MOLECULAR AND CELLULAR BIOLOGY, May 2009, p. 2346–2358
Copyright © 2009, American Society for Microbiology. All Rights Reserved.
Vol. 29, No. 9
Role of the Histone Variant H2A.Z/Htz1p in TBP Recruitment,
Chromatin Dynamics, and Regulated Expression of
Yakun Wan,1Ramsey A. Saleem,1Alexander V. Ratushny,1Oriol Roda,1Jennifer J. Smith,1
Chan-Hsien Lin,1,2Jung-Hsien Chiang,1,2and John D. Aitchison1*
Institute for Systems Biology, Seattle, Washington 98103,1and Department of Computer Science and Information Engineering,
National Cheng Kung University, Tainan, Taiwan2
Received 5 August 2009/Returned for modification 9 September 2008/Accepted 21 January 2009
The histone variant H2A.Z (Htz1p) has been implicated in transcriptional regulation in numerous organ-
isms, including Saccharomyces cerevisiae. Genome-wide transcriptome profiling and chromatin immunopre-
cipitation studies identified a role for Htz1p in the rapid and robust activation of many oleate-responsive genes
encoding peroxisomal proteins, in particular POT1, POX1, FOX2, and CTA1. The Swr1p-, Gcn5p-, and Chz1p-
dependent association of Htz1p with these promoters in their repressed states appears to establish an
epigenetic marker for the rapid and strong expression of these highly inducible promoters. Isw2p also plays a
role in establishing the nucleosome state of these promoters and associates stably in the absence of Htz1p. An
analysis of the nucleosome dynamics and Htz1p association with these promoters suggests a complex mech-
anism in which Htz1p-containing nucleosomes at fatty acid-responsive promoters are disassembled upon
initial exposure to oleic acid leading to the loss of Htz1p from the promoter. These nucleosomes reassemble at
later stages of gene expression. While these new nucleosomes do not incorporate Htz1p, the initial presence of
Htz1p appears to mark the promoter for sustained gene expression and the recruitment of TATA-binding
The organization of DNA into chromatin provides cells with
a key regulatory mechanism for gene expression by limiting the
access of the genome to the transcriptional machinery. The
nucleosome represents a basic structural unit of chromatin and
posttranslational modifications of histones serve as signals to
define active, repressed, or inert chromatin states. In addition,
chromatin states and gene expression can be influenced by the
dynamics of histones and their nonallelic variants. Indeed, the
exchange of canonical histones for histone variants appears to
be a key mechanism by which the transcriptional machinery
overcomes the restricted access imposed by nucleosome posi-
tioning (1). Of the many classes of histone variants discovered,
the Z variant of H2A is perhaps the best characterized. H2A.Z
differs from the canonical H2A histone in both the length and
sequence of the C terminus (37) and is conserved from yeast to
mammals (15). Early studies with H2A.Z in Tetrahymena spp.
showed that H2A.Z incorporation is linked with transcription-
ally active chromatin (35). The Saccharomyces cerevisiae ortho-
logue of H2A.Z is called Htz1p and is encoded by HTZ1.
Although HTZ1 is not an essential gene under standard labo-
ratory growth conditions, Htz1p is implicated in transcriptional
regulation. Global chromatin studies have revealed that Htz1p
preferentially associates with the two nucleosomes flanking the
nucleosome free region of promoters (12, 18, 26, 41), and this
association is inversely proportional to transcription rates
Studies of the role of Htz1p in transcriptional regulation at
specific promoters, such as those of GAL1 and PHO5 (1, 30)
indicate that the presence of Htz1p at promoters is dynamic;
Htz1p is bound in their repressed states but dissociates during
the activation process. Accordingly, it is proposed that nucleo-
somes containing Htz1p are poised to undergo nucleosome
displacement, allowing for rapid transcriptional responses
The yeast S. cerevisiae is an excellent model for understand-
ing the mechanisms of cellular responses to induced perturba-
tions. Upon the exposure of yeast to fatty acids, such as oleate,
cells respond by dramatically altering their gene expression
patterns, inducing genes required for peroxisomal ?-oxidation
and peroxisome biogenesis (32). Genetic screens to identify
proteins specifically required for efficient fatty acid metabolism
in S. cerevisiae (34) identified metabolic enzymes, proteins re-
quired for the biogenesis of the organelle, signaling proteins
and transcriptional regulators, and chromatin modifiers.
Among this latter class of proteins, this approach identified
genes encoding Htz1p, RNA polymerase II, mediator subunits,
and components of chromatin remodeling complexes. We thus
seek to understand the nature of how chromatin is regulated
and remodeled in response to exposure to fatty acids and the
specific role Htz1p plays in these regulation/remodeling pro-
In this report, transcriptomes of wild-type (WT) and htz1?
strains were compared during exposure to oleic acid. While the
loss of Htz1p reduced the expression of many genes, genes
involved in the fatty acid response were particularly sensitive.
A model is proposed in which Htz1p-containing nucleosomes
at fatty acid-responsive promoters are disassembled upon ini-
* Corresponding author. Mailing address: Institute for Systems
Biology, 1441 N 34th St., Seattle, WA 98103. Phone: (206) 732-1344.
Fax: (206) 732-1299. E-mail: email@example.com.
?Published ahead of print on 9 March 2009.
tial exposure to oleic acid, leading to the loss of Htz1p from the
promoter. These nucleosomes reassemble at later stages of
gene expression. While these nucleosomes do not incorporate
Htz1p, the initial presence of Htz1p appears to mark the pro-
moter for sustained gene expression and the recruitment of
TATA-binding protein (TBP).
MATERIALS AND METHODS
Strains and growth conditions. All yeast strains used in this study are indicated
in Table 1. Haploid strains with myc-tagged genes were made by genomically
tagging target genes with the sequence encoding 13 copies of the c-myc epitope
from pFA6a-13MYC (20) by homologous recombination into BY4742 (WT)
using a previously described PCR-based procedure (2). The strains were verified
by PCR analysis of the tagged gene loci and Western blot analysis of the fusion
proteins. An examination of the growth characteristics of each strain suggests
that the chimeras did not alter protein function. For all experiments, the control
strains were otherwise isogenic to the test strains. The strains were cultured at
30°C in the following media: YPD (1% yeast extract, 2% peptone, 2% glucose)
and SCIM (0.17% yeast nitrogen base without amino acids and ammonium
sulfate, 0.5% yeast extract, 0.5% peptone, 0.079% complete supplement mixture,
0.5% ammonium sulfate) containing 0.5% Tween 40 (wt/vol) and 0.2% (wt/vol)
RNA preparation and microarray analysis. Yeast cultures were grown at 30°C
to a density of ?1 ? 107cells/ml. The cells were collected and immediately
frozen in liquid nitrogen. Total RNA was isolated by hot acid phenol extraction.
