNA because so much of the DNA is protected by histones: the inner surface of the DNA is completely occluded by the central core, the external surface is at least partly protected by the core histone tail domains, and the DNA coils are so close together that their apposed surfaces are also unavailable. To cope with the intrinsically repressive nature of the nucleosome structure, regulatory proteins recruit two types of chromatin remodeling complex to promoters: (i) chromatin remodeling machines, which use ATP to move nucleosomes, effect nucleosomal conformational changes, and exchange core histones with variants (33); and (ii) chromatin modifying enzymes, which catalyze posttranslational modifications of the histones, mostly in their tail domains. These modifications are proposed to represent a “histone code” which is read by regulatory proteins that recognize particular combinations of modifications, resulting in activation or silencing of chromatin (41). Histone acetylation is generally associated with gene activation and is catalyzed by histone acetyltransferases (HATs). The identification of the Gcn5p coactivator as a HAT led to a breakthrough in the field, connecting transcription factors with chromatin (3). The current paradigm is that histone modifying complexes are cofactors recruited to promoters by sequence-specific activators or repressors.

Our studies have focused on the CUP1 gene of Saccharomyces cerevisiae as a model for the role of chromatin in gene regulation (35-37). CUP1 encodes a metallothionein responsible for protecting cells from the toxic effects of copper. The induction of CUP1 by copper results in targeted acetylation of nucleosomes at the CUP1 promoter. This acetylation is dependent on SPT10 (23, 37), which encodes a putative HAT related to Gcn5p (29). SPT10 was originally identified as one of a set of SPT genes, mutations in which suppress insertion mutations due to the Ty1 transposable element (8). The SPT genes encode many proteins important in transcription, including subunits of the SAGA histone modifying complex (12), TATA-binding protein, and histones (44, 47). SPT10 is not an essential gene, but the null allele is associated with very slow growth and defects in gene regulation (7, 27, 28). SPT10 has been implicated as a global regulator of core promoter activity, acting at or near the TATA box (5, 34, 48). Recently, it has been shown that SPT10 is required for cell cycle-specific acetylation of lysine-56 of H3 at the histone gene promoters (46). However, the mechanism by which Spt10p exerts its effects through basal promoters is unclear.

Spt10p has also been identified as an activator of the histone genes, which it regulates in conjunction with Spt21p, the Hir corepressor (7, 15, 38), and SWI/SNF (6, 46). In S. cerevisiae, there are four loci encoding the major core histones, each containing a pair of divergent genes transcribed from a central promoter (31). Two of these loci, HHT1-HHF1 and HHT2-HHF2, carry genes for H3 (HHT1 and -2) and H4 (HHF1 and -2) (40); the others, HTA1-HTB1 and HTA2-HTB2, carry genes for H2A (HTA1 and -2) and H2B (HTB1 and -2) (14). Chromatin immunoprecipitation (ChIP) experiments have shown that Spt10p is present at all four core histone promoters in vivo (15, 46).

Here we address the mechanism by which Spt10p acts as a global regulator. Using microarray analysis, we identified hundreds of genes affected by Spt10p. However, we could demonstrate the presence of Spt10p only at the histone gene promoters in vivo, suggesting that Spt10p acts indirectly on the other genes through regulation of the histone genes. In support of this hypothesis, we present evidence that chromatin is incorrectly assembled in spt10Δ cells. Remarkably, we found that Spt10p is a sequence-specific DNA-binding protein which binds with a high affinity to the upstream activating sequences (UAS elements) in all four major core histone promoters. Due to the highly cooperative nature of the binding, no other high-affinity sites are predicted in the yeast genome.

MATERIALS AND METHODS

Yeast strains.

BJ-spt10Δ was constructed by transformation of BJ5459 (MATa ura3-52 trp1 lys2-801 leu2Δ1 his3Δ200 pep4Δ::HIS3 prb1Δ1.6R can1 GAL cir+; ATCC 208284) (18) with the 1,794-bp spt10Δ::URA3 KpnI-PacI fragment obtained by digestion of pNEB-SPT10ΔURA3 (37). BJ-SPT10-HA/FLAG was constructed by transformation of BJ5459 with a SacI-HindIII digest of p400 to insert three hemagglutinin (HA) and three FLAG tags at the C terminus of SPT10, with URA3 downstream. This strain had no obvious phenotype.

Plasmids. (i) Construction of p400.

The 2,156-bp BamHI-AvrII fragment containing the SPT10 coding region from pNEB-SPT10B (37) was inserted into pRS416 (Stratagene) cut with BamHI and XbaI to create p361. The SPT10 promoter was inserted by replacing the HindIII-BstEII fragment with a version made by PCR using the primers 5′-GCGCGCGGATCCAATAGTGTCTCGTTCACGG and 5′-GCGCGCAAGCTTACTAGTCATTTGCGGTG to create p374, which contained the entire SPT10 gene from −290 relative to the start codon to +2159 in the 3′ flanking region. The 107-bp HpaI-BsrGI fragment encompassing the SPT10 stop codon was replaced with a 210-bp version including three HA tags followed by a BamHI site inserted just before the stop codon. This fragment was made by ligating three synthetic oligonucleotides together in frame to create p375 (the sequence of the three HA tags was GGATCGTACCCATACGATGTTCCAGATTACGCTGGATCTTACCCATACGATGTTCCAGATTACGCTGGATCTTACCCATACGATGTTCCAGATTACGCTGGATCC). The URA3 gene was inserted at the filled BsrGI site downstream of SPT10 as an 1,162-bp PmeI-SmaI fragment from pNEB-URA3 (36) in the opposite orientation to SPT10 to create p376. The 3,743-bp SacI-HindIII SPT10-3HA fragment was transferred to pNEB193 (New England Biolabs) cut with the same enzymes to create p355. A synthetic oligonucleotide encoding three FLAG tags with sticky ends for BamHI (GATCGGATTATAAAGATGACGATGACAAGGATTATAAAGATGACGATGACAAGGATTAT AAAGATGACGATGACAAGG) was inserted at the BamHI site after the three HA tags in p355 to create p400; the sequence of the entire insert was verified.

(ii) Construction of pRS425-HHT1-HHF1-HTA1-HTB1.

The HHT1-HHF1 locus was obtained as an 1,870-bp PmeI-BsaAI fragment from pCC67 (4) and inserted at the SmaI site of pNEB193 to create p367. The HHT1-HHF1 locus was then transferred as an 1,894-bp BamHI-SacI fragment from p367 to pRS425 (a 2μm vector carrying LEU2 as a selection marker) to create p482. The HTA1-HTB1 locus was obtained as a 2,903-bp BamHI-MscI fragment from pCC67 and inserted into pRS414 (Stratagene) to create p474. The HTA1-HTB1 locus was then transferred as a 2,948-bp BamHI-XhoI fragment to pRS425 cut with the same enzymes to form pRS-HHT1-HHF1-HTA1-HTB1 (p484).

