MOLECULAR AND CELLULAR BIOLOGY, Sept. 2009, p. 4604–4611
Copyright © 2009, American Society for Microbiology. All Rights Reserved.
Vol. 29, No. 17
A Noncanonical Bromodomain in the AAA ATPase Protein Yta7
Directs Chromosomal Positioning and Barrier Chromatin Activity?
Angeline Gradolatto,1Sherri K. Smart,1† Stephanie Byrum,1† Lauren P. Blair,1†‡ Richard S. Rogers,2
Elizabeth A. Kolar,3Heather Lavender,1Signe K. Larson,1John D. Aitchison,2
Sean D. Taverna,3and Alan J. Tackett1*
University of Arkansas for Medical Sciences, Department of Biochemistry and Molecular Biology, Little Rock, Arkansas 722051;
Institute for Systems Biology, Seattle, Washington 981032; and Johns Hopkins School of Medicine, Baltimore, Maryland 212053
Received 4 February 2009/Returned for modification 7 March 2009/Accepted 21 June 2009
Saccharomyces cerevisiae Yta7 is a barrier active protein that modulates transcriptional states at the silent
mating locus, HMR. Additionally, Yta7 regulates histone gene transcription and has overlapping functions with
known histone chaperones. This study focused on deciphering the functional role of the noncanonical Yta7
bromodomain. By use of genetic and epistasis analyses, the Yta7 bromodomain was shown to be necessary for
barrier activity at HMR and to have overlapping functions with histone regulators (Asf1 and Spt16). Canonical
bromodomains can bind to acetylated lysines on histones; however, the Yta7 bromodomain showed an asso-
ciation with histones that was independent of posttranslational modification. Further investigation showed
that regions of Yta7 other than the bromodomain conferred histone association. Chromatin immunoprecipi-
tation-chip analyses revealed that the Yta7 bromodomain was not solely responsible for histone association but
was also necessary for proper chromosomal positioning of Yta7. This work demonstrates that the Yta7
bromodomain engages histones for certain cellular functions like barrier chromatin maintenance and partic-
ular Spt16/Asf1 cellular pathways of chromatin regulation.
Cellular processes such as gene transcription are governed
on one level by epigenetics. Chemical modifications occur on
chromosomal DNA and proteins to establish a given epige-
netic state (e.g., CpG DNA methylation and histone acetyla-
tion/methylation/phosphorylation). Histones are rich with
modifiable amino acids and thus can be found with particular
sets of posttranslational modifications (PTMs) (16). The par-
ticular mark or set of marks establishes regions of active (eu-
chromatin) and silent (heterochromatin) gene transcription.
Chromatin binding proteins can utilize a variety of domains to
recognize these modifications. One highly specialized domain
is the bromodomain, which binds specifically to acetylated ly-
sines and serves as a protein-protein interaction domain (10).
The bromodomain has been found in proteins ranging from
histone acetyltransferases to transcriptional activators. The re-
ported three-dimensional structures of bromodomains are
highly conserved, consisting of a four-helix bundle (?Z, ?A, ?B,
and ?C) that has interhelical ZA and BC loops (10). The loops
form a hydrophobic pocket that interacts with the acetylated
lysine, with the ZA loop likely determining the binding dynam-
ics. ZA and BC loops show high sequence diversity, but three
amino acids are conserved for acetyllysine recognition (in Sac-
charomyces cerevisiae histone acetyltransferase Gcn5, these are
Y364, Y406, and N407) (10, 11). Interestingly, the conserved
Asn is not found in human TIF1?—suggesting that this tran-
scription intermediate factor may not have acetyllysine speci-
ficity but may still serve as a non-PTM-dependent protein-
protein interaction domain (10).
The yeast tat-binding analog 7 (Yta7) protein falls into the
aforementioned category of proteins with potential noncanoni-
cal bromodomains, which may utilize the bromodomain as a
protein-engaging module in a PTM-independent manner.
Within the ZA loop, Yta7 has only two of five conserved amino
acids necessary for Gcn5 binding to acetylated lysine but does
contain two of three conserved amino acids needed for Gcn5
binding to the histone peptide backbone (11). This simple
homology analysis led us to hypothesize that Yta7 contains a
noncanonical bromodomain that modulates protein-protein
interactions in an acetyllysine-independent manner. We pos-
tulated that the protein-protein association mediated by the
Yta7 bromodomain (Yta7BD) directs a histone interaction
because affinity purification of Yta7 yields a strong interaction
with core histones (14). We also proposed that the Yta7BD is
noncanonical, as in vitro binding studies revealed an associa-
tion with the unmodified N-terminal tail of histone H3 (6). To
test these hypotheses, we performed a functional analysis of
the Yta7BD and further explored histone PTMs that may
enhance bromodomain binding.