The total RNA was treated with RNase-free DNase I and purified with a Qiagen
RNeasy kit. Microarray labeling and hybridization reactions were performed as
previously described (7). Two color microarrays, comparing RNA from the
experimental conditions (WT and htz1? cells grown in oleate [SCIM] for 6 h) to
RNA from the control WT cells grown in glucose-containing medium (YPD),
were performed using Agilent whole-genome S. cerevisiae arrays. All experiments
were performed with duplicate experimental and duplicate technical replicates of
each condition, and the log10of the average mRNA abundance ratios are re-
ported. Differentially expressed genes were identified by maximum-likelihood
analysis (? ? 100) (14, 32) and significantly affected genes in the mutants were
identified by a change in expression of twofold or more compared to the expres-
sion in the relevant WT strains.
For the quantitative reverse transcription-PCR (qRT-PCR), total RNA was
directly reverse transcribed using the First Strand cDNA synthesis kit from
Fermentas (catalog no. K1611). cDNAs were treated by RNase H and diluted
1/100 for the qPCR. The RT-PCR was done using a 7900HT fast real-time PCR
system and a DyNAmo Flash SYBR green qPCR kit (F-415L; NEB) with gene-
specific oligonucleotides. mRNA levels were normalized relative to ACT1
mRNA levels from three independent RT-PCR analyses. A list of primers used
for the qRT-PCRs is available on request.
ChIP and real-time PCR. For each chromatin immunoprecipitation (ChIP)
experiment, yeast strains were first grown in glucose medium (YPD) to a density
of ?1 ? 107cells/ml and then transferred to oleate medium (SCIM) for the times
indicated in each figure. The ChIP experiments were performed as described in
reference 33 with the following modifications. For the hemagglutinin (HA)-
Htz1p ChIP, cells were crossed-linked with 1% formaldehyde for 45 min at room
temperature. Two micrograms of anti-HA antibody (12CA5) was prebound to 50
?l of pan-mouse immunoglobulin G Dynabeads (Dynal Biotech) and then incu-
bated with 1 mg (protein) of supernatant from the sheared chromatin overnight
at 4°C. The TBP (Spt15p-Myc) ChIP was performed as described in reference 33.
Cells were cross-linked with 1% formaldehyde for 2 h at room temperature. Two
microliters of anti-Myc antibody (9E11; Abcam) was prebound to 50 ?l of
pan-mouse immunoglobulin G Dynabeads and then incubated with 1 mg (pro-
tein) of supernatant from sheared chromatin overnight at 4°C.
All ChIP experiments were performed in triplicate. The purified ChIP samples
were used in the qPCR analysis. Real-time qPCR was performed by using an
iCycler instrument (ABI 7900) and a DyNAmo Flash SYBR green qPCR kit.
The average of the results of three independent replicates is reported as the
relative amplification of each target of interest compared to a normalization
control amplicon, within the nonpromoter IGRi YMR325W. The primer se-
quences are available on request. The occupancy level was determined by divid-
ing the relative abundance of an experimental target by the relative abundance
of a control target. This ratio represents the enrichment of ChIP DNA over the
input DNA for a specific target versus the control target.
FACS analysis. The fluorescence-activated cell sorter (FACS) analysis proce-
dures were performed as previously described (27). The fluorescence intensities
of individual cells were measured using a FACSCalibur flow cytometer (BD
Biosciences). The data analysis was performed using WinMDI 2.8 (available
Nucleosome scanning assay (NuSA). A total of 200 ml of cells at an optical
density at 600 nm of 1.0 in glucose medium or after transfer to oleate-containing
medium prior to treatment with 1% formaldehyde for 20 min, followed by a
5-min incubation in 125 mM glycine. The cell permeabilization, micrococcal
nuclease digestion, protein degradation, and DNA purification steps were per-
formed as described previously (38). The DNA samples were then treated with
RNase A and analyzed in a 2% agarose gel to quantify the nucleosomal content.
The bands corresponding to mononucleosomal DNA were extracted using a
TABLE 1. Strains and plasmids used in this study
Strain or plasmid Genotype or description
MATa his3?1 leu2?0 met15?0 ura3?0
MAT? his3?1 leu2?0 met15?0 ura3?0
MAT? his3?1 leu2?0 met15?0 ura3?0 htz1::kanMX4
MAT? his3?1 leu2?0 met15?0 ura3?0 POT1-GFP::natMX
MAT? his3?1 leu2?0 met15?0 ura3?0 POT1-GFP::natMX pRS416
MAT? his3?1 leu2?0 met15?0 ura3?0 htz1::kanMX4 POT1-GFP::natMX
MAT? his3?1 leu2?0 met15?0 ura3?0 htz1::kanMX4 POT1-GFP::natMX pRS416
MAT? his3?1 leu2?0 met15?0 ura3?0 htz1::kanMX4 POT1-GFP::natMX pCM305
MAT? his3?1 leu2?0 met15?0 ura3?0 htz1::kanMX4 POT1-GFP::natMX pCM330
MAT? his3?1 leu2?0 met15?0 ura3?0 htz1::kanMX4 POT1-GFP::natMX pCM314
MAT? leu2?0 ura3?0 HA-HTZ1
MAT? leu2?0 ura3?0 HA-HTZ chz1::kanMX4
MAT? leu2?0 ura3?0 HA-HTZ gcn5::kanMX4
MAT? leu2?0 ura3?0 HA-HTZ swr1::kanMX4
MAT? his3?1 leu2?0 met15?0 ura3?0 SPT15-13MYC::kanMX4
MAT? his3?1 leu2?0 met15?0 ura3?0 SPT15-13MYC::kanMX4 htz1::hphMX
CEN6-ARS4 URA3 HA-htz1K14A
CEN6-ARS4 URA3 HA-htz1K14R
CEN6-ARS4 URA3 HA-HTZ1
VOL. 29, 2009 CONTROL OF GENE INDUCTION BY Htz1p2347
Qiagen gel extraction kit. A qPCR analysis of the digested DNA was performed.
A list of the qPCR primers is available on request; they cover the promoter
regions of POT1, CTA1, POX1, and FOX2 with overlapping amplicons averaging
100 bp in size. To define nucleosome occupancy, the protection value of each
amplicon was normalized to the CEN3 values as described previously (6). The
N?1 nucleosome refers to the first nucleosome downstream of the transcription
start site, which is located at the open reading frame regions. The N?1 nucleo-
some refers to the first nucleosome upstream of the transcription start site, which
is located at the promoter regions.
Htz1p is required for the transcriptional activation of a
subset of oleic acid-responsive genes. Transcriptome profiling
was used to obtain a global understanding of how Htz1p con-
tributes to gene expression in response to external stimuli. To
do so, we focused on gene induction upon the shift from
glucose to oleic acid growth conditions. This condition was
chosen because we and others have previously shown that this
transition leads to dramatic alterations in the gene expression
patterns (16, 32, 33), because genes involved in fatty acid me-
tabolism are significantly induced under these conditions, and
because it has been shown that S. cerevisiae htz1? strains have
a specific growth defect when grown on fatty acids (19, 34). In
accordance with previous genome-wide analyses of oleate re-
sponses (16, 32, 33), a large portion of the genome responds to
the transition (Fig. 1A, column 1). Reflecting the nonfermen-
tative metabolism of oleate by the coordinated activities of
peroxisomes and mitochondria, the most significantly enriched
classes of induced genes include genes linked to mitochondrial
respiration and peroxisomal lipid metabolism (hypergeometric
ylation, electron transport chain, and aerobic respiration, P ?