The SPT10-HA/FLAG open reading frame was obtained by PCR using 5′-GGCGCGGAGCTCATGCTAAATCAGCACACAAGTTCAG as the forward primer and 5′-GGCGCGCTCGAGCGTCTCTGGCAGAGTCGC as the reverse primer, with p400 as the template. The purified 2,145-bp fragment was digested with SacI and XhoI and inserted at the same sites in the baculoviral expression vector pFastBac1 (Invitrogen) to create p439.

Microarrays.

RNAs were prepared from BJ5459 and BJ-spt10Δ cells grown to an A600 of 0.6 on three separate occasions using an RNeasy kit (QIAGEN) and were sent to Expression Analysis (High Point, NC) for the synthesis of biotinylated cRNAs. Hybridization to Affymetrix S98 microarrays was detected using streptavidin conjugated to a fluorescent probe. Gene expression levels were analyzed by two-group comparison (see the table in the GEO database [http://www.ncbi.nlm.nih.gov/projects/geo/] under accession number GSE2407).

ChIP assays.

ChIP assays were performed as described previously (26), using 5 μg DNA per sample (DNA was measured using the Hoechst assay) and anti-HA antibody (Roche 12CA5; 7.5 μg was prebound to 20 μl protein G magnetic beads [Dynal] per sample). HA-peptide (Sigma I-2149) was used for the specific elution of bound chromatin, as follows: washed beads were resuspended in 10 μl 0.1 M HEPES-K, pH 7.5, 20% glycerol, 0.1 M KCl, 0.6 mM EDTA, 2 mM MgCl2, and 0.02% NP-40 containing 50 μg HA-peptide, and 40 μl 10 mM Tris-HCl, pH 8.0, 0.25 M LiCl, 0.5% NP-40, 0.5% sodium deoxycholate, and 1 mM sodium EDTA (RIPA buffer) containing protease inhibitors (Roche) was added and placed in an Eppendorf Thermomixer (22°C, 1,200 rpm, 30 min). The elution was repeated, and the eluates were pooled. DNAs in the immunoprecipitates were measured by quantitative PCR with the primers listed in Table Table11.

TABLE 1.
PCR primers used for this study

Chromatin experiments.

BJ5459 and BJ-spt10Δ were grown in synthetic complete medium to an A600 of 0.5. Nuclei were prepared from 250 A600 units of cells as described previously (36), resuspended in 2 ml 10 mM HEPES, pH 7.5, 0.5 mM MgCl2, and 0.05 mM CaCl2 with protease inhibitors as described above, divided into six aliquots, and warmed to 37°C. Nuclei were digested with micrococcal nuclease (MNase; Worthington) for 5 min at 37°C, EDTA was added to 1 mM, and sodium dodecyl sulfate was added to 1%. Purified DNAs were dissolved in 40 μl 50 mM Tris-HCl, pH 8.0, 5 mM sodium EDTA with 50 μg/ml RNase. For topology of the 2μm circle, genomic DNA was extracted from spheroplasts and analyzed in two-dimensional chloroquine gels (0.8% agarose) as described previously (21): the first dimension was 10 μg/ml chloroquine diphosphate (55 V, 15 h), and the second dimension was 20 μg/ml chloroquine diphosphate (35 V, 15 h). The gel was blotted and probed with the 645-bp HpaI-SnaBI 2μm DNA fragment from pYES2 (Invitrogen) labeled by random priming.

Recombinant Spt10p.

The SPT10-HA/FLAG gene from p439 was recombined into a baculovirus and expressed in Hi5 cells (Invitrogen) at a multiplicity of infection of 1.0 for 72 h. The presence of Spt10p was confirmed by Western blotting using an anti-HA monoclonal antibody (F7392; Santa Cruz). Spt10p was purified from a 45-ml culture by affinity chromatography as follows. Protein G-Sepharose (100 μl of packed resin; Sigma) was equilibrated with 5-mg/ml immunoglobulin G-free bovine serum albumin (BSA) (Sigma) in phosphate-buffered saline (PBS) and then with 400 μg anti-FLAG M2 monoclonal antibody (Sigma) in BSA-PBS overnight with rotation at 4°C. The resin was washed twice with BSA-PBS. Cells were lysed in 3 ml 0.5 M NaCl, 50 mM Tris-HCl, pH 8.0, 5 mM sodium EDTA, 1% Triton X-100, 10% glycerol, 1 mM 2-mercaptoethanol (2-ME) with protease inhibitors as described above, using a Dounce homogenizer. Cell debris was pelleted in a microcentrifuge (5 min, 14,000 rpm, 4°C). The lysate was incubated with antibody-resin and mixed by rotation (2.5 h, 4°C). The resin was washed twice with 1 ml lysis buffer (5 min on rotator at 4°C) and twice with 1 ml 0.15 M NaCl, 50 mM Tris-HCl, pH 8.0, 5 mM sodium EDTA, 0.2% Triton X-100, 10% glycerol, 1 mM 2-ME with protease inhibitors as described above. Spt10p was eluted (30 min of rotation, 4°C) using a three-FLAG-containing peptide (Sigma) at 5 mg/ml (dissolved in wash buffer with 0.2 M Tris-HCl, pH 8.0, to maintain the pH): 50 μl peptide was added to the drained pellet, followed by 50 μl wash buffer. The elution was repeated. The Spt10p eluates were pooled and stored at −80°C. Protein concentrations were measured by comparison with BSA standards (Bio-Rad) in a protein gel.

Gel shift assays.

Promoter fragments were prepared by PCR using yeast genomic DNA as the template. Synthetic oligonucleotides were gel purified and annealed to their complements. Fragments and oligonucleotides were end labeled using T4 kinase (the oligonucleotides had single-base 5′ extensions to facilitate labeling). Spt10p was added to labeled DNA at 2 nM in 15 μl of 20 mM Tris-HCl, pH 8.0, 0.15 M KCl, 5 mM MgCl2, 0.1 mM zinc acetate, 10% glycerol, 0.1 mg/ml BSA, 5 mM 2-ME, 0.1 mg/ml poly(dI-dC)-poly(dI-dC) (Amersham) and incubated at room temperature for 30 min. Samples were analyzed in 5% (19:1) polyacrylamide gels containing 20 mM Tris-acetate, 1 mM sodium EDTA, pH 8.3 (0.5× TAE), and 5% glycerol, with 0.5× TAE as the running buffer (150 V, 2 h).

Bioinformatics.

The histone UAS motif was defined using Gibbs (43), AlignAce (17), and GLAM (11), with the HHF1-HHT1, HHT2-HHF2, HTA2-HTB2, and HTA1-HTB1 intergenic regions as input (all programs gave similar results). A position frequency matrix derived from the Gibbs top scoring alignment (see Fig. 6A and B) was converted to a position weight matrix and used to scan all yeast intergenic regions using Clover (10).

FIG. 6.
Bioinformatic analysis of yeast histone promoters, with a consensus sequence for the histone UAS element. (A) Alignment of histone motif sequences found in the histone promoters using Gibbs sampling. This method searches for statistically significant ...