Yta7 is a barrier active protein modulating transcriptional
states at the silent mating locus, HMR (8, 14). Deletion of the
YTA7 gene leads to silencing of genes near the HMR region,
likely due to a misregulation of barrier chromatin maintenance
(14). Further studies found that double deletion of the YTA7
gene with histone chaperones or chromatin remodeling pro-
teins impairs cell growth under stress conditions, suggesting
that Yta7 plays a role in the maintenance of chromatin struc-
ture and transcriptional regulation (6). We recently showed
that Yta7 binds histone proteins, associates with histone loci,
* Corresponding author. Mailing address: University of Arkansas
for Medical Sciences, Department of Biochemistry and Molecular
Biology, 4301 West Markham Street, Little Rock, AR 72205. Phone:
(501) 686-8152. Fax: (501) 686-8169. E-mail: firstname.lastname@example.org.
† These authors contributed equally.
‡ Present address: Yale University School of Medicine, Department
of Pathology, New Haven, CT 06520.
?Published ahead of print on 6 July 2009.
and acts as a repressor of histone gene transcription (6). In
addition to the bromodomain region, Yta7 contains one well-
conserved and a second putative AAA ATPase domain (8).
AAA ATPase domains contain Walker motifs for ATP hydro-
lysis as well as regions for protein-protein interactions (7).
Since Yta7 is implicated in barrier chromatin maintenance,
genetically overlaps with histone chaperone activities, and re-
presses histone transcription, we wanted to better understand
how the noncanonical bromodomain affects these cellular
functions. Our results not only better define how the bromo-
domain affects Yta7 activity but also shed light onto the role of
noncanonical bromodomains in modulating chromatin struc-
ture and function.
MATERIALS AND METHODS
Yeast strains. Yeast strains are listed in Table 1. Genomic tagging and domain
deletions with PrA or Myc9were done as reported previously (5, 6). To check for
expression of Yta7 and Yta7BD in deletion strains, equal amounts of cell lysates
from the strains indicated in Fig. 2B were subjected to Western blotting first for
the PrA tag (Dako P0450; 1:100 dilution) on Yta7 (or the internal PrA tag used
to delete the bromodomain) and second for general histone H3 (Abcam 1791;
1:5,000 dilution) as a loading control.
Epistasis analysis. For temperature sensitivity assays, cells were grown to
saturation at 25°C, 10-fold serially diluted, spotted on yeast extract-peptone-
dextrose plates, and incubated at increasing temperatures (25°C and 34°C). For
damage assays, cells were grown to saturation at 25°C, 10-fold serially diluted,
and spotted on yeast extract-peptone-dextrose plates containing hydroxyurea
(HU; 0.1 and 0.2 M).
Barrier activity assay. The YTA7 gene or the YTA7 bromodomain was deleted
in cells harboring a URA3 reporter in the transcriptionally expressed barrier
(ROY508) and repressed (ROY648) areas of HMR (3). Cells were 10-fold serially
diluted on plates with or without 5-fluoroacetic acid (5-FOA; 1 g/liter) and
incubated at 30°C for 3 days. Expression of the URA3 gene in the presence of
5-FOA leads to cell death.
High-resolution ChIP-chip. Chromatin immunoprecipitation (ChIP)-chip
analyses of YTA7-MYC and YTA7bd?-MYC cells were performed in quadrupli-
cate with dye swapping (13, 15). Full-genome S. cerevisiae microarrays were
utilized (4 ? 44K slides; Agilent). The ratio of immunopurified DNA signal to
whole-cell extract was considered significant beyond 2 standard deviations from
the mean array signal. Standard ChIP with real-time PCR readout was used to
validate selected ChIP-chip results as previously described (14).
Transcription assay. ROY508, ROY508 yta7::KAN, and ROY508 YTA7bd?::
MYC9cells were grown to mid-log phase, and pellets were frozen in liquid
nitrogen after being washed with cold water. Hot acidic phenol extraction of
RNA was performed, and reverse transcription was conducted with the Invitro-
gen Superscript first-strand kit (14). Transcription levels of the HTB1 and ACT1
genes were determined by real-time PCR (6).
Affinity purification of YTA7 truncations. Strains shown in Fig. 5A were made
by replacing segments of the YTA7 gene with a PrA (or Myc) tag or by inserting
the tag at the indicated positions (5). Affinity purifications were done as previ-
ously reported (14). The amount of cell lysate used in the truncation analysis was
normalized by Western blotting for the PrA tag. Histone- and Yta7-containing
gel bands were identified by tandem mass spectrometry (MS).
Yta7BD binding to histones. For histone binding studies, 30 ?g of recombinant
Yta7BD–glutathione S-transferase (GST) (or GST control) was incubated with
S. cerevisiae core histones (gift from Ines Pinto, University of Arkansas at Fay-
etteville). Core histones were purified in the presence of sodium butyrate to
prevent histone deacetylase activity (12, 14). Histones and recombinant proteins
were incubated in 500 ?l of binding buffer (20 mM HEPES, pH 7.9, 25%
glycerol, 100 mM NaCl, 1.5 mM MgCl2, 0.2 mM EDTA, 1% Triton X-100, 0.2
?g/?l bovine serum albumin, 1:100 Sigma yeast protease inhibitor cocktail) for
2 h at room temperature. Yta7BD-GST or GST alone was collected for 2 h at
room temperature with glutathione Sepharose 4B (Amersham Biosciences) and
washed four times with low-salt buffer (4 mM HEPES, pH 7.9, 10 mM NaCl).