10?10; components of the mitochondrial respiratory chain, P ?
10?13; fatty acid oxidation and peroxisome organization and
biogenesis, P ? 10?6; and the peroxisomal compartment, P ?
10?12). By comparison, there were many genes that were rel-
atively unresponsive in htz1? cells (Fig. 1A, column 2; Fig. 1B).
This included genes that were both poorly induced and genes
that were poorly repressed in htz1? cells compared to the WT
(Fig. 1A). Among the genes induced upon oleate exposure, 292
were expressed at least twofold less in htz1? cells than in WT
cells (Fig. 1B). Interestingly, these poorly induced genes were
most enriched for those annotated with peroxisomal functions
and components but were not enriched for annotations of
mitochondrial components or aspects of mitochondrial respi-
ration (fatty acid and lipid oxidation, P ? 4.0 ? 10?12; peroxi-
somes, P ? 9 ? 10?20) (Fig. 1C). Indeed, 26 genes (of 57 total)
encoding peroxisomal proteins showed significantly reduced
transcription in an htz1? background (Fig. 1C). These data
suggest that Htz1p is required for the regulated expression of
a large number of genes upon the transition from one state to
another. In the case of the transition to oleate, genes linked to
peroxisomal fatty acid oxidation are normally highly induced
and their expression is the most significantly affected in the
absence of Htz1p.
HTZ1 is required for normal peroxisomal ?-oxidation. The
finding that normally highly induced genes linked to fatty-acid
oxidation are poorly expressed in htz1? cells is consistent with
the finding that cells lacking HTZ1 show a specific impairment
of fatty acid metabolism (19, 34). Like mutants defective in
peroxisomal function (e.g., pex3?), htz1? cells exhibit a growth
defect on fatty acid-containing medium (YPBO) but not on
glucose-containing medium (YPD) (Fig. 2A) or on other non-
fermentable carbon sources such as glycerol (YPG) or acetate
(YPA) requiring mitochondrial function (34). As expected, the
WT cells grew normally on different carbon sources.
To examine the effect of Htz1p on the organelle itself, we
examined peroxisomes by fluorescence microscopy. WT and
htz1? cells expressing peroxisomal thiolase Pot1p, tagged by
genomic integration with green fluorescent protein (GFP),
were incubated in oleate medium and observed over a time
course of induction by direct fluorescence confocal microscopy
(Fig. 2B). In glucose-containing medium, peroxisomes were
barely detectable. However, upon the shift to oleic acid, WT
cells induced the expression and import of Pot1p-GFP as in-
dicated by the accumulation of punctate fluorescent structures
(29). However, there was a dramatic delay in the appearance of
punctate GFP fluorescence in htz1? cells compared with that
in the WT cells induced over the same time period. Together,
these data suggest that peroxisome biogenesis per se is not
defective in htz1? cells. Rather, the defect in the ability to
metabolize oleate effectively is a result of the relatively poor
expression of genes required for (peroxisomal) fatty acid me-
Transcriptional response of POT1, POX1, FOX2, and CTA1.
To further examine the molecular defects associated with the
loss of Htz1p, we focused on four strongly induced peroxisomal
matrix enzymes encoded by POT1, FOX2, POX1, and CTA1
(Fig. 3A, red). These genes are normally repressed on glucose
and strongly induced on oleic acid (32). qRT-PCR of these
mRNAs demonstrated that in the absence of Htz1p, each of
these genes was repressed as in the WT cells, but their induc-
tion was impaired upon the transition to oleate medium (Fig.
3A). Interestingly, the expression of each of these genes ap-
peared to be most significantly affected at the later time points
after the transition to oleate (compare 4 and 6 h of induction
to 0.5 and 1 h of induction). These data suggest that the loss of
Htz1p did not dramatically alter the initial response but was
important for the sustained expression of these four genes.
Having demonstrated a role for Htz1p in the normal regu-
lation of POT1, FOX2, POX1, and CTA1 expression in the
presence of oleic acid, we next sought to determine if Htz1p
binds the cognate promoters of these genes using the ChIP of
a strain expressing an HA-tagged version of Htz1p. The cells
were grown under either repressed (glucose) or activated
(oleate) conditions. Htz1p-HA was immunoprecipitated with
anti-HA antibody, and the isolated DNA was analyzed by
PCR. This analysis revealed that Htz1p was bound to each of
the four promoters (POT1, FOX2, POX1, and CTA1) in their
repressed states (Fig. 3B). These data are consistent with the
genome-wide characterization of the levels of Htz1p associa-
tion with these promoters (41). The association of Htz1p with
these promoters was dynamic; when cells were shifted to oleic
acid-activating conditions, Htz1p levels on the POT1, POX1,
and FOX2 promoters were dramatically reduced. Dissociation
from the CTA1 promoter was not observed. These data suggest
that the loss of Htz1p from promoters is coincident with gene
activation, but that dissociation is not required for the induc-
tion of all genes.
2348 WAN ET AL.MOL. CELL. BIOL.
Swr1p-, Chz1p-, and Gcn5p-dependent association of Htz1p
to promoters. Swr1p, Chz1p, and Gcn5p have been implicated
in modulating Htz1p association at promoter regions. Swr1p is
part of the SWR1-C multisubunit protein complex, necessary
for Htz1p deposition at repressed promoters (24). Chz1p was
recently identified as a histone chaperone that preferentially
interacts with Htz1p (21), and Gcn5p is the histone acetyltrans-
ferase subunit of the SAGA complex (36). To investigate
whether these factors affect Htz1p binding to the oleate-re-
sponsive promoters and subsequent expression, the association
FIG. 1. The robust expression of oleate-responsive genes expression is dependent on HTZ1. (A) Comparison of changes in the mRNA levels
of all yeast genes in WT (left column) and htz1? cells (middle column) after induction in oleate medium for 6 h. Shown are the relative expression
levels (log10) of genes that were determined to be significantly (? ? 100) altered in cells on oleate (compared to WT cells on glucose). Relative
expression levels are shown using the scale of the yellow-blue heat map (top). Genes are ordered top to bottom based on relative expression in
WT cells on oleate. Approximately 1,000 genes were significantly induced, and 1,000 genes were repressed and changed in expression at least
twofold. Genes that were reduced in expression were significantly enriched for functions related to ribosomal biogenesis (hypergeometric
distribution analysis of gene ontology terms, P ? 10?50). Induced genes were enriched for oxidative phosphorylation, the electron transport chain,
and aerobic respiration (P ? 10?10); components of the mitochondrial respiratory chain (P ? 10?13); fatty acid oxidation and peroxisome
organization and biogenesis (P ? 10?6); and the peroxisomal compartment (P ? 10?12). For comparison, the relative expression of each gene in
htz1? cells in oleate (middle column) and glucose (right column) is shown. (B) The same as described for panel A, but shown are the relative
expression levels of 292 genes significantly (? ? 100) altered in WT cells on oleate and expressed at least twofold less than their expression levels
in WT cells. This list is enriched for genes linked to fatty acid and lipid oxidation (P ? 4.0 ? 10?12) and peroxisomes (P ? 9 ?10?20). (C) The
same as described for panel B, but shown are genes encoding peroxisomal proteins significantly (? ? 100) altered in WT cells on oleate and
expressed at least twofold less in htz1? cells.