RESULTS

Spt10p behaves as a global regulator of gene expression.

To determine whether Spt10p is indeed a global regulator of gene expression, we compared differences in gene expression between wild-type and spt10Δ cells using expression microarray analysis. RNAs were prepared from both strains on three separate occasions. The average signal for each gene was computed for wild-type and spt10Δ cells, and a P value was calculated to measure the significance of the comparison (see the table in the GEO database [http://www.ncbi.nlm.nih.gov/projects/geo/] under accession number GSE2407). The data are summarized in Fig. Fig.1.1. A relatively large number of genes (827, or 13% of all open reading frames) were affected more than twofold in spt10Δ cells (441 of these genes had P values of <0.05). Thus, Spt10p behaved as a global regulator. Of the 827 genes affected more than twofold, a substantial majority were up-regulated in spt10Δ cells (77%), indicating that Spt10p repressed more genes than it activated.

FIG. 1.
Spt10p behaves as a global regulator. The plot shows a comparison of gene expression levels in wild-type and spt10Δ cells derived from microarray data. The plot includes all open reading frames (noncoding genes were excluded). Each point represents ...

The gene most affected in spt10Δ cells was SPT10 itself (down-regulated 61-fold, to background levels), as expected because SPT10 had been deleted. The expression of URA3 was up 16-fold in spt10Δ cells, reflecting the fact that URA3 was used to knock out SPT10.

It was shown previously that spt10Δ cells tend to flocculate (27; our unpublished observations). This can be accounted for using our data by the up-regulation of the flocculin genes FLO1 (10-fold), FLO9 (3-fold), MUC1 (6-fold), and FLO10 (2-fold). We observed increased ADH2 expression in spt10Δ cells (up 13-fold), as shown previously (5). Natsoulis et al. (27) observed increased expression of PHO5 and STE6 in spt10Δ cells; we confirmed the effect on PHO5 (up eightfold), but we observed a decrease in STE6 (down twofold), probably because STE6 is expressed only in MATa strains (our strain was MATa; Natsoulis et al. used a MATα strain and observed derepression). Prelich and Winston (34) reported that spt10 mutations can suppress UAS mutations in SUC2; we observed increased SUC2 expression (fourfold). Our data were less consistent with those of Yamashita (48), who presented evidence that the basal promoters of CYC1, CUP1, HIS3, PUT1, and PUT2 were activated in spt10Δ cells when linked to a lacZ reporter gene. In our study, only PUT1 was significantly affected (up fourfold). We did not observe any effect of the spt10Δ mutation on CUP1 basal levels either previously (37) or in the present study. These discrepancies might reflect the artificial nature of the lacZ reporter. In general, though, our array data were in reasonable agreement with published data. However, a comparison of our data with those for spt10Δ cells in a recent microarray expression study by Xu et al. (46) revealed only a very weak correlation between the two data sets, although some of the most strongly affected genes are the same. This is surprising but might be explained by the fact that we used synthetic complete medium, whereas Xu et al. (46) used yeast extract-peptone-dextrose. The difference in strain background might also be a contributing factor.

More genes that were strongly up-regulated in spt10Δ cells included HUG1 (a highly inducible gene functioning in DNA repair), THI11 (involved in thiamine biosynthesis), and SSA3 (encodes a heat shock protein). Genes down-regulated in spt10Δ cells included NPL3 (encodes a nuclear shuttling protein) and MEP2 (encodes an ammonia transporter). Northern blot analysis of these genes confirmed the array data (not shown).

Some of the histone genes (HHF2, HTA2, and HTB2) are down-regulated in spt10Δ cells (7, 15). Our array data indicated that only HTB2 was affected more than twofold (down fourfold); the other histone genes were down-regulated less than twofold. However, the array data gave only average results for HHT1/HHT2 and HHF1/HHF2 because these gene pairs are so closely related that the probes cannot distinguish them. Hess et al. (15) used reverse transcription-PCR with primers directed against the 5′ untranslated regions to measure the expression of all eight histone genes. Using the same method, we confirmed that HTA2, HTB2, and HHF2 were down-regulated in spt10Δ cells (Fig. (Fig.22).

FIG. 2.
Spt10p is not detected at the promoters of selected genes that are strongly affected by Spt10p but is present at the major core histone promoters in vivo. ChIP experiments were used to detect HA-tagged Spt10p using anti-HA antibody. The wild type (WT; ...

Spt10p is detectable in vivo only at the major core histone promoters.

We tested the promoters of some genes strongly affected in spt10Δ cells for the presence of Spt10p in vivo. ChIP experiments were carried out using a strain in which the wild-type SPT10 gene carried three HA and three FLAG tags at its C terminus. Sonicated chromatin fragments were prepared from untagged (wild-type) and tagged cells that had been fixed with formaldehyde. Anti-HA antibody was used for immunoprecipitation (IP). The amount of DNA in each IP was measured by quantitative PCR (Fig. (Fig.2),2), using an intergenic region from chromosome V as a control (22). A series of input dilutions was used to establish that each sample was in the linear range for PCR. A comparison of the IP and the mock sample from the tagged strain revealed that Spt10p was absent from the promoters of all genes tested (Fig. (Fig.2).2). In contrast, we confirmed the presence of Spt10p at all four major core histone promoters (HTA1-HTB1, HTA2-HTB2, HHT1-HHF1, and HHT2-HHF2), as first reported by Hess et al. (15). Spt10p was not detected at the HTZ1 promoter (which encodes H2AZ).

Thus, Spt10p affected the expression of hundreds of genes, but it was detectable in vivo only at the histone gene promoters, suggesting that Spt10p might act on other genes indirectly. Of course, we have only tested a tiny fraction of the 827 genes affected, but those chosen were among the genes which responded most strongly to the loss of Spt10p. It seemed very unlikely that Spt10p did not cross-link well to all seven promoters. Therefore, we considered an alternative hypothesis, namely, that Spt10p acts primarily at the histone promoters, activating transcription to help regulate the supply of core histones. The deletion of SPT10 might therefore result in a shortage of core histones for assembly of the genome into chromatin. This in turn might result in derepression of basal promoters and aberrant gene expression on a global scale. Thus, global regulation by Spt10p might be the indirect result of reduced histone synthesis and the consequent assembly of defective chromatin.

Similar sets of genes are affected in spt10Δ cells and in cells depleted of histone H4.

The artificial depletion of H4 from yeast cells results in the derepression of many genes in a UAS-independent manner, providing evidence for a repressive role for nucleosomes at some basal promoters, including PHO5 and GAL1 (13). More recently, this experiment was repeated on a genome-wide scale using microarrays (45). Our hypothesis is similar to a histone depletion mechanism in that it supposes that there is a dearth of histones in spt10Δ cells. Accordingly, we reasoned that similar sets of genes might be affected in spt10Δ and H4-depleted cells, particularly because PHO5 (up eightfold) and GAL1 (up threefold) were strongly affected in spt10Δ cells.