Associated histones were eluted after the mixture was heated to 95°C for 10 min
in sodium dodecyl sulfate-polyacrylamide gel electrophoresis sample buffer, re-
solved with a 4 to 20% Tris-glycine gel, and visualized by Coomassie blue
staining. Histone-containing gel bands were chemically treated with d6-acetic
anhydride and subjected to trypsin digestion (2, 14).
Histone peptide binding. Recombinant fusions of GST with the PHD finger of
Yng1 and with the bromodomain of CBP were purified as reported previously
and confirmed with MS (15). Histone peptide binding studies with GST-Yng1
(PHD finger) and GST-CBP (bromodomain) were performed as reported pre-
viously with buffers (and washing) as described above for Yta7BD binding meth-
TABLE 1. S. cerevisiae strains used in this study
GenotypeBackgroundSource or reference
MAT? his3?1 leu2?0 lys2? ura3?0
MATa his3?1 leu2?0 met15?0 ura3?0
MATa ade2-1 ura3-1 his3-11,15 trp1-1 leu2-3,112 can1-100
Frederick Cross, Rockefeller
MAT? ade2 LYS URA3-HMR ppr1::HIS3
MAT? ade2 LYS HMR::URA3 ppr1::HIS3
MAT? yta7::KAN ade2 LYS URA3-HMR ppr1::HIS3
MAT? YTA7bd::MYC ade2 LYS URA3-HMR ppr1::HIS3
MAT? yta7::KAN ade2 LYS HMR::URA3 ppr1::HIS3
MAT? YTA7bd::MYC ade2 LYS HMR::URA3 ppr1::HIS3
MAT? yta7::KAN his3?1 leu2?0 lys2? ura3?0
MAT? hir2::KAN his3?1 leu2?0 lys2? ura3?0
MAT? hir2::KAN yta7::LEU2 his3?1 leu2?0 lys2? ura3?0
MAT? hir2::KAN YTA7bd::PrA his3?1 leu2?0 lys2? ura3?0
MAT? YTA7bd::PrA his3?1 leu2?0 lys2? ura3?0
MATa yta7::KAN ade2-1 ura3-1 his3-11,15 trp1-1 leu2-3,112 can1-100
MATa spt16-11 trp1 leu2 ura3 his7
MATa spt16-11 yta7::KAN ade2-1 ura3-1 his3-11,15 trp1-1 leu2-3,112 can1-100
MATa spt16-11 YTA7bd::PrA ade2-1 ura3-1 his3-11,15 trp1-1 leu2-3,112 can1-100
MATa asf1::KAN his3?1 leu2?0 met15?0 ura3?0
MATa asf1::KAN yta7::LEU2 his3?1 leu2?0 met15?0 ura3?0
MATa asf1::KAN YTA7bd::PrA his3?1 leu2?0 met15?0 ura3?0
MAT? YTA7bd::PrA his3?1 leu2?0 lys2? ura3?0
MATa yta7-1200-MYC9ade2-1 ura3-1 his3-11,15 trp1-1 leu2-3,112 can1-100
MATa yta7bd-MYC9ade2-1 ura3-1 his3-11,15 trp1-1 leu2-3,112 can1-100
MAT? YTA7-1200-PrA his3?1 leu2?0 lys2? ura3?0
aWT, wild type.
VOL. 29, 2009FUNCTIONAL ANALYSIS OF THE Yta7 BROMODOMAIN4605
MS. Mass spectra of the histone peptides were collected with a matrix-assisted
laser desorption ionization (MALDI)–prOTOF MS (PerkinElmerSciex). Pep-
tides were confirmed by tandem MS with a vMALDI-LTQ ion trap MS
(Thermo). m-over-z software was used to extract the monoisotopic peak areas
for the heavy and light versions of each histone peptide. Quantitative measure-
ments of acetylation were calculated from peak areas of fragment ions of samples
treated with d6-acetic anhydride (14). The d6-acetic anhydride treatment con-
verts unmodified lysines to triply deuterated acetyllysines (14). This allows one to
differentiate chemically acetylated lysines (?45 Da) from in vivo-acetylated ly-
sines (?42 Da) (14). Measurements of methylation were semiquantitative, as the
isotopic label was an acetyl group and therefore may have had slightly different
ionization properties. The percentage of methylation was obtained by calculating
the light peptide area relative to total light- plus heavy-peptide area. Enrichment
of acetylated or methylated peptides was evaluated by comparison to levels of the
same modification in the input sample of histones.
The bromodomain of Yta7 is required for barrier activity.