VOL. 29, 2009CONTROL OF GENE INDUCTION BY Htz1p2349
of Htz1p with POT1, POX1, FOX2, and CTA1 promoters was
investigated in cells lacking these proteins under conditions of
repression (2% glucose), and the expression of these genes was
monitored upon oleate induction (Fig. 4). Similar to its role at
the well-studied GAL1 promoter, Swr1p is required for Htz1p
binding to oleate responsive promoters, suggesting a common
role for Swr1p at disparate, highly inducible promoters. Like-
wise, Gcn5p was required for efficient Htz1p binding. This
suggests that Gcn5p, which plays a role as a coactivator of
transcription through histone acetylation (11), controls the
binding or stability of Htz1p at repressed promoters. This may
also be via histone acetylation. In the absence of Chz1p, Htz1p
occupancy at each of the four promoters was decreased. As
expected, the amount of Htz1p on each of these promoters in
mutant strains remained low upon the switch to oleate (data
Microarray analyses of the gcn5?, swr1?, and chz1? mutants
support a model in which initial Htz1p association with the
promoter is required for subsequent full induction. The ex-
pression levels of POT1, FOX2, POX1, and CTA1 were signif-
icantly reduced in the mutant strains compared to those in the
WT upon oleate induction. Moreover, all 26 genes encoding
peroxisomal proteins that showed transcriptional defects in
htz1? cells (Fig. 1C) were similarly reduced in their expression
at least twofold in the gcn5?, swr1?, and chz1? mutants com-
pared to that of the WT (Fig. 4B). Together, these data suggest
that factors functionally associated with Htz1p, such as the
chromatin remodeling complex component Swr1p, histone
acetyltransferase Gcn5p, and chaperone Chz1p, regulate the
deposition or maintenance of Htz1p at repressed promoters,
which, in turn, facilitates the rapid activation of transcription.
The acetylation of Htz1p is required for efficient transcrip-
tional induction. The acetylation of Htz1p is known to occur at
sites of active transcription (23). The finding that Gcn5p is
required for the expression of oleate-responsive genes suggests
that acetylation on Htz1p is required for the oleate response.
To address this question, plasmids expressing either one of two
acetylation mutants of Htz1p (pCM314 [Htz1p K14A] or
pCM330 [Htz1p K14R]) were introduced into htz1? cells and
expression was monitored by FACS, confocal microscopy, and
qRT-PCR. FACS and confocal microscopy demonstrated that
Pot1p-GFP fluorescence in cells carrying pCM305 (WT HTZ1)
was stronger than that in those cells carrying empty plasmid
(pRS416) or acetylation mutants (pCM330 or pCM314) during
a time course of oleate incubation, but the peroxisomes were
morphologically normal (data not shown). mRNA levels of
POT1, CTA1, FOX2, and POX1, determined by qRT-PCR,
were consistent with the GFP reporter analysis (Fig. 5A). The
K14A acetylation mutant of Htz1p showed a defect in the
normal induction of POT1, CTA1, FOX2, and POX1. In addi-
tion, cells expressing Htz1p K14A also exhibited a growth
defect on fatty-acid-containing medium but not on glucose-
containing medium. This growth defect was less pronounced
than that of the null mutant of HTZ1 (data not shown). The
association of Htz1p K14A with these oleate responsive pro-
moters at two time points (0 h and 6 h) also revealed that
FIG. 2. The deletion of HTZ1 leads to delayed peroxisome biogenesis. (A) The deletion of HTZ1 impairs cell growth on oleate-containing
medium. Strains were grown to mid-logarithmic phase in liquid YPD medium, and equal amounts of cells were serially diluted 10-fold onto YPD
and incubated at 30°C for 3 days and onto oleate-containing YPBO and incubated at 30°C for 5 days. (B) The fluorescent images of WT and htz1?
cells shown are expressing the peroxisomal matrix protein Pot1p fused with GFP (Pot1p-GFP) at different time points of oleate incubation and
were captured on a TCS SP2 laser scanning spectral confocal microscope.
2350 WAN ET AL.MOL. CELL. BIOL.
Htz1p K14A association was diminished under glucose condi-
tions (Fig. 5B). Significant differences in the association of
Htz1p K14A on these promoters were not observed during 6 h
of oleate induction. These data indicate that the acetylation of
Htz1p is required for association with oleate-responsive pro-
moters under repressed conditions and that the acetylation of
Htz1p is not required for the dissociation of Htz1p from
oleate-responsive promoters during oleate induction (Fig. 5B).
These data collectively indicate that the acetylation of Htz1p is
important for the expression of fatty acid-responsive genes and
normal peroxisomal matrix protein assembly.
TBP is not efficiently recruited to oleate-inducible promot-
ers in the absence of Htz1p. We next directly analyzed the in
vivo binding of the transcriptional machinery to repressed and
activated promoters in both the WT and htz1? strains (Fig.
6A). As expected, the binding of TBP to the POT1, POX1, and
CTA1 promoters increased with gene expression in oleate in
WT cells. The abundance of TBP did not significantly increase
on the FOX2 promoter following oleic acid induction but was
present at higher initial levels than the other three other pro-
moters studied. Nonetheless, at all four promoters in htz1?
cells, TBP binding was significantly reduced compared to that
of the WT cells. The reduced levels of TBP binding were not
attributable to decreased cellular levels of TBP. Western blot
analysis of both WT and htz1? cells demonstrated that TBP
levels were equivalent between the strains and did not signif-
icantly change during oleate induction (Fig. 5B). These results
suggest a positive function for the Htz1p-containing nucleo-
somes in the recruitment of TBP to the repressed promoters
during the process of transcriptional activation.
FIG. 3. Htz1p dynamically dissociates from oleate-responsive promoters upon induction. (A) POT1, POX1, FOX2, and CTA1 mRNA levels
were determined by RT-PCR in WT and htz1? strains over a time course of oleate induction. The signal obtained from ACT1 mRNA was used
as a loading control for normalization. Error bars represent standard deviations from the means of three independent experimental values.
(B) Htz1p enrichment at four promoters was determined by qPCR during oleate induction. Relative enrichment values (y axes) are the averages
of the results from three independent ChIPs with qPCR determination performed twice per biological replicate. Nonpromoter IGRi YMR325W
was used as an internal control to normalize signals of promoter enrichment. In response to oleate induction, Htz1p was lost from the POT1, POX1,
FOX2, and CTA1 promoters.