To assess the degree of overlap in expression data for spt10Δ and H4-depleted cells, we compared our microarray data with those of Wyrick et al. (45). A plot of expression ratios for each gene (H4-depleted cells/wild-type cells and spt10Δ cells/wild-type cells) revealed a good correlation between the two sets of genes (R2 = 0.41) (Fig. (Fig.3A).3A). Wyrick et al. (45) observed that subtelomeric genes tended to be derepressed in H4-depleted cells; similarly, we have noted some marked clustering of genes affected in spt10Δ cells near telomeres (not shown). As a control, expression array data for gcn5Δ cells (24) were compared in the same way with data for H4-depleted cells; these were very poorly correlated (R2 = 0.03). Thus, similar sets of genes were affected in spt10Δ and H4-depleted cells, suggesting that spt10Δ cells might be suffering from histone depletion.

FIG. 3.
The phenotype of spt10Δ cells is primarily due to reduced histone gene expression. (A) Similar sets of genes are affected in spt10Δ cells and in cells depleted of H4. The left plot shows a comparison of microarray data for spt10Δ ...

Extra copies of HHT1-HHF1 and HTA1-HTB1 rescue the spt10Δ growth phenotype.

If, as we propose, global regulation by Spt10p is the indirect result of reduced histone synthesis, it might be possible to rescue the growth defect of spt10Δ cells by supplying extra copies of the histone genes. Indeed, a partial rescue of the spt10Δ phenotype has been demonstrated previously using a centromeric plasmid carrying the HTA1-HTB1 locus (38). Accordingly, wild-type and spt10Δ cells were transformed with a multicopy yeast plasmid carrying both the HHT1-HHF1 and HTA1-HTB1 loci, and spot dilution tests were performed (Fig. (Fig.3B).3B). The extra histone gene copies had no effect on wild-type cells, as observed previously (4), but they rescued the growth phenotype of spt10Δ cells. This rescue was apparently complete, since the growth of spt10Δ cells carrying the plasmid was very similar to that of wild-type cells. Multiple extra copies of the histone genes were necessary for rescue of the growth phenotype, since a low-copy-number plasmid (CEN/ARS) failed to rescue (not shown), perhaps reflecting the importance of balancing the number of gene copies encoding H3/H4 and H2A/H2B (4, 38). The fact that extra copies of the histone genes could correct the growth defect of spt10Δ cells is strong evidence in support of the hypothesis that the primary defect in spt10Δ cells is indeed reduced histone gene expression.

Defective chromatin is assembled in spt10Δ cells.

To test whether chromatin assembled in spt10Δ cells is defective, its nucleosomal organization was assessed by digesting nuclei with MNase (Fig. (Fig.4A).4A). Bulk chromatin from wild-type cells gave rise to the typical yeast nucleosomal repeat of about 160 bp (42), with clearly defined bands corresponding to one to seven nucleosomes. In contrast, bulk chromatin from spt10Δ cells gave a much less distinct nucleosomal repeat: the bands corresponding to dinucleosomes and above were relatively indistinct due to a strong background smear. (A similar result has been obtained by others [D. Hess and F. Winston, personal communication].) This smear probably reflected the presence of irregularly spaced nucleosomes and/or incompletely assembled or missing nucleosomes. In either case, the organization of chromatin in spt10Δ cells is defective on a global scale.

FIG. 4.
Defective chromatin structure is formed in spt10Δ cells. (A) Nucleosomal organization in spt10Δ cells is defective. Nuclei prepared from wild-type and spt10Δ cells were digested with increasing amounts of MNase; purified DNA was ...

Another assay commonly used to assess chromatin assembly is to measure the number of negative supercoils in a circular plasmid. The assay is based on the fact that a nucleosome protects one negative supercoil from relaxation by topoisomerases (39). Thus, the number of nucleosomes on a circular DNA molecule is equal to the number of negative supercoils. This is measured by resolving plasmid topoisomers in a two-dimensional chloroquine gel (topoisomers differ only in the number of supercoils they contain). Most laboratory strains of yeast contain the 2μm circle, an endogenous 6.3-kb plasmid containing four genes (FLP, D, REP1, and REP2) (2). Our microarray data indicated that FLP (up 3-fold) and D (up 5-fold) were strongly affected in spt10Δ cells, whereas REP1 (up 1.4-fold) and REP2 (up 1.9-fold) were not.

Accordingly, we compared the chromatin structures of the 2μm circle in wild-type and spt10Δ cells. Firstly, we confirmed that the nucleosomal repeat in 2μm chromatin from spt10Δ cells was disrupted by probing a Southern blot of MNase-digested chromatin with 2μm plasmid DNA (Fig. (Fig.4B).4B). 2μm chromatin had a very regular nucleosomal repeat in wild-type cells (up to 13 distinct bands) but was much less regular in spt10Δ cells, exhibiting the same background smear observed for bulk chromatin (compare the scans in Fig. Fig.4B).4B). Secondly, we compared the topologies of 2μm DNAs extracted from wild-type and spt10Δ cells (Fig. (Fig.4C).4C). The wild-type plasmid exhibited the expected Gaussian distribution of negatively supercoiled topoisomers. In contrast, the 2μm plasmid from spt10Δ cells was less negatively supercoiled than that in the wild type; the shift in the center of the topoisomer distribution corresponded to a loss of about three negative supercoils. This is equivalent to the loss of three nucleosomes, or about 10% of the nucleosomes on the 2μm circle. Furthermore, the topoisomer distribution was much broader than that of the wild type (compare the scans in Fig. Fig.4C),4C), indicating that 2μm chromatin was quite heterogeneous in spt10Δ cells and that a subfraction of 2μm circle chromatin was much more poorly assembled.

In conclusion, defective chromatin is formed in spt10Δ cells.

Recombinant Spt10p binds specifically to multiple sites in the core histone promoters.

We propose that Spt10p acts primarily at the core histone promoters to regulate histone gene transcription. Although HATs are generally coactivators, requiring recruitment to promoters by activators, it seemed that Spt10p might be unusual because it contains a possible DNA-binding domain, i.e., a putative basic zinc finger (residues 351 to 388 [unpublished observations]). We addressed whether Spt10p binds specifically to histone promoter DNA using a gel shift assay.

Initially, we attempted to prepare recombinant Spt10p from Escherichia coli, but it was insoluble. Therefore, Spt10p with three HA and three FLAG tags at its C terminus was expressed in insect cells using a baculoviral promoter and then purified by affinity chromatography using anti-FLAG antibody (Fig. (Fig.5A).5A). For gel shift experiments, the HTA1-HTB1 promoter was divided into three overlapping fragments made by PCR (F1, F2, and F3). End-labeled DNA fragments were incubated with increasing amounts of Spt10p (Fig. (Fig.5B).5B). F1 contained all four UAS elements and the cell cycle regulatory (CCR) box (31); F2 contained one UAS element and the region upstream of the TATA box; and F3 corresponded to the basal promoter of HTA1, including the TATA box. Spt10p bound to F1 with a high affinity to form two major complexes. Spt10p formed a complex of much lower affinity with F2 and no complexes at all with F3. Thus, Spt10p bound specifically to the major regulatory region of the HTA1-HTB1 promoter. The fact that Spt10p formed more than one specific complex on F1 indicated the presence of multiple binding sites.