Genomic deletion of YTA7 results in the loss of barrier main-
tenance at the right- and left-hand barriers of HMR (8, 14). In
accordance, we investigated the requirement for the Yta7BD
in HMR barrier maintenance by using a URA3 gene silencing
assay (3, 6, 14). Two different cellular backgrounds lacking the
Yta7BD were used: ROY508, which contains a URA3 gene in
the transcriptionally active left-hand barrier of HMR, and
ROY648, which contains a URA3 gene in the silenced HMR
region (Fig. 1). ROY508 cells died in the presence of 5-FOA,
while ROY648 cells that carry the URA3 gene in the silenced
region grew normally (Fig. 1). In contrast, YTA7 gene deletion
or Yta7BD deletion conferred survival of ROY508 cells on
5-FOA plates. Deletions of YTA7 in the ROY648 background
did not affect growth, thus demonstrating that the silent region
of HMR was not perturbed. These results show that the
Yta7BD is necessary for maintenance of the left-hand barrier
around HMR, as bromodomain deletion resulted in spreading
of the silent transcriptional state and silencing of the ROY508
URA3 gene reporter.
The functional overlap of Yta7 and certain histone regula-
tors is bromodomain dependent. Previous epistasis analyses
revealed that double deletion of the YTA7 gene and certain
histone regulators leads to growth defects under stress condi-
tions. For example, cells harboring both a mutation in the
yFACT (facilitates chromatin transcription) component SPT16
(spt16-11) and a deletion of the YTA7 gene exhibit tempera-
ture sensitivity and a mild HU sensitivity (6). Double deletion
of known nucleosome assembly proteins (HIR1, HIR2, or ASF1
gene products) in yta7? cells leads to temperature, methyl
methanesulfonate, and/or HU sensitivity (6). We used the
same epistasis analysis to investigate the importance of the
The bromodomain of Yta7 was genomically replaced with a
PrA tag (YTA7bd?). Cells were serially diluted and grown at
elevated temperature (34°C) or under HU (0.1 M and 0.2 M)
stress conditions. YTA7bd? cells with hir2?, asf1?, or spt16-11
were subjected to the different treatments. The spt16-11 point
mutant of Spt16 was used because the gene deletion mutant is
inviable and this mutant shows growth defects under the con-
ditions utilized (4, 6). In hir2? cells exposed to high-tempera-
ture or HU treatment, deletion of the bromodomain had no
effect on cell growth relative to controls (not shown). In con-
trast, deletion of the Yta7BD with spt16-11 or asf1? showed an
intermediate growth defect in comparison with the single
(YTA7bd?, spt16-11, or asf1?) and double (spt16-11 yta7? or
asf1? yta7?) gene deletions. The synthetic defect with asf1?
YTA7bd? was mild but reproducible. Western blotting for the
PrA tag on Yta7 or Yta7BD showed that the observed effects
were not due to altered protein expression (Fig. 2B). Our
results show that the Yta7BD contributes to certain Yta7 ac-
tivities and that the bromodomain confers overlapping func-
tion with Spt16 and Asf1.
Deletion of the bromodomain genomically redistributes
Yta7. We have shown that the bromodomain of Yta7 associates
with histones via the N-terminal tail of histone H3 (6). Our
data presented above suggest that the bromodomain impacts
the function of Yta7 in barrier activity and exhibits overlapping
functions with histone chaperones (Fig. 1 and 2). Therefore,
we sought to better understand whether the bromodomain of
Yta7 localized the activity of Yta7 to certain regions of chro-
To determine the importance of the bromodomain for chro-
mosomal positioning of Yta7, genome binding of Yta7 with or
without the bromodomain was assessed using ChIP-chip anal-
yses (Fig. 3A). Deletion of the Yta7BD resulted in a significant
redistribution of Yta7 binding to more chromosomal sites (Fig.
3A). Only 104 binding sites (?15%) were conserved between
YTA7bd?-MYC and YTA7-MYC cells. Bromodomain deletion
led to a loss of binding in some sites, such as GLN4 (Fig. 3B),
or to a gain of binding such as at YCR095C, located next to the
HMR region (Fig. 3C). Gaining and losing binding suggest that
the bromodomain is not solely responsible for chromatin as-
sociation but may actually fine-tune Yta7 chromosomal posi-
tioning. Our previous work shows that Yta7 binds to histone
FIG. 1. The bromodomain of Yta7 is required for barrier activity.
The YTA7 gene or the bromodomain region of the gene was deleted in
strains harboring a URA3 reporter gene at the transcriptionally active
barrier (ROY508) or silent region (ROY648) of HMR. Cells were 10-
fold serially diluted on medium with or without 5-FOA. WT, wild type.
4606GRADOLATTO ET AL.MOL. CELL. BIOL.
loci to regulate S-phase transcriptional activation (6). In the
present ChIP-chip analyses, histone locus binding was not lost
upon deletion of the Yta7BD (Fig. 3D). We observed signifi-
cant association of both Yta7 and Yta7BD? at the HHT1/
HHF1 and HTB1/HTA1 loci. Binding distributions shown in
Fig. 3B, C, and D were confirmed by conventional ChIP (Fig.