VOL. 29, 2009 CONTROL OF GENE INDUCTION BY Htz1p 2351
Htz1 regulates nucleosome-promoter association during ac-
tivation. A NuSA was used to investigate the role of Htz1p in
modulating chromatin structure by measuring nucleosome oc-
cupancy and location within oleate-responsive promoters dur-
ing activation (POT1, POX1, FOX2, and CTA1) (Fig. 7).
Mononucleosome-associated DNA was isolated from yeast
cells before and after oleate induction, and real-time qPCR
was used to measure dynamic nucleosome occupancy during
activation in WT and htz1? cells. The precise positions of the
nucleosomes were determined by qPCR corresponding to their
known positions (17). Overall, the gross nucleosome position
pattern at each of the four promoters under repressed condi-
tions was the same in the WT and htz1? cells. The major
nucleosome changes were observed at position N-1 in each
promoter. These nucleosomes appeared to begin disassembly
from each promoter at the earliest time point measured (5
min) and continued through to the 30-min time point. After
this initial disassembly, nucleosomes were detected to have
begun reassembly after 1 h of induction (Fig. 7). These reas-
sembled nucleosomes likely do not contain Htz1p. As shown in
Fig. 3B, Htz1p is progressively lost from these promoters dur-
ing the 6-h period of induction. Notably, the nucleosomes of
FIG. 4. Swr1p-, Chz1p-, and Gcn5p-dependent association of Htz1p with promoters. (A) In vivo association of Htz1p with the POT1, POX1,
FOX2, and CTA1 promoters was measured by ChIP in the WT, chz1?, gcn5?, and swr1? strains in 2% glucose medium. ChIP was performed in
glucose-containing medium. Error bars represent standard deviations from the means of three independent experimental values and two technical
replicates of each. (B) Comparison of changes in the mRNA levels of all yeast genes in WT, chz1?, gcn5?, and swr1? strains after induction in
oleate medium for 6 h. As described in the legend to Fig. 1C, genes encoding peroxisomal proteins significantly (? ? 100) altered in WT cells on
oleate and expressed at least twofold less in htz1? cells are shown.
2352WAN ET AL.MOL. CELL. BIOL.
each promoter appeared to be more protected at the later time
points (6 h) in htz1? cells than in the WT. This was most
evident at the N-1 position of the promoters of POX1 and
CTA1 (and at the N-2 position of POX1). These data suggest
that upon oleate treatment, the nucleosome proximal to the
initiation site in each promoter disassembles, leading to Htz1p
loss and initial transcriptional activation. During prolonged
expression, nucleosomes reassemble, but these reassembled
nucleosomes do not contain Htz1p.
Interplay between Htz1p and chromatin-remodeling factor
Isw2p. While Htz1p is proposed to contribute to nucleosome
disassembly during induction, the results presented above in-
dicate that the overall chromatin structure at the promoters
was not extensively perturbed in htz1? cells. To gain insight
into the potential additional mechanisms at play during the
transcriptional induction, we considered additional chromatin-
bound proteins. One such protein is Isw2p. Isw2p is an ATP-
dependent chromatin-remodeling factor that has previously
FIG. 5. The acetylation of Htz1p is required for efficient transcriptional induction. (A) POT1, POX1, FOX2, and CTA1 mRNA levels were
determined by RT-PCR in the WT, htz1?, and Htz1p K14A mutant strains over a time course of oleate induction. The signal obtained from ACT1
mRNA was used as a loading control for normalization. Error bars represent standard deviations from the means of three independent
experimental values. (B) Enrichment of WT Htz1p and the Htz1p K14A mutant at four promoters was determined by qPCR during glucose and
oleate induction for 6 h. Relative enrichment values (y axes) are the averages of the results from three independent ChIP experiments with two
technical replicates of each. Nonpromoter IGRi YMR325W was used as an internal control to normalize signals of promoter enrichment.
VOL. 29, 2009 CONTROL OF GENE INDUCTION BY Htz1p2353
been shown to be required for the maintenance of chromatin
structure at the POT1 promoter (8, 9, 39). We therefore in-
vestigated Isw2p function at the POT1, POX1, FOX2, and
CTA1 promoters in WT and htz1? cells.
Nucleosome protection assays of isw2? cells led to signifi-
cant changes in the nucleosome structure in all four promoters
(Fig. 8A). These data indicate that Isw2p plays a role in the
chromatin structure of these four promoters and are consistent
with previous work on the POT1 promoter (8, 9, 39).
Next, we used ChIP to assay the ability of Isw2p to associate
with the four oleate-responsive promoters. Previous work has
shown that in WT cells, Isw2p does not stably associate with
these promoters, which suggests that under normal conditions,
Isw2p contributes to the nucleosome structure at these Htz1p-
containing POT1, POX1, FOX2, and CTA1 promoters through
transient interactions (9). Similarly, we found very low levels of
Isw2p association with these promoters in WT cells, in glucose
and after oleate induction. However, substantial amounts of
Isw2p were observed in association with each of these promot-
ers in the htz1? cells. Isw2p remained associated with these
promoters during their activation, suggesting a role in estab-
lishing chromatin dynamics in the absence of Htz1p (Fig. 8B).
The exposure of yeast cells to oleate results in a large-scale
reorganization of gene expression regulatory networks and
provides an excellent experimental system for understanding
the mechanisms of gene expression at both the molecular and
network levels (28, 33). The resulting changes in gene expres-
sion are widespread, representing the reorganization of regu-
latory networks governing numerous categories of gene func-
tion. For example, genes involved in protein translation and
glycolysis are repressed, reflecting the shift in growth rates and
metabolism (16, 32, 33). Likewise, genes linked to mitochon-
drial respiration and peroxisomal fatty acid metabolism are
FIG. 6. Recruitment of TBP during oleate induction requires Htz1p. (A) The association of TBP with the POT1, POX1, FOX2, and CTA1 promoters
was determined by ChIP using anti-Myc antibodies, followed by gene-specific PCR. The relative enrichment ratio is plotted at four time points (0, 1, 4,
from three independent experimental values with two technical replicates of each. (B) Deletion of HTZ1 did not affect TBP expression during oleate
induction. The WT strain and htz1? strains expressing genomically integrated TBP were grown in 2% glucose overnight and then transferred to
oleate-containing SCIM medium at the indicated time points. Samples containing equal amounts of protein were analyzed by Western blotting with
anti-Myc antibody to visualize TBP expression. A polyclonal antibody directed against Gsp1p was used as a loading control.
2354 WAN ET AL.MOL. CELL. BIOL.
induced, reflecting the shift of the cells to nonfermentative
?-oxidation as an energy source (16, 32, 33). The ability of
yeast cells to adapt to this shift is dependent on the HTZ1 gene
encoding the histone variant Htz1p/H2A.Z (19, 32). The data
presented here demonstrate that Htz1p plays a critical role in
this transition by contributing to the recruitment of TBP to
oleate-responsive genes leading to the rapid and robust expres-
sion of highly inducible genes.