FIG. 5.
Spt10p binds specifically to multiple sites in all four major core histone promoters. (A) Purified recombinant Spt10p. A Coomassie-stained gel of Spt10p affinity purified using anti-FLAG antibody from insect cells infected with a baculovirus expressing ...

The other histone promoters were tested for binding to Spt10p (Fig. (Fig.5C).5C). All four major core histone promoters contained multiple high-affinity binding sites for Spt10p (HHT1-HHF1, HHT2-HHF2, HTA1-HTB1, and HTA2-HTB2). The dissociation constant (KD) for the binding of Spt10p to these promoter fragments was estimated to be 20 to 60 nM. The promoters of the genes encoding the other yeast histones (H2AZ, centromeric H3, and H1) were also examined, but none contained binding sites for Spt10p.

The promoters of various genes that were strongly affected in spt10Δ cells were tested for binding of Spt10p (HUG1, ADH2, PHO5, THI11, SSA3, MEP2, and NPL3) (Fig. (Fig.5D).5D). None were recognized by Spt10p. We have shown previously that CUP1 is poorly induced in spt10Δ cells (37). However, the CUP1 promoter also lacked Spt10p binding sites.

The presence of Spt10p binding sites in the major core histone promoters and their absence in the promoters of other strongly affected genes are consistent with our observations in vivo (Fig. (Fig.2).2). These results lend further support to our hypothesis that Spt10p acts primarily through regulation of the histone genes and that its effects on other genes are indirect and mediated through alterations in chromatin structure.

Spt10p binds with high cooperativity to pairs of histone UAS elements.

The presence of multiple binding sites in all four major core histone promoters led us to analyze their sequences for common motifs. A sequence element present in multiple copies in all of the major core histone promoters was identified using the Gibbs sampling method (Fig. (Fig.6A).6A). This element corresponded to the UAS identified many years ago (9, 32), but the protein which recognizes it has remained elusive. The consensus sequence is TTCN2AN4TTC(G/T)C (Fig. (Fig.6B).6B). It is almost the same as those inferred previously (9, 31). The consensus of Osley (31), GTTCN2ANTTTTTCGC, differs from ours only in specifying more bases. However, the use of our consensus revealed several previously unrecognized copies of this element (Fig. (Fig.6C).6C). The UAS elements tend to be clustered in the center of each divergent promoter. The HHT1-HHF1 and HTA1-HTB1 promoters both have four perfect consensus sequences, but the HHT2-HHF2 and HTA2-HTB2 promoters are more complex in their organization, having six possible UAS elements (although some are relatively weak matches to the consensus).

To address the question of whether Spt10p recognizes the histone UAS, synthetic double-stranded oligonucleotides were used in the gel shift assay. Initially, we tested oligonucleotides corresponding to a UAS element in the HTB1 promoter. There was no detectable binding to the wild-type UAS oligonucleotide under the conditions used for the promoter DNA fragments (not shown). However, a complex was formed on the wild-type UAS at very high concentrations of both Spt10p and UAS-DNA; this complex was not formed on a mutated UAS (Fig. (Fig.7A).7A). The same result was obtained for one of the UAS elements near HTA1 (not shown). Thus, Spt10p bound specifically but very weakly to a single UAS element, with an apparent KD of [dbl greater-than sign]1 μM, in sharp contrast to the KD of 20 to 60 nM observed for histone promoter DNA.

FIG. 7.
Spt10p binds specifically and highly cooperatively to pairs of histone UAS elements. Gel shift assays were performed using end-labeled synthetic double-stranded oligonucleotides and recombinant Spt10p. Specific complexes are indicated by arrowheads; asterisks ...

A possible explanation for this result was that the binding of Spt10p to promoter DNA might be highly cooperative, involving interactions between Spt10p and more than one UAS element. To address this, longer oligonucleotides containing the two UAS elements in the HTA1 promoter were examined (Fig. (Fig.7B).7B). Spt10p bound tightly to the wild-type oligonucleotide with a high affinity, but only very weakly to oligonucleotides containing one intact and one mutated UAS element and not at all to an oligonucleotide with both UAS elements mutated. The KD for the wild-type oligonucleotide was approximately 50 nM in 150 mM KCl (used in all experiments described here). Spt10p bound more tightly in 50 mM KCl (KD = 10 nM [not shown]). Thus, Spt10p bound to a pair of UAS elements much more strongly than to a single UAS; this cooperativity was very strong, resulting in a >20-fold increase in binding affinity. Competition experiments using labeled promoter DNA indicated that the single UAS was a very poor competitor for Spt10p, whereas the UAS pair was very effective, as expected (not shown).

The mutations described above abolished Spt10p binding, but since almost the entire consensus was mutated, they proved only that the binding site included at least some of these base pairs. To assess the accuracy of the consensus sequence, we undertook a systematic point mutation analysis, changing one base at a time and assessing the effect of each mutation on binding. Since high-affinity binding required a pair of UAS elements, synthetic oligonucleotides containing the same point mutation in both UAS elements were used (Fig. (Fig.7C).7C). Beginning at the 5′ end of the consensus, it was found that the nucleotide at position 3 (Fig. (Fig.6B)6B) must be G or A for Spt10p to bind at all. In the first TTC motif in the consensus (positions 4, 5, and 6), the first T was essential for binding, the second T was much less important (though still very significant), and the C was very important (mutation resulted in a >20-fold decrease in affinity, i.e., a KD of >1 μM). Mutations in the bases flanking the TTC motifs (positions 7, 8, and 13) resulted in more modest reductions in binding affinity; mutating the other bases between the two TTC motifs had no effect (including the A at position 9, which was strong in the consensus). All three bases of the second TTC motif (positions 14 to 16) were very important (KD > 1 μM). Mutation of the following G/T and C nucleotides (positions 17 and 18) had modest effects (KD = 100 to 200 nM). Mutation of the C at position 19 had no effect. This binding study was in remarkable agreement with the weighted consensus sequence (Fig. (Fig.6B);6B); the most significant difference was that the predicted A between the TTC and TTC(G/T)C motifs was not important. In conclusion, according to our gel shift analysis, the Spt10p binding site is best represented by the sequence (G/A)TTCCN6TTCNC.