3E). To functionally verify Yta7BD? binding at the HTB1/
HTA1 locus, we used real-time PCR to measure the transcrip-
tion level of HTB1 relative to that of the control gene ACT1
(Fig. 3F). We observed that YTA7 gene deletion caused an
increase in HTB1 transcription, while deletion of the bromo-
domain mimicked wild-type transcription. These results sug-
gest that the Yta7BD is not required for physical retention or
activity at histone loci.
The Yta7BD binds to histones in a PTM-independent man-
ner. Functional and localization data suggest that the bromo-
domain of Yta7 is necessary for particular cellular functions
(Fig. 1 to 3). To evaluate whether binding of the Yta7BD to
histones is dependent on a particular set of histone PTMs,
recombinant Yta7BD-GST fusion protein or GST alone was
incubated with yeast core histones and affinity isolated on gluta-
thione resin (Fig. 4A). Histones copurifying with Yta7BD-GST
or GST alone were resolved on a denaturing gel and visualized
with Coomassie blue staining. Compared to the GST control,
all four histones were strongly enriched with Yta7BD (Fig.
4A). Under the conditions of the assay, histone octamers are
likely dissociating; thus, the Yta7BD may have affinity for
H3-H4 pairs, H2A-H2B pairs, and/or individual histones. In
previous work, we observed a Yta7BD enrichment of Tetrahy-
mena thermophila histone H3 and H2B (6). The conditions
used in the experiment shown in Fig. 4A have been optimized
relative to our initial work, which is evident in our now-mini-
mal binding of histones to the GST control (and use of histones
from S. cerevisiae). Thus, under the conditions used, Yta7BD
was found to purify the core histones.
To determine if the buffer composition and resin washing
procedure used in this approach would preserve protein-his-
tone interactions, we performed histone peptide binding stud-
ies with two protein domains that bind histone PTMs (Fig. 4B).
Recombinant GST fusions were made with the PHD finger
from the yeast Yng1 protein and the bromodomain of human
CBP. The Yng1 PHD finger shows a relatively strong and
specific interaction with H3K4me3 (KD[equilibrium dissocia-
tion constant] ? 9 ?M) (15). Canonical bromodomains show a
weaker binding to the acetyllysine moieties on histones (KD?
100 to 400 ?M) (1, 17). To test binding, the biotinylated his-
tone peptides listed in Fig. 4B were coupled to streptavidin
Dynabeads and incubated with the respective protein (6). Un-
der the binding and washing conditions used in this work for
Yta7BD, both the higher-affinity interaction of the Yng1 PHD
finger with H3K4me3 and the lower-affinity interaction of the
CBP bromodomain with a tetra-acetylated version of the H4
N-terminal tail were detected (Fig. 4B). These results indicate
that any histone PTM enrichment with the Yta7BD pulldown
should be detectable, since the relative association of histones
with Yta7BD is greater than that with the CBP canonical
bromodomain (Fig. 4A) (6).
Gel bands corresponding to histones in Fig. 4A were excised,
treated with d6-acetic anhydride for quantitative analysis of
histone acetylation, and subjected to in-gel trypsin digestion (6,
14). This d6-acetic anhydride treatment will heavy acetylate
(45-Da) unmodified lysines while not affecting in vivo acetyla-
tions (42 Da), thereby providing an isotopic label for quanti-
tation. Resulting tryptic histone peptides were subjected to
quantitative MALDI-MS and MALDI-tandem MS analysis (6,
14). MS analysis identified all four histones, but posttransla-
tionally modified peptides were identified only for H3/H2B/
H4. As an example, the MALDI mass spectra of the histone
FIG. 2. The functional overlap of Yta7 and certain histone regula-
tors is bromodomain dependent. (A) To assay temperature sensitivity,
indicated cells were 10-fold serially diluted and incubated at 25°C or
34°C for the indicated number of days (d). To assay HU sensitivity,
cells were 10-fold serially diluted and incubated at 25°C for the indi-
cated number of days. (B) Deletion/mutation of the Yta7BD, Asf1, or
Spt16 does not affect protein levels of Yta7 or Yta7 with a bromodo-
main deletion. Western blotting was used to detect the PrA tag on
Yta7 or the internal PrA tag used to delete the Yta7BD in the indi-
cated strains. Western blotting for histone H3 served as a loading
VOL. 29, 2009FUNCTIONAL ANALYSIS OF THE Yta7 BROMODOMAIN 4607
H4 4-17 peptide is shown for the input and affinity-purified
histone (Fig. 4C). The MS analysis of Yta7BD-associated his-
tones relative to input histones did not reveal any enriched
acetylation or methylation (Table 2), suggesting that histone
association was PTM independent for the detected histone
Yta7 contains histone binding regions distinct from the bro-
modomain. The ChIP-chip studies in Fig. 3 suggest that Yta7
can maintain chromatin association without the bromodomain.