Transcriptome profiling studies presented here demonstrate
oleate is impaired in the absence of Htz1p. Because, under these
conditions, many of the most strongly induced genes are required
for peroxisomal ?-oxidation and peroxisome proliferation, lack of
Htz1p renders cells unable to respond efficiently to the transition
and metabolize the fatty acids.
In order to elucidate the stepwise molecular function of
Htz1p in the transcriptional regulation of these genes, we gen-
erated and compared various time course datasets to analyze
chromatin states before and after the switch to oleic acid. We
used ChIP to assay the dynamic association of Htz1p at the
promoters of four model genes (POT1, POX1, FOX2, and
CTA1) encoding peroxisomal matrix enzymes, the expression
of which was perturbed by deletions of HTZ1. Htz1p has been
proposed to preferentially bind repressed promoters, facilitat-
ing the rapid activation of the associated genes (41). Consistent
with the current models, we demonstrate that Htz1p tends to
be bound to these promoters in their repressed states (glucose)
and disassociates from these promoters once the cells are ex-
posed to oleate; however, this association and dissociation
pattern occurs at levels that are promoter specific. The meth-
ods employed here did not reveal a significant dissociation of
Htz1p from the CTA1 promoter. The data suggest that Htz1p
levels on the CTA1 promoter are lower (approximately twofold
lower than the control regions) than the other promoters ex-
amined. Thus, Htz1p does not appear to dissociate from the
CTA1 promoter following oleic acid induction. The mecha-
nisms underlying the promoter-specific effects of Htz1p and
other epigenetic factors remain fertile ground for future study.
both yeast and mammalian cells that demonstrate that Htz1p is
deposited at promoters by the chromatin-remodeling protein
Swr1p (24, 40). Similarly, as in other transcriptional responses
(23), Gcn5p/Esa1p-mediated acetylation at Lys14 of Htz1p is
Esa1p-mediated acetylation at Lys14 of Htz1p is required for the
efficient association of Htz1p at some oleate responsive promot-
ers. The significantly decreased association of the Htz1p K14R
mutant was observed under repressive conditions (i.e., glucose),
FIG. 7. Htz1p regulates the occupancy of specific nucleosomes on the POT1, POX1, FOX2, and CTA1 promoters. The NuSA was used to
determine the nucleosome positioning and density at the POT1, POX1, FOX2, and CTA1 promoters during oleate induction (the time of induction
is indicated on the left) in the WT and HTZ1 deletion strains. Each point represents the relative protection of each PCR amplicon, quantified by
real-time PCR and normalized to a centromeric control. The position of each amplicon (referenced to the middle of each amplicon) within the
promoter is shown on the x axis. The approximate location of a nucleosome is represented by a gray circle with the nucleosome number referred
to in the text shown above the circle.
VOL. 29, 2009CONTROL OF GENE INDUCTION BY Htz1p2355
and this decreased binding of Htz1p was also observed in cells
lacking the enzyme (Gcn5p) responsible for Htz1p acetylation.
Htz1p acetylation mutant cells also displayed defects in peroxi-
some proliferation and growth on oleic acid, similar to an HTZ1
null mutant. These data demonstrate that the acetylation of
Htz1p, mediated by Gcn5p, is required for association with
oleate-responsive promoters during repressed conditions and for
normal transcriptional induction contributed by Htz1p.
FIG. 8. Isw2p can associate with oleate-responsive promoters in the absence of Htz1p. (A) The NuSA was used to determine the nucleosome
positioning and density at the POT1, POX1, FOX2, and CTA1 promoters during repression (2% glucose) in WT, htz1?, and isw2? strains. Each
point represents the relative protection of each PCR amplicon, quantified by real-time PCR and normalized to a centromeric control. The
approximate location of a nucleosome is represented by a gray circle with the nucleosome number referred to in the text shown above the circle.
(B) The association of Isw2p (as a C-terminal myc fusion) with the POT1, POX1, FOX2, and CTA1 promoters was determined by ChIP using
anti-Myc antibodies, followed by gene-specific PCR. The relative enrichment ratio is plotted at four time points (0, 1, 4, and 6 h) of induction in
oleate. ACT1 was used as an internal control to normalize the signals of promoter enrichment. Error bars show the standard deviations from three
independent experimental values with two technical replicates of each.
2356 WAN ET AL.MOL. CELL. BIOL.
Among the known effectors of Htz1p, Chz1p is relatively less
well characterized. Luk et al. (21) identified a role for Chz1p as
a nuclear chaperone for Htz1p, though the functional relation-
ship between Chz1p and Htz1p with respect to transcriptional
regulation remained uncharacterized. Here, we report that
Chz1p, like Swr1p and Gcn5p, is also involved in the deposi-
tion of Htz1p at repressed promoters.
With respect to the role of Htz1p in TBP recruitment, stud-
ies of the GAL promoters have drawn different conclusions. In
a recent study, TBP recruitment to the GAL1 promoter in
htz1? strains was indistinguishable from that of the WT cells
(10). However, earlier studies showed the Htz1p-dependent
enrichment of TBP to GAL1 and GAL10 promoters during a
time course of galactose induction (1). In the case of the fatty
acid-inducible promoters tested here, the absence of Htz1p led
to a significant reduction in the recruitment of TBP during
oleate induction. We did not observe an increased enrichment
of TBP at the FOX2 promoter during oleate induction. The
dynamics of TBP binding appear to be promoter specific. The
relative abundance of TBP at the FOX2 promoter prior to
induction by oleic acid (compared to later time points) sug-
gests that the activation of FOX2 does not require additional
TBP binding and that the gene exists in a state poised for its
activation upon receiving the correct signals (i.e., oleic acid). A
comprehensive investigation of the dynamic and quantitative
role of Htz1p in the recruitment of factors such as mediator
and TBP to different promoters throughout the genome re-
mains for future studies.
Our data suggest that the activation of repressed genes leads
to a dynamic reorganization of chromatin structure. Specifi-
cally, upon oleate treatment, the nucleosome proximal to the
initiation site in each promoter disassembles. This coincides
with the ejection of Htz1p from the promoter. These data are
in agreement with previous studies that indicate that nucleo-
some disassembly from promoters during activation provides
access to the transcriptional machinery (25). While Htz1p is
proposed to contribute to nucleosome disassembly during in-
duction, surprisingly, the apparent rate of nucleosome disas-
sembly at the oleate-responsive promoters was not dramati-
cally different in htz1? cells. After initial disassembly, the
nucleosomes reassemble (?1 h after induction). Interestingly,
these new nucleosomes do not appear to contain Htz1p, but
the levels of transcription are nonetheless higher in WT cells
than in cells lacking Htz1p. These data suggest that the initial
presence of Htz1p ensures a normal transcriptional response
and provides an epigenetic mark that persists after its loss,
ensuring high levels of expression. A close examination of the
data from nucleosome protection assays suggest that, in the
absence of Htz1p, nucleosomes in the promoter regions of
oleate-responsive genes are relatively more assembled, which
may cause reduced expression levels at these later time points.