This analysis enabled us to assess the predicted UAS elements (Fig. (Fig.6A)6A) for likely binding to Spt10p. All four UAS elements in the HHT1-HHF1 and HTA1-HTB1 promoters are perfect consensus sequences and should bind Spt10p. They occur in pairs, as required for cooperative, high-affinity binding. The single site in the HHO1-NAN1 promoter did not bind Spt10p (Fig. (Fig.5C)5C) and would not be expected to bind with a high affinity because it is a single element. In the HHT2-HHF2 promoter, elements 1, 4, and 6 lack complete TTC motifs and would not be expected to bind Spt10p; element 2 has TGC rather than TTC in the first motif, which would weaken but not abolish binding; and elements 3 and 5 are perfect matches. Since two major complexes were formed on this promoter (Fig. (Fig.5C),5C), it is likely that one or two of the weaker sites are functional when close to a perfect UAS element; this is currently under investigation. In the HTA2-HTB2 promoter, there are three perfect elements and one weaker element (element 4, which has TGC rather than TTC in the first motif)—these four elements probably account for the two complexes formed on this promoter (Fig. (Fig.5C).5C). Elements 1 and 6 are unlikely to bind Spt10p since they both lack important nucleotides.

In conclusion, Spt10p binds strongly and highly cooperatively to pairs of UAS elements in the major core histone promoters.

DISCUSSION

Mechanism of global regulation by Spt10p.

Our analysis of gene expression in spt10Δ cells has shown that many genes are affected by more than the arbitrary cutoff value of twofold, demonstrating that Spt10p behaves as a global regulator. As is generally the case in microarray experiments, some genes are activated and some are repressed. Spt10p behaves primarily as a repressor. At first glance, this is surprising for a putative HAT, because histone acetylation is usually associated with gene activation. However, it is important to distinguish direct from indirect effects; activation of a target gene encoding a repressor (e.g., histones) would result in repression of its target genes, resulting indirectly in opposite effects on expression. Thus, gene regulation should be considered an integrated network of functionally interacting components; the loss of a component is likely to result in ripple effects throughout the regulatory system. In the case of Spt10p, direct and indirect effects can be distinguished by determining whether Spt10p is present at a particular promoter by ChIP. Our results suggest that Spt10p represents an extreme case of indirect regulation.

We propose that Spt10p acts primarily at the histone gene promoters, which it activates, and that the global effects of the spt10Δ mutation result indirectly from reduced levels of histone synthesis resulting in incomplete assembly of the genome into chromatin. In support of this hypothesis, we have shown that (i) a similar set of genes is affected by H4 depletion (Fig. (Fig.3A),3A), (ii) the spt10Δ growth phenotype can be rescued by supplying extra histone genes (Fig. (Fig.3B),3B), (iii) the nucleosomal organization of chromatin is generally defective in spt10Δ cells (Fig. (Fig.4),4), and (iv) Spt10p binds specifically to pairs of UAS elements in the core histone promoters not found elsewhere in the genome (Fig. (Fig.55 to to7;7; see below). A more direct test of the hypothesis would be to measure the relative histone levels in wild-type and spt10Δ cells. However, the expected difference is only about 10 to 20%, based on the topology data (Fig. (Fig.4C),4C), which cannot be measured with any confidence.

The global effects of Spt10p are mediated through basal promoters (34, 37, 48) and perhaps cryptic promoters (19). This could be accounted for by relief of nucleosomal repression of basal promoters in spt10Δ cells. This predicts a general activation of genes in spt10Δ cells, but as discussed above, the activation of regulatory genes would complicate the pattern.

A puzzling observation made by Hess et al. (15) and confirmed by us is that the expression of only two of the four major histone loci (HHT2-HHF2 and HTA2-HTB2) is apparently affected in spt10Δ cells, even though Spt10p binds to all four promoters in vivo. In addition, we have demonstrated the presence of binding sites for Spt10p in all four promoters. However, Xu et al. (46) used synchronized cells and observed that all eight major core histone genes are strongly down-regulated in spt10Δ cells. Thus, the lack of synchronization in our experiments is the most likely explanation for the apparent absence of an effect on some of the histone genes; effects on HHT1-HHF1 and HTA1-HTB1 could be masked by the long doubling time of spt10Δ cells in normal culture. Another possible explanation for the apparent lack of an effect on HHT1-HHF1 and HTA1-HTB1 in spt10Δ cells is that the loss of histone promoter activation by Spt10p might be offset by an increased basal activity resulting from relief of nucleosomal repression at these promoters. In addition, HTA1-HTB1 normally exhibits dosage compensation: a deletion of HTA2-HTB2 is compensated for posttranscriptionally by increased HTA1-HTB1 expression (30). We observed reduced HTA2-HTB2 expression in spt10Δ cells but no compensatory increase in HTA1-HTB1 expression, implying that dosage compensation is impaired in spt10Δ cells.

Histone UAS element and Spt10p.

The histone UAS element was identified many years ago as a multicopy element that can activate a reporter gene in a cell cycle-dependent manner (32). We have shown here that Spt10p recognizes this element. The major core histone genes are cell cycle regulated to provide histones for S phase. However, the binding of Spt10p at the histone promoters is not cell cycle dependent, except at HTA2-HTB2 (15), suggesting that the activating function of Spt10p is regulated, perhaps by the Hir corepressor complex through the CCR box (6) or by Spt21p. Spt21p expression is cell cycle dependent, and it appears to regulate Spt10p binding in vivo (15). However, Spt10p function is only partially dependent on Spt21p, since spt10Δ cells grow much more slowly than spt21Δ cells and genetic evidence indicates that Spt10p and Spt21p have other, distinct roles in vivo (16).

The DNA-binding domain of Spt10p is likely to include a predicted H2C2 zinc finger (residues 351 to 388). We have confirmed that the binding of Spt10p to DNA is zinc dependent (in preparation). In addition, the C388S mutation results in very poor growth, similar to the null phenotype (28; confirmed by us), and the mutant protein does not bind to DNA in vitro or in vivo (in preparation).

The fact that Spt10p binds with very strong cooperativity to pairs of UAS elements might be of great biological significance: Spt10p binds to a single UAS specifically but very weakly, making it very unlikely that a single UAS would be recognized in vivo. Thus, Spt10p is expected to bind in vivo only where there is a pair of UAS elements. A computational search of yeast intergenic regions for the histone UAS consensus revealed just four examples of likely binding sites for Spt10p outside the major core histone promoters, but these were all single elements. More importantly, a search of the entire yeast genome revealed that only the major core histone promoters contain more than one UAS element. Thus, cooperativity is likely to guarantee that Spt10p binds with a high affinity only at the four major core histone promoters and nowhere else.

Putative HAT activity of Spt10p.