To determine other regions of Yta7 that facilitate histone
association, we constructed a series of truncations of or inser-
tions in the YTA7 gene (Fig. 5A). Each of these Yta7 proteins
was affinity purified via the PrA tag, and copurifying histones
were identified by MS (Fig. 5B). Interestingly, the two smallest
constructs (1 to 400 and 1 to 650) copurified with the largest
amount of histones (Fig. 5B). Each construct longer than 650
amino acids exhibited histone binding comparable to that of
full-length Yta7. In concurrence with those results, insertion of
the PrA tag within the first 400 amino acids of the gene (at
position 106) greatly impaired histone purification, suggesting
that there is at least one additional histone binding region in
the N-terminal 400 amino acids of the protein. The largely
unstructured first 200 amino acids of Yta7 are 35% acidic
amino acids with a pI of 4.3, while the remaining portion of the
protein is 15% acidic amino acids with a pI of 5.7. This suggests
that the acidic N-terminal region of Yta7 may have electro-
static interactions with charged, unmodified lysine and arginine
residues on histones. One possibility is that the N-terminal
region and the bromodomain function synergistically to pro-
vide histone binding and proper in vivo localization.
FIG. 3. Deletion of the bromodomain genomically redistributes Yta7. (A) ChIP-chip analysis of Yta7-Myc and Yta7BD?-Myc. A Venn
diagram shows a redistribution of Yta7 binding with deletion of the bromodomain. Binding sites from the microarray are reported at the 95%
confidence level. (B) Yta7 binding at GLN4 was lost with bromodomain deletion. Binding is reported as the log2of the immunopurified signal to
whole-cell lysate array signals. The asterisk indicates significant binding (95% confidence level). (C) Bromodomain deletion resulted in detectable
Yta7 binding at YCR095C. The asterisk indicates significant binding (95% confidence level). (D) Deletion of the bromodomain did not affect Yta7
association at the HTB1-HTA1 histone locus. The asterisks indicate significant binding (95% confidence level). (E) Conventional ChIP for the Myc
tag on Yta7 and Yta7BD? confirms the redistribution shown in panels B, C, and D. Error bars are the standard errors from triplicate analyses.
(F) Bromodomain deletion does not affect HTB1 transcription. Transcript levels of HTB1 relative to ACT1 in asynchronous cells harboring
full-length Yta7 or the bromodomain deletion. Error bars are the standard deviations. Asterisks show significant differences determined by
one-tailed testing (P ? 0.001). WT, wild type.
4608GRADOLATTO ET AL.MOL. CELL. BIOL.
The Yta7 protein has reported activities in (i) the mainte-
nance of barrier chromatin surrounding HMR and (ii) path-
ways involving histone-regulating proteins (like Spt16, Hir1/2,
and Asf1) (6, 8). The functionality of the Yta7BD was tested
for these various activities. Deletion of the bromodomain mim-
icked deletion of the entire YTA7 gene for barrier maintenance
activity at HMR (Fig. 1). Since an in vivo effect was observed
upon deletion of a histone binding module, one would antici-
pate that Yta7 association at HMR had been perturbed. In
contrast, upon removal of the bromodomain, Yta7 showed
significant association with the barrier relative to that of wild-
type Yta7 (Fig. 3C). This suggests that the barrier maintenance
activity of Yta7 may be more of a transient association that is
attenuated by the bromodomain. In support, removal of the
bromodomain leaves the N-terminal acidic region of Yta7 to
modulate the histone association, which showed increased hi-
stone binding relative to that observed with the bromodomain
(Fig. 5). In accordance with these data, removal of the Yta7BD
favors Yta7 association at the HMR barrier, which in turn
FIG. 4. The Yta7BD binds to histones in a PTM-independent manner. (A) Recombinant Yta7BD-GST or GST alone was incubated with
purified histones and isolated on glutathione resin. Associated histones were resolved on Tris-glycine gels (4 to 20%) and visualized by Coomassie
blue staining. (B) Yng1 PHD finger and the CBP bromodomain show specific binding under the Yta7BD binding assay conditions. Recombinant
GST, GST-Yng1 PHD finger, or GST-CBP bromodomain was incubated with the indicated biotinylated histone peptides immobilized on
streptavidin-coated Dynabeads. Bound protein was resolved by denaturing gel electrophoresis and visualized by Coomassie blue staining.
(C) Quantitative analysis of histone acetylation copurifying with Yta7BD-GST. Copurifying and input histones shown in panel A were treated with
d6-acetic anhydride and digested with trypsin. Shown are overlaid MALDI spectra of the N-terminal tail of histone H4 corresponding to peptide
4–17 containing K5, K8, K12, and K16 that have been chemically (45 Da) or in vivo (42 Da) acetylated. No significant difference in histone
acetylation was observed. The complete set of results is in Table 2.
TABLE 2. Yta7BD does not enrich for specific histone PTMs
H2BK12, K15 ac1
H4K5, K8, K12, K16
aRatio of Yta7BD enriched to input histone PTMs.
bNR, very low to nonrecordable levels in both input and enriched.