The reassembly of nucleosomes during the coincident high
levels of gene expression suggests that transcriptional activity is
not simply related to an overall openness of chromatin at
activated promoters and obstruction at repressed promoters.
Rather, the precise dynamic placement and specific constitu-
ents of individual nucleosomes at promoters mechanistically
regulate transcription by modulating the access of transacting
factors to specific sites. Further, the characterization of the
dynamics of the epigenetic marks, protein components of the
nucleosomes, and chromatin-remodeling complexes at these
promoters is required to delineate the mechanistic basis of the
links between chromatin state and transcriptional activation.
The observed lower expression levels in cells lacking Htz1p
may also be contributed by Isw2p. Isw2p is an energy-depen-
dent chromatin-remodeling factor and negative regulator of
gene expression (8, 9). When assayed for genome binding by
ChIP-chip Isw2p association with the POT1, POX1, FOX2, and
CTA1 promoters was not detected (9). Similarly, we found no
enrichment of Isw2p at these promoters in WT cells. However,
Isw2p bound to each promoter in the absence of Htz1p, and
this association persisted through the 6 hours of induction.
Therefore, the increased association of Isw2p to these four
oleate-responsive promoters may account for the reduced ex-
pression levels in htz1? cells. It is also possible that in the WT
cells Isw2p provides a complementary mechanism for chroma-
tin structural changes independent of Htz1p. The loss of Htz1p
provides an opportunity for Isw2p binding that is not normally
functional in Htz1p-containing regions of chromatin. Further
studies are required to understand the global relationship be-
tween Isw2p and Htz1p.
In mammalian cells histone variant H2A.Z can serve as a
novel epigenetic marker of breast cancer progression as it is
associated with lymph node metastasis and decreased breast
cancer survival (13). In the plant Arabidopsis thaliana, histone
H2A.Z is required for immune resistance to the phytopatho-
genic bacterium Pseudomonas syringae pv. tomato (22). In ze-
brafish, histone variant 2a z (H2afza) is essential for larval
development through the generation of a lethal locus with a
truncation of conserved carboxy-terminal residues in the pro-
tein (31). Taken together, these studies implicate histone
H2A.Z in a number of diverse functions in different organisms.
Because peroxisomes are highly dynamic and responsive eu-
karyotic organelles whose dysfunction is linked to a host of
human conditions (4, 5), it is important to understand the roles
of proteins like Htz1p that control aspects of chromatin struc-
ture and transcriptional responses preceding the proliferation
of peroxisomes and fatty acid metabolism in S. cerevisiae.
We thank Bradley R. Cairns, Haiying Zhang, and Michael Grun-
stein for providing plasmids and strains and Jeff Ranish and members
of the Aitchison laboratory for helpful comments and discussion dur-
ing the course of this project.
This work was supported by grants GM075152, GMO76547, and
RR022220 from the U.S. National Institutes of Health.
1. Adam, M., F. Robert, M. Larochelle, and L. Gaudreau. 2001. H2A.Z is
required for global chromatin integrity and for recruitment of RNA poly-
merase II under specific conditions. Mol. Cell. Biol. 21:6270–6279.
2. Aitchison, J. D., M. P. Rout, M. Marelli, G. Blobel, and R. W. Wozniak. 1995.
Two novel related yeast nucleoporins Nup170p and Nup157p: complemen-
tation with the vertebrate homologue Nup155p and functional interactions
with the yeast nuclear pore-membrane protein Pom152p. J. Cell Biol. 131:
3. Albert, I., T. N. Mavrich, L. P. Tomsho, J. Qi, S. J. Zanton, S. C. Schuster,
and B. F. Pugh. 2007. Translational and rotational settings of H2A.Z nu-
cleosomes across the Saccharomyces cerevisiae genome. Nature 446:572–
4. Bensinger, S. J., and P. Tontonoz. 2008. Integration of metabolism and
inflammation by lipid-activated nuclear receptors. Nature 454:470–477.
5. Berger, J., and D. E. Moller. 2002. The mechanisms of action of PPARs.
Annu. Rev. Med. 53:409–435.
6. Biddick, R. K., G. L. Law, and E. T. Young. 2008. Adr1 and Cat8 mediate
VOL. 29, 2009 CONTROL OF GENE INDUCTION BY Htz1p2357
coactivator recruitment and chromatin remodeling at glucose-regulated Download full-text
genes. PLoS ONE 3:e1436.
7. Dudley, A. M., J. Aach, M. A. Steffen, and G. M. Church. 2002. Measuring
absolute expression with microarrays with a calibrated reference sample and
an extended signal intensity range. Proc. Natl. Acad. Sci. USA 99:7554–7559.
8. Fazzio, T. G., M. E. Gelbart, and T. Tsukiyama. 2005. Two distinct mecha-
nisms of chromatin interaction by the Isw2 chromatin remodeling complex in
vivo. Mol. Cell. Biol. 25:9165–9174.
9. Gelbart, M. E., N. Bachman, J. Delrow, J. D. Boeke, and T. Tsukiyama. 2005.
Genome-wide identification of Isw2 chromatin-remodeling targets by local-
ization of a catalytically inactive mutant. Genes Dev. 19:942–954.
10. Gligoris, T., G. Thireos, and D. Tzamarias. 2007. The Tup1 corepressor
directs Htz1 deposition at a specific promoter nucleosome marking the
GAL1 gene for rapid activation. Mol. Cell. Biol. 27:4198–4205.
11. Govind, C. K., F. Zhang, H. Qiu, K. Hofmeyer, and A. G. Hinnebusch. 2007.
Gcn5 promotes acetylation, eviction, and methylation of nucleosomes in
transcribed coding regions. Mol. Cell 25:31–42.
12. Guillemette, B., A. R. Bataille, N. Gevry, M. Adam, M. Blanchette, F. Robert,
and L. Gaudreau. 2005. Variant histone H2A.Z is globally localized to the
promoters of inactive yeast genes and regulates nucleosome positioning.
PLoS Biol. 3:e384.
13. Hua, S., C. B. Kallen, R. Dhar, M. T. Baquero, C. E. Mason, B. A. Russell,
P. K. Shah, J. Liu, A. Khramtsov, M. S. Tretiakova, T. N. Krausz, O. I.
Olopade, D. L. Rimm, and K. P. White. 2008. Genomic analysis of estrogen
cascade reveals histone variant H2A.Z associated with breast cancer pro-
gression. Mol. Syst. Biol. 4:188.
14. Ideker, T., V. Thorsson, A. F. Siegel, and L. E. Hood. 2000. Testing for
differentially-expressed genes by maximum-likelihood analysis of microarray
data. J. Comput. Biol. 7:805–817.
15. Jackson, J. D., V. T. Falciano, and M. A. Gorovsky. 1996. A likely histone
H2A.F/Z variant in Saccharomyces cerevisiae. Trends Biochem. Sci. 21:466–
16. Koerkamp, M. G., M. Rep, H. J. Bussemaker, G. P. Hardy, A. Mul, K.
Piekarska, C. A. Szigyarto, J. M. De Mattos, and H. F. Tabak. 2002. Dis-
section of transient oxidative stress response in Saccharomyces cerevisiae by
using DNA microarrays. Mol. Biol. Cell 13:2783–2794.