Attempts to demonstrate the HAT activity of Spt10p have been unsuccessful so far, but additional proteins might be required to activate it. The acetylation of H3-K56 at the histone promoters is dependent on Spt10p, suggesting that Spt10p might acetylate H3-K56 (46). Although this is strong circumstantial evidence, direct proof that Spt10p possesses H3-K56 acetylase activity is lacking. Our previous work demonstrated that acetylation of H3-K9/K14 and hyperacetylation of the H4 tail in nucleosomes at the CUP1 promoter is similarly dependent on Spt10p (37), but there are no binding sites for Spt10p in the CUP1 promoter (Fig. (Fig.5D)5D) and we were unable to demonstrate the presence of Spt10p at the CUP1 promoter by ChIP (not shown), suggesting an indirect effect at CUP1. Xu et al. (46) studied the SUC2 promoter, which also exhibits H3-K56 acetylation, but whether this acetylation is dependent on Spt10p or whether Spt10p is present at SUC2 was not reported. Since H3-K56 acetylation is not limited to the histone promoters (46), Spt10p might also be recruited as a coactivator to promoters lacking binding sites for Spt10p (although we could not detect it by ChIP). Alternatively, since there is only a small reduction in H3-K56 acetylation in spt10Δ cells, there must be another HAT for H3-K56 in yeast (46), which could be responsible for K56 acetylation at other promoters.

The putative HAT domain is certainly important for Spt10p function, because mutations designed to inactivate the HAT activity result in slow growth and effects on histone mRNA levels similar to those observed in spt10Δ cells (15). In addition, a protein containing the HAT domain fused to the LexA DNA-binding domain can activate a reporter gene, indicating that the HAT domain has an activating function (15). If Spt10p is an acetyltransferase, then it belongs to the novel class of regulatory proteins defined by the mammalian transcription factor ATF-2, in which a HAT enzyme is fused to a sequence-specific DNA-binding domain (20). Since Spt10p is a sequence-specific DNA-binding protein and possesses an activation function, it is best classified as an activator rather than a coactivator. Coactivators are generally recruited to promoters by activators and do not bind specifically to DNA by themselves.

In conclusion, global regulation by Spt10p is probably an indirect consequence of insufficient histone synthesis resulting from a reduced expression of HTA2-HTB2 and HHT2-HHF2. The implied decrease in histone supply in the absence of Spt10p is proposed to explain the poor assembly of the genome into chromatin, resulting in relief of nucleosomal repression of basal transcription and global changes in gene expression.

Acknowledgments

We thank Ambarish Bhat and Ali Naim for technical help, Rohinton Kamakaka for discussion, Fred Winston and Jerry Workman for comments on the manuscript, and David Hess and Fred Winston for sharing unpublished results.