VOL. 29, 2009 FUNCTIONAL ANALYSIS OF THE Yta7 BROMODOMAIN4609
perturbs the maintenance of the barrier (Fig. 1 and 3C). One
explanation of these observations is that Yta7 functions as a
chromatin-organizing protein that serves a transient function in
establishing or maintaining the barrier chromatin. In the case of
a bromodomain deletion, the prolonged Yta7 activity serves as a
bottleneck along the pathway for barrier maintenance.
In a similar situation of transient association, Yta7 shows
binding at the HTB1-HTA1 histone locus prior to gene tran-
scription but dissociates during S-phase transcription of HBT1
(A. Gradolatto and A. J. Tackett, unpublished observations).
Thus, Yta7 binding and activity can be transient during the cell
cycle. The transient association of Yta7 at the histone genes
does affect gene transcription (Fig. 3F); thus, the normal dos-
age of histones is altered in the S phase and could have impli-
cations for the repackaging of chromatin during replication.
Additional links of Yta7 to chromatin packaging were found
with our epistasis analyses (Fig. 2). Deletion of the Yta7BD
with spt16-11 or asf1? showed an intermediate effect relative to
that of full deletion of the YTA7 gene and controls. These
synthetic effects link the Yta7BD to pathways of chromatin
assembly and disassembly for gene transcription.
To test whether the bromodomain targets Yta7 to particular
chromatin sites, a ChIP-chip analysis was performed. Removal
of the bromodomain did perturb wild-type binding, as
Yta7BD? was largely redistributed across the chromosomes
(Fig. 3A). We previously reported the avid association of Yta7
with histone loci (6). The histone locus association was pre-
served upon deletion of the bromodomain, and the bound
Yta7BD? protein was functionally active for the regulation of
histone gene transcription (Fig. 3D and F). The redistribution
supports the finding that histone association is not solely con-
ferred through the Yta7BD but additionally via the N-terminal
acidic region (Fig. 5). Chromatin association through an acidic
(?50%) C-terminal half of a protein has also been seen for the
histone regulator Asf1 (9). Furthermore, the redistribution ef-
fect correlates with the mixed phenotypes observed with our ep-
istasis analyses. These data support the idea that redistribution of
Yta7 upon deletion of the bromodomain can alter particular
activities of Yta7; however, retention of Yta7 through bromodo-
main-independent mechanisms occurs.
In previous work, we were unable to uncover a histone PTM
that would enhance Yta7BD association with histones by using an
in vitro system with synthetic histone peptides but rather found
that additional PTMs perturbed the Yta7BD-histone H3 interac-
tion (6). These analyses were expanded upon by utilizing purified
yeast histones (Fig. 4A and Table 2). These studies also did not
reveal a histone acetylation that was enriched with Yta7. It is
possible that a given acetylation precluded our MS analysis; how-
ever, the femtomole-level sensitivity of the MS combined with the
avid enrichment of histones with Yta7BD would be expected to
identify any observable acetylated peptide. These results support
the concept that the Yta7BD serves as a protein-protein interac-
tion domain for PTM-independent histone association.
The data presented provide evidence that the Yta7 protein
contains a noncanonical bromodomain that serves primarily as
a histone recognition domain. The function of the bromodo-
main appears to be for chromatin localization in some cases
and attenuation of binding in others. Specifically, the ability of
Yta7 to engage histones via the bromodomain is needed for
certain cellular functions like barrier maintenance at HMR and
particular Spt16/Asf1 cellular pathways of chromatin regula-
tion. Future work will focus on understanding the molecular
details of how the bromodomain and the AAA ATPase do-
main confer in vivo activity.
The following National Institutes of Health grants supported this
work: R01DA025755, P20RR015569, P20RR016460, R01GM075152,
FIG. 5. Yta7 contains histone binding regions distinct from the bromodomain. (A) Schematic of Yta7 constructs used to determine histone
binding. The shapes represent: PrA tag (circles), AAA ATPase domain (filled squares), putative AAA ATPase domain (striped squares), and
bromodomain (open squares). Constructs were made genomically. (B) Affinity purification of the constructs in panel A. Yta7 constructs were
purified on immunoglobulin G-coated Dynabeads under conditions that preserve in vivo protein associations. PrA-tagged Yta7 and associated
proteins were resolved by denaturing gel electrophoresis and visualized by Coomassie blue staining. Yta7 constructs and histones were identified
4610 GRADOLATTO ET AL.MOL. CELL. BIOL.
We acknowledge support from the Arkansas Biosciences Institute,
David Terrano for critical reading, Ana Raman for technical assis-
tance, Haitao Li and Dinshaw Patel for the GST-Yng1 (PHD) expres-
sion construct, and Phillip A. Cole and Blair Dancy for Escherichia coli
containing induced GST-CBP (bromodomain).