17. Lee, W., D. Tillo, N. Bray, R. H. Morse, R. W. Davis, T. R. Hughes, and C.
Nislow. 2007. A high-resolution atlas of nucleosome occupancy in yeast. Nat.
18. Li, B., S. G. Pattenden, D. Lee, J. Gutierrez, J. Chen, C. Seidel, J. Gerton,
and J. L. Workman. 2005. Preferential occupancy of histone variant H2AZ
at inactive promoters influences local histone modifications and chromatin
remodeling. Proc. Natl. Acad. Sci. USA 102:18385–18390.
19. Lockshon, D., L. E. Surface, E. O. Kerr, M. Kaeberlein, and B. K. Kennedy.
2007. The sensitivity of yeast mutants to oleic acid implicates the peroxisome
and other processes in membrane function. Genetics 175:77–91.
20. Longtine, M. S., A. McKenzie III, D. J. Demarini, N. G. Shah, A. Wach, A.
Brachat, P. Philippsen, and J. R. Pringle. 1998. Additional modules for
versatile and economical PCR-based gene deletion and modification in Sac-
charomyces cerevisiae. Yeast 14:953–961.
21. Luk, E., N. D. Vu, K. Patteson, G. Mizuguchi, W. H. Wu, A. Ranjan, J.
Backus, S. Sen, M. Lewis, Y. Bai, and C. Wu. 2007. Chz1, a nuclear chap-
erone for histone H2AZ. Mol. Cell 25:357–368.
22. March-Diaz, R., M. Garcia-Dominguez, J. Lozano-Juste, J. Leon, F. J.
Florencio, and J. C. Reyes. 2008. Histone H2A.Z and homologues of com-
ponents of the SWR1 complex are required to control immunity in Arabi-
dopsis. Plant J. 53:475–487.
23. Millar, C. B., F. Xu, K. Zhang, and M. Grunstein. 2006. Acetylation of
H2AZ Lys 14 is associated with genome-wide gene activity in yeast. Genes
24. Mizuguchi, G., X. Shen, J. Landry, W. H. Wu, S. Sen, and C. Wu. 2004.
ATP-driven exchange of histone H2AZ variant catalyzed by SWR1 chroma-
tin remodeling complex. Science 303:343–348.
25. Petesch, S. J., and J. T. Lis. 2008. Rapid, transcription-independent loss of
nucleosomes over a large chromatin domain at Hsp70 loci. Cell 134:74–84.
26. Raisner, R. M., P. D. Hartley, M. D. Meneghini, M. Z. Bao, C. L. Liu, S. L.
Schreiber, O. J. Rando, and H. D. Madhani. 2005. Histone variant H2A.Z
marks the 5? ends of both active and inactive genes in euchromatin. Cell
27. Ramsey, S. A., J. J. Smith, D. Orrell, M. Marelli, T. W. Petersen, P. de
Atauri, H. Bolouri, and J. D. Aitchison. 2006. Dual feedback loops in the
GAL regulon suppress cellular heterogeneity in yeast. Nat. Genet. 38:1082–
28. Ratushny, A. V., S. A. Ramsey, O. Roda, Y. Wan, J. J. Smith, and J. D.
Aitchison. 2008. Control of transcriptional variability by overlapping feed-
forward regulatory motifs. Biophys. J. 8:3715–3723.
29. Saleem, R. A., B. Knoblach, F. D. Mast, J. J. Smith, J. Boyle, C. M. Dobson,
R. Long-O’Donnell, R. A. Rachubinski, and J. D. Aitchison. 2008. Genome-
wide analysis of signaling networks regulating fatty acid-induced gene ex-
pression and organelle biogenesis. J. Cell Biol. 181:281–292.
30. Santisteban, M. S., T. Kalashnikova, and M. M. Smith. 2000. Histone
H2A.Z regulates transcription and is partially redundant with nucleosome
remodeling complexes. Cell 103:411–422.
31. Sivasubbu, S., D. Balciunas, A. E. Davidson, M. A. Pickart, S. B. Herman-
son, K. J. Wangensteen, D. C. Wolbrink, and S. C. Ekker. 2006. Gene-
breaking transposon mutagenesis reveals an essential role for histone H2afza
in zebrafish larval development. Mech. Dev. 123:513–529.
32. Smith, J. J., M. Marelli, R. H. Christmas, F. J. Vizeacoumar, D. J. Dilworth,
T. Ideker, T. Galitski, K. Dimitrov, R. A. Rachubinski, and J. D. Aitchison.
2002. Transcriptome profiling to identify genes involved in peroxisome as-
sembly and function. J. Cell Biol. 158:259–271.
33. Smith, J. J., S. A. Ramsey, M. Marelli, B. Marzolf, D. Hwang, R. A. Saleem,
R. A. Rachubinski, and J. D. Aitchison. 2007. Transcriptional responses to
fatty acid are coordinated by combinatorial control. Mol. Syst. Biol. 3:115.
34. Smith, J. J., Y. Sydorskyy, M. Marelli, D. Hwang, H. Bolouri, R. A. Rachu-
binski, and J. D. Aitchison. 2006. Expression and functional profiling reveal
distinct gene classes involved in fatty acid metabolism. Mol. Syst. Biol.
35. Stargell, L. A., J. Bowen, C. A. Dadd, P. C. Dedon, M. Davis, R. G. Cook,
C. D. Allis, and M. A. Gorovsky. 1993. Temporal and spatial association of
histone H2A variant hv1 with transcriptionally competent chromatin during
nuclear development in Tetrahymena thermophila. Genes Dev. 7:2641–2651.
36. Sterner, D. E., and S. L. Berger. 2000. Acetylation of histones and transcrip-
tion-related factors. Microbiol. Mol. Biol. Rev. 64:435–459.
37. Suto, R. K., M. J. Clarkson, D. J. Tremethick, and K. Luger. 2000. Crystal
structure of a nucleosome core particle containing the variant histone
H2A.Z. Nat. Struct. Biol. 7:1121–1124.
38. Whitehouse, I., O. J. Rando, J. Delrow, and T. Tsukiyama. 2007. Chromatin
remodelling at promoters suppresses antisense transcription. Nature 450:
39. Whitehouse, I., and T. Tsukiyama. 2006. Antagonistic forces that position
nucleosomes in vivo. Nat. Struct. Mol. Biol. 13:633–640.
40. Wong, M. M., L. K. Cox, and J. C. Chrivia. 2007. The chromatin remodeling
protein, SRCAP, is critical for deposition of the histone variant H2A.Z at
promoters. J. Biol. Chem. 282:26132–26139.
41. Zhang, H., D. N. Roberts, and B. R. Cairns. 2005. Genome-wide dynamics of
Htz1, a histone H2A variant that poises repressed/basal promoters for acti-
vation through histone loss. Cell 123:219–231.
2358 WAN ET AL.MOL. CELL. BIOL.