REFERENCES

1. Broach, J. R. 1983. Construction of high copy yeast vectors using 2-micron circle sequences. Methods Enzymol. 101:307-325. [PubMed]
2. Broach, J. R., and F. C. Volkert. 1991. Circular DNA plasmids of yeast, p. 297-331. In The molecular and cellular biology of the yeast Saccharomyces cerevisiae: genome dynamics, protein synthesis and energetics. Cold Spring Harbor Press, Cold Spring Harbor, N.Y.
3. Brownell, J. E., J. Zhou, T. Ranalli, R. Kobayashi, D. G. Edmondson, S. Y. Roth, and C. D. Allis. 1996. Tetrahymena histone acetyltransferase A: a homologue to yeast Gcn5p linking histone acetylation to gene activation. Cell 84:843-851. [PubMed]
4. Clark-Adams, C. D., D. Norris, M. A. Osley, J. S. Fassler, and F. Winston. 1988. Changes in histone gene dosage alter transcription in yeast. Genes Dev. 2:150-159. [PubMed]
5. Denis, C. L., and T. Malvar. 1990. The CCR4 gene from Saccharomyces cerevisiae is required for both non-fermentative and spt-mediated gene expression. Genetics 124:283-291. [PMC free article] [PubMed]
6. Dimova, D., Z. Nackerdien, S. Furgeson, S. Eguchi, and M. A. Osley. 1999. A role for transcriptional repressors in targeting the yeast SWI/SNF complex. Mol. Cell 4:75-83. [PubMed]
7. Dollard, C., S. L. Ricupero-Hovasse, G. Natsoulis, J. D. Boeke, and F. Winston. 1994. SPT10 and SPT21 are required for transcription of particular histone genes in Saccharomyces cerevisiae. Mol. Cell. Biol. 14:5223-5228. [PMC free article] [PubMed]
8. Fassler, J. S., and F. Winston. 1988. Isolation and analysis of a novel class of suppressor of Ty insertion mutations in Saccharomyces cerevisiae. Genetics 118:203-212. [PMC free article] [PubMed]
9. Freeman, K. B., L. R. Karns, K. A. Lutz, and M. M. Smith. 1992. Histone H3 transcription in Saccharomyces cerevisiae is controlled by multiple cell cycle activation sites and a constitutive negative regulatory element. Mol. Cell. Biol. 12:5455-5463. [PMC free article] [PubMed]
10. Frith, M. C., Y. Fu, L. Yu, F. Chen, U. Hansen, and Z. Weng. 2004. Detection of functional DNA motifs via statistical over-representation. Nucleic Acids Res. 32:1372-1381. [PMC free article] [PubMed]
11. Frith, M. C., U. Hansen, J. L. Spouge, and Z. Weng. 2004. Finding functional sequence elements by multiple local alignment. Nucleic Acids Res. 32:189-200. [PMC free article] [PubMed]
12. Grant, P. A., D. E. Sterner, L. J. Duggan, J. L. Workman, and S. L. Berger. 1998. The SAGA unfolds: convergence of transcription regulators in chromatin-modifying complexes. Trends Cell Biol. 8:193-197. [PubMed]
13. Han, M., and M. Grunstein. 1988. Nucleosome loss activates yeast downstream promoters in vivo. Cell 55:1137-1145. [PubMed]
14. Hereford, L., K. Fahrner, J. Woolford, M. Rosbash, and D. B. Kaback. 1979. Isolation of yeast histone genes H2A and H2B. Cell 18:1261-1271. [PubMed]
15. Hess, D., B. Liu, N. R. Roan, R. Sternglanz, and F. Winston. 2004. Spt10-dependent transcriptional activation in Saccharomyces cerevisiae requires both the Spt10 acetyltransferase domain and Spt21. Mol. Cell. Biol. 24:135-143. [PMC free article] [PubMed]
16. Hess, D., and F. Winston. 2005. Evidence that Spt10 and Spt21 of S. cerevisiae play distinct roles in vivo and functionally interact with MCB-binding factor, SCB-binding factor, and Snf1. Genetics 170:87-94. [PMC free article] [PubMed]
17. Hughes, J. D., P. W. Estep, S. Tavazoie, and G. M. Church. 2000. Computational identification of cis-regulatory elements associated with groups of functionally related genes in Saccharomyces cerevisiae. J. Mol. Biol. 296:1205-1214. [PubMed]
18. Jones, E. W. 1991. Tackling the protease problem in Saccharomyces cerevisiae. Methods Enzymol. 194:428-453. [PubMed]
19. Kaplan, C. D., L. Laprade, and F. Winston. 2003. Transcription elongation factors repress transcription initiation from cryptic sites. Science 301:1096-1099. [PubMed]
20. Kawasaki, H., L. Schiltz, R. Chiu, K. Itakura, K. Taira, Y. Nakatani, and K. K. Yokoyama. 2000. ATF-2 has intrinsic histone acetyltransferase activity which is modulated by phosphorylation. Nature 405:195-200. [PubMed]
21. Kim, Y., and D. J. Clark. 2002. SWI/SNF-dependent long-range remodeling of yeast HIS3 chromatin. Proc. Natl. Acad. Sci. USA 99:15381-15386. [PMC free article] [PubMed]
22. Krogan, N. J., J. Dover, A. Wood, J. Schneider, J. Heidt, M. A. Boateng, O. W. Ryan, A. Golshani, M. Johnston, J. F. Greenblatt, and A. Shilatifard. 2003. The Paf1 complex is required for histone H3 methylation by COMPASS and Dot1p: linking transcriptional elongation to histone methylation. Mol. Cell 11:721-729. [PubMed]
23. Kuo, H. C., J. D. Moore, and J. E. Krebs. 2005. Histone H2A and Spt10 cooperate to regulate induction and autoregulation of the CUP1 metallothionein. J. Biol. Chem. 280:104-111. [PubMed]
24. Lee, T. I., H. C. Causton, F. C. Holstege, W. C. Shen, N. Hannett, E. G. Jennings, F. Winston, M. R. Green, and R. A. Young. 2000. Redundant roles for the TFIID and SAGA complexes in global transcription. Nature 405:701-704. [PubMed]
25. Luger, K., A. W. Mäder, R. K. Richmond, D. F. Sargent, and T. M. Richmond. 1997. Crystal structure of the nucleosome core particle at 2.8Å resolution. Nature 389:251-260. [PubMed]
26. McConnell, A. D., M. E. Gelbart, and T. Tsukiyama. 2004. Histone fold protein Dls1p is required for Isw2-dependent chromatin remodeling in vivo. Mol. Cell. Biol. 24:2605-2613. [PMC free article] [PubMed]
27. Natsoulis, G., C. Dollard, F. Winston, and J. D. Boeke. 1991. The products of the SPT10 and SPT21 genes of Saccharomyces cerevisiae increase the amplitude of transcriptional regulation at a large number of unlinked loci. New Biol. 3:1249-1259. [PubMed]
28. Natsoulis, G., F. Winston, and J. D. Boeke. 1994. The SPT10 and SPT21 genes of Saccharomyces cerevisiae. Genetics 136:93-105. [PMC free article] [PubMed]
29. Neuwald, A. F., and D. Landsman. 1997. GCN5-related histone N-acetyltransferases belong to a diverse superfamily that includes the yeast SPT10 protein. Trends Biochem. Sci. 22:154-155. [PubMed]
30. Norris, D., and M. A. Osley. 1987. The two gene pairs encoding H2A and H2B play different roles in the Saccharomyces cerevisiae life cycle. Mol. Cell. Biol. 7:3473-3481. [PMC free article] [PubMed]
31. Osley, M. A. 1991. The regulation of histone synthesis in the cell cycle. Annu. Rev. Biochem. 60:827-861. [PubMed]
32. Osley, M. A., J. Gould, S. Kim, M. Kane, and L. Hereford. 1986. Identification of sequences in a yeast histone promoter involved in periodic transcription. Cell 45:537-544. [PubMed]
33. Peterson, C. L. 2002. Chromatin remodeling enzymes: taming the machines. EMBO Rep. 3:319-322. [PMC free article] [PubMed]
34. Prelich, G., and F. Winston. 1993. Mutations that suppress the deletion of an upstream activating sequence in yeast: involvement of a protein kinase and histone H3 in repressing transcription in vivo. Genetics 135:665-676. [PMC free article] [PubMed]
35. Shen, C.-H., and D. J. Clark. 2001. DNA sequence plays a major role in determining nucleosome positions in yeast CUP1 chromatin. J. Biol. Chem. 276:35209-35216. [PubMed]
36. Shen, C.-H., B. P. Leblanc, J. A. Alfieri, and D. J. Clark. 2001. Remodeling of yeast CUP1 chromatin involves activator-dependent repositioning of nucleosomes over the entire gene and flanking sequences. Mol. Cell. Biol. 21:534-547. [PMC free article] [PubMed]
37. Shen, C.-H., B. P. Leblanc, C. Neal, R. Akhavan, and D. J. Clark. 2002. Targeted histone acetylation at the yeast CUP1 promoter requires the transcriptional activator, the TATA boxes and the putative histone acetylase encoded by SPT10. Mol. Cell. Biol. 22:6406-6416. [PMC free article] [PubMed]
38. Sherwood, P. W., and M. A. Osley. 1991. Histone regulatory (hir) mutations suppress δ insertion alleles in Saccharomyces cerevisiae. Genetics 128:729-738. [PMC free article] [PubMed]
39. Simpson, R. T., F. Thoma, and J. M. Brubaker. 1985. Chromatin reconstituted from tandemly repeated cloned DNA fragments and core histones: a model system for study of higher order structure. Cell 42:799-808. [PubMed]
40. Smith, M. M., and O. S. Andrésson. 1983. DNA sequences of yeast H3 and H4 histone genes from two non-allelic gene sets encode identical H3 and H4 proteins. J. Mol. Biol. 169:663-690. [PubMed]
41. Strahl, B. D., and C. D. Allis. 2000. The language of covalent histone modifications. Nature 403:41-45. [PubMed]
42. Thomas, J. O., and V. Furber. 1976. Yeast chromatin structure. FEBS Lett. 66:274-280. [PubMed]
43. Thompson, W., E. C. Rouchka, and C. E. Lawrence. 2003. Gibbs Recursive Sampler: finding transcription factor binding sites. Nucleic Acids Res. 31:3580-3585. [PMC free article] [PubMed]
44. Winston, F., and P. Sudarsanam. 1998. The SAGA of Spt proteins and transcriptional analysis in yeast: past, present and future. Cold Spring Harbor Symp. Quant. Biol. 63:553-561. [PubMed]
45. Wyrick, J. J., F. C. P. Holstege, E. G. Jennings, H. C. Causton, D. Shore, M. Grunstein, E. S. Lander, and R. A. Young. 1999. Chromosomal landscape of nucleosome-dependent gene expression and silencing in yeast. Nature 402:418-421. [PubMed]
46. Xu, F., K. Zhang, and M. Grunstein. 2005. Acetylation in histone H3 globular domain regulates gene expression in yeast. Cell 121:1-11. [PubMed]
47. Yamaguchi, Y., T. Narita, N. Inukai, T. Wada, and H. Handa. 2001. SPT genes: key players in the regulation of transcription, chromatin structure and other cellular processes. J. Biochem. 129:185-191. [PubMed]
48. Yamashita, I. 1993. Isolation and characterization of the SUD1 gene, which encodes a global repressor of core promoter activity in Saccharomyces cerevisiae. Mol. Gen. Genet. 241:616-626. [PubMed]

Articles from Molecular and Cellular Biology are provided here courtesy of American Society for Microbiology (ASM)
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...