1. Dhalluin, C., J. E. Carlson, L. Zeng, C. He, A. K. Aggarwal, and M. M. Zhou.
1999. Structure and ligand of a histone acetyltransferase bromodomain.
2. Dilworth, D. J., A. J. Tackett, R. S. Rogers, E. C. Yi, R. H. Christmas, J. J.
Smith, A. F. Siegel, B. T. Chait, R. W. Wozniak, and J. D. Aitchison. 2005.
The mobile nucleoporin Nup2p and chromatin-bound Prp20p function in
endogenous NPC-mediated transcriptional control. J. Cell Biol. 171:955–
3. Donze, D., C. R. Adams, J. Rine, and R. T. Kamakaka. 1999. The boundaries
of the silenced HMR domain in Saccharomyces cerevisiae. Genes Dev.
4. Formosa, T., S. Ruone, M. D. Adams, A. E. Olsen, P. Eriksson, Y. Yu, A. R.
Rhoades, P. D. Kaufman, and D. J. Stillman. 2002. Defects in SPT16 or
POB3 (yFACT) in Saccharomyces cerevisiae cause dependence on the Hir/
Hpc pathway: polymerase passage may degrade chromatin structure. Genet-
5. Gauss, R., M. Trautwein, T. Sommer, and A. Spang. 2005. New modules for
the repeated internal and N-terminal epitope tagging of genes in Saccharo-
myces cerevisiae. Yeast 22:1–12.
6. Gradolatto, A., R. S. Rogers, H. Lavender, S. D. Taverna, C. D. Allis, J. D.
Aitchison, and A. J. Tackett. 2008. Saccharomyces cerevisiae Yta7 regulates
histone gene expression. Genetics 179:291–304.
7. Hanson, P. I., and S. W. Whiteheart. 2005. AAA? proteins: have engine, will
work. Nat. Rev. Mol. Cell Biol. 6:519–529.
8. Jambunathan, N., A. W. Martinez, E. C. Robert, N. B. Agochukwu, M. E.
Ibos, S. L. Dugas, and D. Donze. 2005. Multiple bromodomain genes are
involved in restricting the spread of heterochromatic silencing at the Sac-
charomyces cerevisiae HMR-tRNA boundary. Genetics 171:913–922.
9. Mousson, F., F. Ochsenbein, and C. Mann. 2007. The histone chaperone
Asf1 at the crossroads of chromatin and DNA checkpoint pathways. Chro-
10. Mujtaba, S., L. Zeng, and M. M. Zhou. 2007. Structure and acetyl-lysine
recognition of the bromodomain. Oncogene 26:5521–5527.
11. Owen, D. J., P. Ornaghi, J. C. Yang, N. Lowe, P. R. Evans, P. Ballario, D.
Neuhaus, P. Filetici, and A. A. Travers. 2000. The structural basis for the
recognition of acetylated histone H4 by the bromodomain of histone acetyl-
transferase Gcn5p. EMBO J. 19:6141–6149.
12. Pinto, I., and F. Winston. 2000. Histone H2A is required for normal cen-
tromere function in Saccharomyces cerevisiae. EMBO J. 19:1598–1612.
13. Ren, B., F. Robert, J. J. Wyrick, O. Aparicio, E. G. Jennings, I. Simon, J.
Zeitlinger, J. Schreiber, N. Hannett, E. Kanin, T. L. Volkert, C. J. Wilson,
S. P. Bell, and R. A. Young. 2000. Genome-wide location and function of
DNA binding proteins. Science 290:2306–2309.
14. Tackett, A. J., D. J. Dilworth, M. J. Davey, M. O’Donnell, J. D. Aitchison,
M. P. Rout, and B. T. Chait. 2005. Proteomic and genomic characterization
of chromatin complexes at a boundary. J. Cell Biol. 169:35–47.
15. Taverna, S. D., S. Ilin, R. S. Rogers, J. C. Tanny, H. Lavender, H. Li, L.
Baker, J. Boyle, L. P. Blair, B. T. Chait, D. J. Patel, J. D. Aitchison, A. J.
Tackett, and C. D. Allis. 2006. Yng1 PHD finger binding to histone H3
trimethylated at lysine 4 promotes NuA3 HAT activity at lysine 14 of H3 and
transcription at a subset of targeted ORFs. Mol. Cell 24:785–796.
16. Taverna, S. D., B. M. Ueberheide, Y. Liu, A. J. Tackett, R. L. Diaz, J.
Shabanowitz, B. T. Chait, D. F. Hunt, and C. D. Allis. 2007. Long-distance
combinatorial linkage between methylation and acetylation on histone H3 N
termini. Proc. Natl. Acad. Sci. USA 104:2086–2091.
17. Zeng, L., Q. Zhang, G. Gerona-Navarro, N. Moshkina, and M. M. Zhou.
2008. Structural basis of site-specific histone recognition by the bromodo-
mains of human coactivators PCAF and CBP/p300. Structure 16:643–652.
VOL. 29, 2009 FUNCTIONAL ANALYSIS OF THE Yta7 BROMODOMAIN4611